RESPONSE OF AQUATIC BIOTA TO CHANGING LAND USE PATTERN IN 
STREAMS OF WEST GEORGIA, USA 
 
 
Except where reference is made to the work of others, the work described in this 
dissertation is my own or was done in collaboration with my advisory committee.  This 
dissertation does not include proprietary or classified information  
 
 
 
________________________________ 
Brian S. Helms 
 
 
 
 
 
 
Certificate of Approval 
 
 
 
 
 
__________________________              _____________________________ 
B. Graeme Lockaby     Jack W. Feminella, Chair 
Professor      Professor 
Forestry and Wildlife Sciences   Biological Sciences 
 
 
 
__________________________              _____________________________ 
Dennis R. DeVries     Joe F, Pittman 
Profesor      Interim Dean 
Fisheries and Allied Aquacultures   Graduate School 
 
RESPONSE OF AQUATIC BIOTA TO CHANGING LAND USE PATTERN IN 
STREAMS OF WEST GEORGIA, USA 
 
 
Brian S. Helms 
 
 
 
A Dissertation 
Submitted to 
the Graduate Faculty of 
Auburn University 
in Partial Fulfillment of the 
Requirements for the 
Degree of 
Doctor of Philosophy 
 
 
Auburn, Alabama 
May 10, 2008 
 iii
RESPONSE OF AQUATIC BIOTA TO CHANGING LAND USE PATTERN IN 
STREAMS OF WEST GEORGIA, USA 
 
 
Brian S. Helms 
 
 
 
Permission is granted to Auburn University to make copies of this dissertation at its 
discretion, upon request of individuals or institutions and at their expense.  The author 
reserves all publication rights 
 
 
 
 
 
 
 
 
 
 
___________________________ 
      Signature of Author 
 
 
___________________________ 
      Date of Graduation 
 
 
 
 
 
 iv
VITA 
 Brian Scott Helms, son of R. Gary Helms and Frances P. Helms, brother of 
Shannon N. Helms, husband of Kristen T. Helms and father of Isaac L. Helms, was born 
on February 27, 1973, in Gaston County, NC.  He graduated from Ashbrook High 
School, Gastonia, NC in 1991.  He then moved to Raleigh, NC, where he earned a 
Bachelor of Science in Biology and a Bachelor of Science in Zoology from NC State 
University in 1995.  In 2000, he received a Master of Science in Biology from 
Appalachian State University, Boone, NC, under the guidance of Dr. Robert P. Creed.  
Following 2 years with the North Carolina Botanical Garden, he entered the Department 
of Biological Sciences under the guidance of Dr. Jack Feminella in 2002 and earned a 
Doctor of Philosophy in Biological Sciences in 2008. 
 v
DISSERTATION ABSTRACT 
RESPONSE OF AQUATIC BIOTA TO CHANGING LAND USE PATTERN IN 
STREAMS OF WEST GEORGIA, USA 
 
Brian S. Helms 
Ph.D., Auburn University, May 10, 2008 
(M.S., Appalachian State University, 2000) 
(B.S. North Carolina State University, 1995) 
(B.S. North Carolina State University, 1995) 
 
224 Typed Pages 
 
Directed by Jack W. Feminella 
 
Watershed land use change is one of the most pervasive influences facing streams 
in the southeastern United States.  In particular, watershed urbanization has a potentially 
dramatic impact on stream biota.  Through direct alteration of several interrelated 
instream processes including hydrology, geomorphology, and physicochemistry, land use 
change can indirectly influence stream biotic composition and quality.  Starting in 2002, I 
examined aquatic biotic responses to changing watershed land use/cover (LU/LC) and 
subsequent abiotic alterations in the Lower Piedmont ecoregion north of Columbus, 
Georgia, USA, in 25 small streams (2 ? 3
rd
 order) whose watersheds (500 ? 2500 ha) 
varied in their degree of urbanization, residential development, pasture, managed and 
unmanaged forest cover.  
 vi
Benthic macroinvertebrates, as taxa richness and Shannon?s diversity (H?), were 
negatively associated with increasing urban cover, peak stream discharge, and water 
temperature, and positively associated with increasing forest cover and dissolved oxygen 
(DO) levels.  Further, macroinvertebrate biomass increased while seasonal variation 
decreased with increasing urban cover. Fish assemblages were largely explained by a 
combination of land use and landscape (stream hydrology, geomorphology) variables that 
vary seasonally, with stronger land use signals in the summer and a stronger fish response 
to landscape variables in the winter and spring.  Fish assemblages shifted from cyprinid-
based in forested (managed and unmanaged) and pasture watersheds to centrarchid-based 
in urban and suburban watersheds; this shift was correlated with increasing spate 
frequency, stream temperature and decreasing DO.  Stable isotopes analyses revealed 
food webs in watersheds with increased LU/LC disturbance were significantly different 
than food webs in forested watersheds.  Further, overall food chain length and mean 
trophic position decreased with increasing impervious surface.  Specifically, food chain 
length increased with stream habitat quality whereas chain length and mean trophic 
position decreased with increasing maximum stream discharge and duration of high 
flows. 
My results suggest that changing land use exerts multiple direct and indirect 
stressors on stream biota that vary depending on the biology of the organism or 
assemblage under consideration.  Specifically, streams in urbanized watersheds are harsh 
environments that have less diverse assemblages dominated by few tolerant taxa, minimal 
seasonal variation, and simplified food webs.   
 vii
ACKNOWLEDGEMENTS 
I would like to thank my major advisor, Jack Feminella, for his insight, guidance 
and friendship during this project.  I also thank my committee members, Graeme 
Lockaby and Dennis DeVries, for keeping my perspectives sufficiently broad, and Rusty 
Wright who provided an astute review of this document.  Many of the ideas in these 
studies come from countless hours of fieldwork and discussions with Jon Schoonover, 
whereas Kyle Barrett, Richard Mitchell, and Kelly Maloney were instrumental in 
identifying the bad ideas.  I also thank Nate Bickford, Michael Buntin, Michele Burton, 
Robert Creed, Jacqueline Crim, Ken Fortino, Michael Gangloff, Emily Hartfield, Laura 
Heck, Sharon Hermann, Steve Herrington, Michael Holley, Brian Lowe, Stephanie 
Miller, Molli Newman, Jeff Stratford, and Abbie Tomba for their integral contributions to 
this research and my development.   Without the help of Amy Dalrymple, Sarah 
Dalrymple, Ray Fisher, Courtney Ford, Talitha Thompson, Nathan Tubbs, Laura 
Wasylenko, and Christy Waits in the field and/or laboratory, none of this work would be 
possible.  I thank Shufen Pan for generating the impervious surface data used throughout 
and Sandra Riddle and Tamara Knowles for keeping me sufficiently informed.  I would 
also like to thank the west Georgia landowners who granted access to their property.   
This project was supported by grants from the Auburn University Environmental Institute 
and the Auburn University Center for Forest Sustainability.  Finally, I would like to thank 
my wife, Kristen Helms, for her incredible love and support during this endeavor. 
 ix
Style manual of journal used:  Ecological Applications (Chapter 1, 4, 5), Urban 
Ecosystems (Chapter 2), Journal of the American Water Resources Association (Chapter 
3). 
Computer software used:  Microsoft Word 2003, Microsoft Excel 2003, Microsoft 
Access 2000, SAS System for Windows 8.0, Minitab 13, PC-ORD 4.25, ArcView 3.2a, 
Sigma Plot 8.0, Endnote 9.0. 
 
 
 
 x
TABLE OF CONTENTS 
 
LIST OF TABLES .................................................................................................xiii 
 
LIST OF FIGURES...............................................................................................xvii 
 
1.   INTRODUCTION TO LAND USE EFFECTS ON STREAM BIOTA...............1 
 
 1.1 INTRODUCTION.................................................................................1 
 1.2 REFERENCES......................................................................................6
 
2. DETECTION OF FISH ASSEMBLAGE RESPONSE TO URBANIZATION 
FROM SMALL STREAMS OF WESTERN GEORGIA, USA ........................11 
 
  2.1 ABSTRACT.......................................................................................11 
  2.2  INTRODUCTION..............................................................................12 
  2.3  METHODS ........................................................................................14 
   2.3.1  STUDY AREA ........................................................................14 
   2.3.2  LANDSCAPE CLASSIFICATION..........................................15 
   2.3.3  FIELD SAMPLING .................................................................16 
   2.3.4  DATA ANALYSES.................................................................17 
  2.4 RESULTS..........................................................................................19 
  2.5 DISCUSSION....................................................................................22 
   2.5.1  SEASONALITY OF FISH RESPONSE...................................23 
   2.5.2  IMPERVIOUS SURFACE AND FISH RESPONSE ................24 
   2.5.3  URBANIZATION AND ASSEMBLAGE STRUCTURE ........26 
  2.6 REFERENCES ..................................................................................29 
 
3.   ASSESSING INFLUENCES OF HYDROLOGY, PHYSICOCHEMISTRY, AND 
HABITAT ON STREAM FISH ASSEMBLAGES ACROSS A CHANGING 
LANDSCAPE ..................................................................................................47 
   
  3.1 ABSTRACT ......................................................................................47 
  3.2 INTRODUCTION .............................................................................48 
  3.3  STUDY AREA ..................................................................................50 
  3.4 METHODS........................................................................................51 
   3.4.1  LAND COVER ANALYSIS ....................................................51 
   3.4.2  HYDROLOGY MEASURES...................................................52 
   3.4.3  PHYSICOCHEMISTRY MEASURES.....................................53 
   3.4.4   FISH AND HABITAT SAMPLING........................................53
 
 xi
   3.4.5  DATA ANALYSES.................................................................56 
  3.5 RESULTS..........................................................................................57 
   3.5.1  STREAM HYDROLOGY........................................................57 
   3.5.2  PHYSICOCHEMISTRY AND HABITAT...............................57 
   3.5.3  FISH ASSEMBLAGE STRUCTURE ......................................58 
 3.5.4  INFLUENCE OF ENVIRONMENTAL VARIABLES....................59 
  3.6 DISCUSSION....................................................................................59 
   3.6.1  HYDROLOGY AND FISH ASSEMBLAGES.........................60 
   3.6.2 CONCLUSIONS......................................................................64 
  3.7 REFERENCES...................................................................................66 
 
4. LAND USE IMPACTS ON STREAM MACROINVERTEBRATE 
ASSEMBLAGES: TEMPORAL VARIABILITY AND THE INTERPLAY OF 
HYDROLOGY, PHYSICOCHEMISTRY, AND GEOMORPHOLOGY..........88 
   
  4.1 ABSTRACT ......................................................................................88 
  4.2 INTRODUCTION .............................................................................89 
  4.3  METHODS ........................................................................................91 
   4.3.1  STUDY AREA ........................................................................91 
   4.3.2  LAND COVER / LAND USE ANALYSIS ..............................92 
   4.3.3  STREAM HYDROLOGY VARIABLES .................................93 
   4.3.4  PHYSICOCHEMISTRY VARIABLES ...................................93 
   4.3.5 MACROINVERTEBRATES ...................................................94 
   4.3.6  STATISTICAL ANALYSES ...................................................96 
  4.4 RESULTS..........................................................................................98 
   4.4.1  ENVIRONMENTAL VARIABLES AND LU/LC ...................98 
   4.4.2  MACROINVERTEBRATE ASSEMBLAGES.........................99 
   4.4.3 MULTIVARIATE RELATIONSHIPS................................... 100 
  4.5  DISCUSSION.................................................................................. 102 
   4.5.1 RELATIVE IMPACTS OF DIFFERENT LU/LC................... 102 
   4.5.2  MACROINVERTEBRATE SEASONALITY ........................ 105 
   4.5.3 POTENTIAL ASSEMBLAGE DRIVERS.............................. 105 
   4.5.4 CONCLUSION AND IMPLICATIONS................................. 107 
  4.6  REFERENCES ................................................................................ 109 
 
5.  LAND USE EFFECTS AND THE INFLUENCE OF ECOSYSTEM SIZE, 
RESOURCE AVAILABILITY, AND DISTURBANCE ON STREAM FOOD 
WEBS ........................................................................................................ 134 
 
  5.1 ABSTRACT .................................................................................... 134 
  5.2  INTRODUCTION ........................................................................... 135 
  5.3  METHODS...................................................................................... 138 
   5.3.1  STUDY AREA ...................................................................... 138 
   5.3.2 RESOURCE AVAILABILITY .............................................. 139 
   5.3.3 ENVIRONMENTAL STABILITY......................................... 140 
 
 xii
   5.3.4 SYSTEM SIZE ...................................................................... 142 
   5.3.5 FISH COLLECTION ............................................................. 142 
   5.3.6  SAMPLE PREPARATION AND ANALYSIS....................... 143 
   5.3.7 DATA ANALYSES............................................................... 144 
  5.4 RESULTS........................................................................................ 146 
   5.4.1  TROPHIC POSITION AND FOOD CHAIN LENGTH.......... 146 
   5.4.2  MEAN FOOD WEBS ............................................................ 147 
  5.5 DISCUSSION.................................................................................. 149 
   5.5.1 ?
13
C AND ?
15
N DIFFERENCES AMONG WATERSHEDS.. 149 
   5.5.2  CORRELATES OF FOOD CHAIN LENGTH ....................... 150 
   5.5.3  FOOD CHAIN LENGTH AND WEB STRUCTURE............. 152 
   5.5.4 CONCLUSION...................................................................... 154 
  5.6 REFERENCES ................................................................................ 156  
 
APPENDIX ........................................................................................................ 180 
 
 
 xiii
LIST OF TABLES 
 
Chapter 2 
 
Table 1: Characteristics of the 15 study watersheds.  LU/LC refers to the dominant land 
use/land cover in the watershed based on GIS.  A watershed was classified only 
if it contained >30% of a given land use category, except for Developing, which 
showed <30% urban cover but contained active residential development.  UTM 
coordinates were taken at the sampling site furthest downstream. Trib. = 
tributary. ...................................................................................................37 
 
Table 2: Fish metrics quantified by stream and season in the 15 study watersheds.   
Complex breeders that lack parental care were simply classified ?Complex?. 
P/C = complex breeders showing parental care, No P/C = complex breeders 
showing no parental care. See text for definitions of breeding guilds and 
tolerance values.........................................................................................38 
 
Table 3: Initial land use and natural basin variables quantified in the 15 study 
watersheds.  Asterisks indicate those variables used in final analyses. .......39 
 
Table 4: Summary fish data for the 15 study watersheds.  Values are cumulative for the 
sites (seasons combined).  IBI= GA Index of Biotic Integrity (see Schleiger 
2000). % DELT = proportion of fish in the sample with eroded fins, lesions, or 
tumors. P/C = proportion of complex breeders showing parental care, No P/C = 
proportion of complex breeders showing no  parental care. See text for further 
explanation of variables.............................................................................40 
 
Table 5: Pearson product-moment correlation coefficients between selected fish variables 
and land use variables (based on 30-m resolution Landsat data, see text).  % 
Herbivore = proportion of herbivorous species in the sample, % Simple spawner 
= proportion of species in the sample showing no complex spawning behavior, 
% Lithophils = proportion of species in the sample requiring clean, gravel 
substrate for spawning, %Tolerant = proportion of tolerant species in the 
sample, % DELT = proportion of fish in sample with eroded fins, lesions, or 
tumors in the sample, Bray-Curtis = stream faunal similarity to average 
reference condition, IBI= Georgia Index of Biotic Integrity. * p = 0.05, ** p = 
0.01, *** p < 0.001. ns = nonsignificant. ...................................................41
 
 xiv
Table 6: Seasonal multiple regression models for select fish metrics and simple linear 
regressions of fish metrics with 1-m resolution % impervious surface. % 
Herbivore =proportion of herbivorous species in the sample, % Omnivore 
=proportion of omnivorous species in the sample, % Cyprinid insectivore 
=proportion of insectivorous species in the family Cyprinidae, % Pioneer = 
proportion of pioneer species in the sample, % DELT = proportion of fish in 
sample with eroded fins, lesions, or tumors in the sample, Bray-Curtis = stream 
faunal similarity to reference stream, and IBI= Georgia Index of Biotic 
Integrity.  Direction of relationships for simple regressions are denoted by (+) 
and (-?) in the R2 column when appropriate................................................42 
 
Table 7: Watershed and fish variables regressed against Axes 1 and 2 in non-metric 
multidimensional scaling (NMDS) space based on abundance of 20 most 
common fish species in the 15 study watersheds.  Values are R2 that were 
significant (p < 0.05).................................................................................43 
 
Table 8: Pearson product-moment correlation coefficients and associated p values of fish 
species associations with Axes 1 and 2 of a non-metric multidimensional scaling 
(NMDS) ordination based on abundance of 20 most common fish species in the 
15 study watersheds. Significant correlations and associated p values are shown 
in bold.......................................................................................................44 
 
Chapter 3 
 
Table 1: Land cover and physical characters of study watersheds.  IS = % impervious 
surface cover, Pasture = % pasture cover, and Mixed Forest =  % managed and 
unmanaged forest cover, and LU/LC = dominant land use/land cover in 
watershed, and Spate f = spate frequency category assigned each watershed (see 
Results).  ..................................................................................................77 
 
Table 2: List of hydrological variables used in analyses and their range of values, 
Pearson correlations to predominant land cover classes in the 18 watersheds, 
and loadings on Principal Components Analysis (PCA) axes.  IS = proportion of 
impervious surface cover, Forest = proportion of forest (managed + unmanaged) 
cover, Pasture = proportion of pasture cover, PCA1 and PCA2 = Principal 
Components Analysis axes 1 and 2.  *  p < 0.05, ** p < 0.01.....................78 
 
Table 3: Physicochemical and habitat (environmental) variables used in analyses, the 
range of values, and their Pearson correlations to predominant land cover 
classes in the 18 watersheds.  IS = proportion of impervious surface cover, 
Forest = proportion of forest cover (managed + unmanaged), Pasture = 
proportion of pasture cover.  *  p < 0.05, ** p < 0.01, p < 0.001***. .........80 
 
 
 xv
Table 4: Associations between select hydrologic variables and environmental variables.  
Values are Pearson correlation coefficients, and only those with a significant 
correlation are shown.  *  p < 0.05, ** p < 0.01, p < 0.001***...................81 
 
 
Table 5: Fish species collected and Pearson correlation coefficients to non-metric 
multidimensional scaling (NMDS) axes 1 and 3.  Rare species excluded from 
NMDS analyses are denoted by dashes. Breeding guild abbreviations: S = 
Simple miscellaneous, L = Simple lithophil, Cpc = Complex with parental care, 
Cnc = Complex with no parental care.  Feeding guild abbreviations: P = 
Predator, I = Insectivore, H = Herbivore, O = Omnivore, F = Filterer........82 
 
Table 6: Fish metrics and their correlations with watershed LU/LC and best NMDS axes.   
Only significant correlations are shown.  IS = proportion of impervious surface 
cover, Forest = proportion of forest cover, Pasture = proportion of pasture 
cover, NMDS
1
 and NMDS
2
 are NMDS axes 1 and 2, respectively.  *  p < 0.05, 
** p < 0.01................................................................................................83 
Table 7: Multiple regression models for fish variables with standardized regression 
coefficients and R
2
adj
.  Fish metric definitions are as in text, others  are as in 
Tables 2 and 3.  *  p < 0.05, ** p < 0.01, *** p < 0.001.............................84 
 
Chapter 4 
 
Table 1: Land cover and size of study watersheds.  IS = % impervious surface cover, 
Grass/Transistion = % pasture cover, Deciduous = % deciduous forest cover, 
Evergreen =  % evergreen forest cover, and LU/LC = dominant land use/land 
cover in watershed .................................................................................. 120 
 
Table 2: List of hydrological variables used in analyses, their range of values, and 
significant Pearson correlations to percentage of watershed as impervious (% 
IS), pasture (% P), evergreen forest (% EF) and deciduous forest (% DF).  *  p < 
0.05, ** p < 0.01 ..................................................................................... 121 
 
Table 3: Physicochemical variables used in analyses, their range of values, and 
significant Pearson correlations to percentage of watershed as impervious (% 
IS), pasture (% P), evergreen forest (% EF) and deciduous forest (% DF).          
*  p < 0.05, ** p < 0.01. .......................................................................... 123 
 
Table 4: Geomorphic variables used in analyses, their range of values, and significant 
Pearson correlations to percentage of watershed as impervious (% IS), pasture 
(% P), evergreen forest (% EF) and deciduous forest (% DF).  * p < 0.05, ** p < 
0.01......................................................................................................... 124 
 
Table 5: Mean seasonal and annual values for macroinvertebrate variables and their 
ranges over the study.  Seasonal values for each stream were used for Range 
 
 xvi
values and to acquire averages for Summer, Winter, and Spring values and total 
stream averages were used for Annual values.......................................... 125 
 
Table 6: Results from General Linear Model testing the effects of land use/land cover 
(LU/LC) and season on macroinvertebrate metrics across the study watersheds.  
For LU/LC, F
4, 55
, for Season, F
2, 55
, and for LU/LC-season interactions, F
6, 55
. 
GA-BMI = Georgia Benthic Macroinvertebrate Index............................. 126 
 
Table 7: Best models for Composite and Community macroinvertebrate responses.  
LU/LC column shows the best single land use / land cover predictor and 
Multivariate column shows the best model from multiple regressions with PCA 
axes as predictors, and.  For the Multivariate column, C-PC1 and C-PC2 are 
physicochemical PCA axes 1 and 2, respectively, and S-PC1 and S-PC2 are 
substrate PCA axes 1 and 2, respectively while LU/LC classes are as defined in 
Table 1.  GA-BMI = Georgia Benthic Macroinvertebrate Index and NMDS1 = 
axis 1 from NMDS ordination .  Models are based on mean annual data.  All 
models are of the form y = ax + b. *p < 0.05, **p < 0.01, ***p < 0.001. . 127 
 
Table 8: Loadings (eigenvalues) and axis correlations for 1st and 2nd axes of 
Hydrologic, Geomorphic, and Chemical PCA ordinations.  * p < 0.05, ** p < 
0.01, *** p < 0.001. ................................................................................ 128 
 
Table 9: Environmental variable and macroinvertebrate metric Pearson correlations with 
1st and 2nd axes of NMDS ordination.  Only variables with significant 
correlations to NMDS Axis 1 or Axis 2 are shown.  * p < 0.05, ** p < 0.01, *** 
p < 0.001. ICI = Invertebrate Community Index  and GA-BMI = Georgia 
Benthic Macroinvertebrate Index ............................................................ 129 
 
Chapter 5 
 
Table 1: Land cover and physical characters of study watersheds.  Watershed is the 
watershed size, Stream = mean depth x width for the stream draining the 
watershed, %IS = % impervious surface cover, %Pasture = % pasture cover, 
and % Forest = % forest cover, and LU/LC = dominant land use/land cover in 
watershed................................................................................................ 166 
 
Table 2: Resource, environmental, and size variables used in analyses, their range of 
values, and their Pearson correlations to predominant land cover classes in the 
12 watersheds.  %IS = proportion of impervious surface cover, %Forest = 
proportion of forest cover, %Pasture = proportion of pasture cover.  *  p < 0.05, 
** p < 0.01.............................................................................................. 167 
 
Table 3: Taxa collected for stable isotope analysis over the 12 watersheds............ 168 
 
 
 xvii
Table 4:  Number of different taxa collected from each Functional Group in each 
watershed.  For invertebrates, CF = collector/fileters, CG = collector/gatherers, 
PA = parasites, PR = predators, SC = scrapers, SH = shredders, and SV = 
scavengers (crayfish).  For vertebrates, HBV = herbivores, ISV = insectivores, 
OMN = omnivores, PSV = piscivores, SAL = salamander, TOAD =  toad.  Site 
codes are as in Table 1.  Only invertebrates that could be identified to family-
level were included. ................................................................................ 170 
 
Table 5: Stable isotope results for 12 watersheds.  ?
13
C and ?
15
N values are site averages, 
TP is trophic position, FCL is maximum trophic position (food chain length), 
Top Consumer is the organism with the maximum trophic position, and FG is 
the functional group of the top consumer (see Table 3 for abbreviation 
explanations)........................................................................................... 171 
 
Table 6: Significant correlation coefficients between stable isotope results and land 
cover, resource, environmental and system size.  Variable definitions are as in 
Tables 1 ? 2  *  p < 0.05, ** p < 0.01....................................................... 172 
 
 
 
 
  
 
 
 
 xviii
LIST OF FIGURES 
 
 
Chapter 2 
 
Figure 1: Map of the 15 study watersheds (shaded regions) in the middle Chattahoochee 
River Basin in Muscogee, Harris, Meriwether, and Troup counties, west 
Georgia....................................................................................................45 
 
Figure 2: Non-metric multidimensional scaling (NMDS) ordination of study sites in fish 
species space.  Axes are scaled proportionate to the longest axis (% of Max). 
Symbols are the 15 study streams in different seasons, and fish species are 
labeled according to abbreviations defined in Table 8. Land use classifications 
were based on the predominant land use in the watershed (>30% of a given 
land use category), except for ?Developing? sites, which showed <30% urban 
cover but contained active residential development. Land use and natural basin 
variables significantly correlated with NMDS axes were the proportions of 
watershed impervious surface (=% Impervious Surface) and total forest cover 
(=% Forest Cover) on Axis 1, and nearest distance from the study site to a 
larger downstream tributary (=Distance to Source) on Axis 2. Arrows on x and 
y axes show direction of increasing values...............................................46   
 
Chapter 3 
 
Figure 1: Conceptual model of the potential causal relationships between land use/land 
cover, abiotic conditions, and biotic assemblages of streams....................85 
 
Figure 2: The study area included 18 small watersheds (shaded) of the Chattahoochee 
River basin in 4 counties in the Lower Southern Piedmont ecoregion.  County 
names are in capital letters. Numbers refer to watersheds and correspond to 
codes in Table 1.......................................................................................86 
 
Figure 3: Non-metric multidimensional scaling ordination (NMDS) of sites in ordination 
space.  Axes are scaled proportionate to the longest axis (% of max).  Symbols 
are the 18 study sites coded by land use classifications as described in text with 
vectors that show relative direction and strength of correlated environmental 
variables.  Arrows on x and y axes show direction of correlated fish 
assemblage values.  Axes 1 and 3 explain 57.1 and 25.3% of the total variation, 
respectively .............................................................................................87 
 
 
xix 
Chapter 4 
 
Figure 1: The study area included 20 small watersheds (shaded sections) of the 
Chattahoochee River Basin in 4 counties in the Lower Southern Piedmont 
ecoregion of western Georgia, USA.  Numbers refer to watersheds and 
correspond to Table 1. The city of Columbus is located in western Muscogee 
County.  County names are in capital letters. ......................................... 130 
 
Figure 2: Composite and community invertebrate metrics as a function of land use /land 
cover (LU/LC) category.  Horizontal lines are based on annual site means, 
symbols are seasonal site means.  Letters identify significant LU/LC category 
differences from annual General Linear Model...................................... 131 
 
Figure 3: GA-BMI (y = 11.658x
-0.3104
) as function of Impervious surface , illustrating 
potential threshold beyond which biotic integrity is compromised. ........ 132 
 
Figure 4:  Nonmetric multidimensional scaling ordination (NMDS) of sites in species 
space.  Axes are scaled proportionate to the longest axis (% of max).  Symbols 
are study sites coded by land use classifications as described in text Arrows on 
x and y axes show direction of significant correlated LU/LC and 
macroinvertebrate community / composite metric scores and vectors show 
relative direction and strength of correlated values for A) macroinvertebrate 
functional feeding and taxonomic groups and B) environmental variables.  
Axes 1 and 2 explained 79.3 and 9.8% of the total variation, respectively.  
Stress = 13.37, Final instability = 0.00001, number iterations = 66 ........ 133 
 
Chapter 5 
 
Figure 1: The study area included 12 small watersheds (shaded sections) of the 
Chattahoochee River Basin in 4 counties in the Lower Southern Piedmont 
ecoregion of western Georgia, USA.  Numbers refer to watersheds and 
correspond to Table 1. The city of Columbus is located in western Muscogee 
County.  County names are in capital letters. ......................................... 173 
 
Figure 2:  Relationship between mean trophic position and taxa richness and maximum 
food chain length and taxa richness in 12 streams.................................. 174 
 
Figure 3: Standard errors of mean ?13C and ?15N functional group values for 4 land use 
types...................................................................................................... 175 
 
Figure 4: Mean standard deviations for averaged ?
13
C and ?
15
N values in each LU/LC 
category.  There is no significant difference in the variation between 
categories. ............................................................................................. 176 
 
Figure 5: Biplots of average ?13C and ?15N (+ 1 SE) values for functional groups for 4 
 
 
xx 
land use categories (3 streams per category).  A given functional group had to 
be present in at least 2 of the 3 streams per category to be included.  Dark grey 
values are vertebrates, light grey are invertebrates, and black values are basal / 
terrestrial resources.  ............................................................................. 177 
 
Figure 6: Arrow diagrams of direction and magnitude of change for consumer trophic 
groups that correspond between compared foodwebs.  Each diagram represents 
a comparison of 1 mean foodweb to another with (A) Forest to Pasture, (B) 
Forest to Developing, and (C) Forest to Urban.  The length of the arrows 
represent the magnitude of change for each group shared in both foodwebs, 
concentric circles correspond to magnitude of change, dashed lines are mean 
vector of change, and curved dashed lines are 95% confidence intervals of the 
vector of change. ................................................................................... 178 
 
 
 
 
 
1 
CHAPTER 1.  Introduction to land use effects on stream biota. 
1.1 INTRODUCTION 
 Biotic composition and quality of aquatic systems are strongly affected by 
watershed land use (Karr et al. 1985, Lenat and Crawford 1994, Weaver and Garman 
1994, Wang et al. 2001).  In the southeastern United States, silviculture and agriculture 
are common types of traditional land use, and disturbance from each type may exert its 
own unique pressures on receiving streams (Brown et al. 2005).  However, one of the 
more pervasive and rapidly growing forms of land use change is the conversion of natural 
vegetation or agriculture to urban environments (urbanization).  Urban land use has more 
than tripled in the United States since 1950 (8,065 mi
2
 to 27,838 mi
2
 in 1990,(Wang and 
Lyons 2003a).  However, population density in these urbanized areas has actually 
decreased (6,121 to 3,411 people / mi
2
, US Census Bureau 2001) because of the 
phenomenon of increasing lower-density, primarily residential, populations (urban 
sprawl).  As a result of increased urban land use and associated urban sprawl, it is 
estimated that >60% of the world?s population will live in urbanized areas by 2030 (US 
Census Bureau 2001). 
 The inevitable change in land use/land cover (LU/LC) associated with human 
population expansion has demonstrated, far-reaching effects on stream biotic integrity.  
In general, increased LU/LC influence aquatic organisms through           
                                                           
 
 
2 
interrelated impacts on water quality, hydrology, and habitat (Hirsch et al. 1990, Herlihy 
et al. 1998, Paul and Meyer 2001, Allan 2004, Walsh et al. 2005b).  Effects of increased 
LU/LC disturbance on biota often are manifested as increased flood magnitude and 
frequency (Booth and Jackson 1997), altered streambed morphology and stability (Booth 
and Jackson 1997, Wang and Lyons 2003b) altered sediment and chemical inputs 
(Wernick et al. 1998, Allan 2004), and degraded riparian zones (Finkenbine et al. 2000, 
Groffman et al. 2003).  Many studies have described the negative effects of LU/LC on 
stream benthic macroinvertebrates (Klein 1975, Kennen 1999) and, to a lesser degree, 
periphyton (Chessman et al. 1999, Sonneman et al. 2001) and fish (Weaver and Garman 
1994, Wang et al. 2000).  However, key aspects remain unclear about specific tolerance 
thresholds of stream communities to urbanization.   The complex nature of streams has 
caused the specific mechanisms of LU/LC change, particularly urbanization, on biota and 
community interactions to be poorly understood (Paul and Meyer 2001).  Further, most 
humans dwell in urbanized areas, so it is essential, both from ecological and 
socioeconomic viewpoints, to understand the associated impacts of changing land use on 
water quality and ecosystem services of surrounding streams (Groffman et al. 2003).  
 LU/LC exerts dramatic physical and chemical effects on stream ecosystems, but 
wholesale impairment may not be evident for years or decades following development 
(Fitzgerald et al. 1998, Harding et al. 1998).  Effects often are episodic, becoming more 
pronounced seasonally or with increased precipitation, which thus requires continuous 
and expensive monitoring (Wang and Lyons 2003a).  A relatively cost-effective means of 
evaluating stream integrity is to document responses biota, particularly 
 
 
3 
macroinvertebrates and fish (Karr and Dudley 1981), which integrate dynamic 
physicochemical conditions and thus indicate stream condition (Karr et al. 1987, Barbour 
et al. 1999).  Beyond its use in biomonitoring, understanding the response of biota to 
development in impaired streams is critical information necessary for the sustainable use 
of ecosystem services. 
Benthic macroinvertebrates are particularly good indicators of localized stream 
conditions because they have limited migration patterns, have complex yet short life 
histories, and exhibit a broad range of environmental tolerance (Barbour et al. 1999).  
They are also a diverse group, filling the roles of stream predator, prey, herbivore, and 
detritivore (Allan 1995).  Examination of fish in biological monitoring efforts also is 
useful, particularly in determining long-term effects (Karr et al. 1986) because fish are 
relatively long-lived.   As a result of their varied life histories and tolerances, fish 
assemblages can reflect a broad range of environmental conditions.  Further, assemblages 
often incorporate multiple trophic levels and, thus, integrate lower trophic level effects 
(Barbour et al. 1999).   
As a result of the relatively extreme stream conditions associated with LU/LC 
change, particularly urbanization, the Harsh-Benign hypothesis (Peckarsky 1983, Menge 
and Sutherland 1987) of community regulation may apply to community interactions 
within these systems.  This hypothesis posits that in relatively harsh environments, 
abiotic regulation of community structure exerts primary control, whereas in relatively 
benign systems, biotic control of community structure is most important (Menge and 
Sutherland 1987).  Specifically, predation regulates prey populations when environmental 
 
 
4 
conditions are benign enough to support a large predator assemblage.  However, as 
environmental harshness increases, the effectiveness of predators decreases and 
competition between prey taxa increases.  As harshness increases to high levels and 
causes physical or physiological stress to taxa, biological interactions decrease in 
importance, and abiotic factors (e.g., flow, temperature, pollution) regulate community 
structure (Peckarsky 1983).    
 One of the more dramatic effects of urbanization on streams is hydrologic 
alteration, which results in increasingly harsh physical conditions, including increased 
spate flows, sedimentation, and streambed scour (Paul and Meyer 2001, Wang and Lyons 
2003b).  Species respond differently to increasing environmental harshness in streams 
(Poff and Ward 1989, Poff and Allan 1995).  As a result of their broad distribution among 
streams of various water quality, tolerant organisms may have different trophic positions 
and/or community impacts depending upon the particular stream they inhabit, which 
could lead to altered food webs in disturbed streams.  Thus, strongly contrasting 
environmental harshness associated with different LU/LC could be an important driver of 
variations in and interactions among biota inhabiting such contrasting streams.   
Many facets of LU/LC effects are known, but there are many aspects that remain 
elusive.  The primary focus of my research was to investigate the role of land use change 
on stream biota in 1
st
- to 3
rd
-order streams across an urban?rural gradient.  The primary 
objectives of this study were: 
1. To determine the relative influence of watershed LU/LC vs. natural watershed 
attributes (size, discharge, etc.) on fish assemblage structure (Chapter 2); 
 
 
5 
2. To quantify the relative explanatory power of stream hydrology, 
physicochemistry, and habitat variables associated with LU/LC change on 
variation in fish assemblages (Chapter 3);  
3. To investigate the relative seasonal influence of various LU/LC conditions on 
stream macroinvertebrate assemblages (Chapter 4); 
4. To evaluate impacts of urbanization on a common species (redbreast sunfish 
([Lepomis auritus]) by examining its feeding, growth, and trophic position 
(Chapter 5); and 
5. To examine the role of watershed LU/LC and instream conditions on food chain 
length and structure of stream food webs 
The overall goal of this research was to provide scientists and resource managers with 
basic and applied information regarding influence of changing LU/LC on aquatic biota 
which can enhance management of small watersheds in the face of anthropogenic change. 
 
 
6 
1.2 REFERENCES 
Allan, J. D. 1995. Stream ecology: structure and function of running waters. Chapman & 
Hall, New York. 
Allan, J. D. 2004. Landscapes and riverscapes: The influence of land use on stream 
ecosystems. Ann. Rev. Ecol. Syst. 35:257-284. 
Barbour, M. T., J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. Rapid bioassessment 
protocols for use in streams and wadeable rivers: periphyton, benthic 
macroinvertebrates and fish, second edition. EPA 841-B-99-002, U.S. 
Environmental Protection Agency; Office of Water, Washington D.C. 
Booth, D. B., and C. R. Jackson. 1997. Urbanization of aquatic systems: degradation 
thresholds, stormwater detection, and the limits of mitigation. Journal of the 
American Water Resources Association 33:1077-1090. 
Brown, D. G., K. M. Johnson, T. R. Loveland, and D. M. Theobald. 2005. Rural land-use 
trends in the conterminous United States, 1950-2000. Ecological Applications 
15:1851-1863. 
Chessman, B., I. Growns, J. Currey, and N. Plunkett-Cole. 1999. Predicting diatom 
communities at the genus level for the rapid biological assessment of rivers. 
Freshwater Biology 41:317-331. 
Finkenbine, J. K., J. W. Atwater, and D. S. Mavinic. 2000. Stream health after 
urbanization. Journal of the American Water Resources Association 36:1149-
1160. 
Fitzgerald, D. G., E. Kott, R. P. Lanno, and D. G. Dixon. 1998. A quarter century of 
 
 
7 
change in the fish communities of three small streams modified by anthropogenic 
activities. Journal of Aquatic Ecosystem Stress and Recovery 6:111-127. 
Groffman, P. M., D. J. Bain, L. E. Band, K. T. Belt, G. S. Brush, J. M. Grove, R. V. 
Pouyat, I. C. Yesilonis, and W. C. Zipperer. 2003. Down by the riverside: urban 
riparian ecology. Frontiers in Ecology and the Environment 1:315-321. 
Harding, J. S., E. F. Benfield, P. V. Bolstad, G. S. Helfman, and E. B. D. Jones. 1998. 
Stream biodiversity: the ghost of land use past. Proceedings of the National 
Academy of Sciences 95:14843-14847. 
Herlihy, A. T., J. L. Stoddard, and C. B. Johnson. 1998. The relationship between stream 
chemistry and watershed land cover data in the mid-Atlantic region, U.S. Water, 
Air, and Soil Pollution 105:377-386. 
Hirsch, R. M., J. F. Walker, J. C. Day, and R. Kallio. 1990. The influence of man on 
hydrologic systems. in M. G. Wolman and H. C. Riggs, editors. Surface water 
hydrology. Geological Society of America, Boulder, Colorado. 
Karr, J. R., and D. R. Dudley. 1981. Ecological perspective on water quality goals. 
Environmental Management 5:55-68. 
Karr, J. R., K. D. Fausch, P. L. Angermeier, P. R. Yant, and I. J. Schlosser. 1986. 
Assessing biological integrity in running waters: a method and its rationale. 
Illinois Natural History Survey Special Publication 5, Champaign, Illinois. 
Karr, J. R., L. A. Toth, and D. R. Dudley. 1985. Fish communities of midwestern rivers: 
a history of degradation. BioScience 35:90-95. 
Karr, J. R., P. R. Yant, and K. D. Faush. 1987. Spatial and temporal variability of the 
 
 
8 
Index of Biotic Integrity in three midwestern streams. Transactions of the 
American Fisheries Society 116:1-11. 
Kennen, J. G. 1999. Relation of macroinvertebrate community impairment to catchment 
characteristics in New Jersey streams. Journal of the American Water Resources 
Association 35:939-955. 
Klein, R. D. 1975. Urbanization and stream quality impairment. Water Resources 
Bulletin 15:948-963. 
Lenat, D. R., and J. K. Crawford. 1994. Effects of land use on water quality and aquatic 
biota of three North Carolina Piedmont streams. Hydrobiologia 294:185-199. 
Menge, B. A., and J. P. Sutherland. 1987. Community regulation: variation in 
disturbance, competition, and predation in relation to environmental stress and 
recruitment. The American Naturalist 130:730-757. 
Paul, M. J., and J. L. Meyer. 2001. Streams in the urban landscape. Annual Review of 
Ecology and Systematics. 32:333-365. 
Peckarsky, B. L. 1983. Biotic interactions or abiotic limitations?  A model of lotic 
community structure. Pages 303-324 in T. D. Fontaine III and S. M. Bartell, 
editors. Dynamics of lotic systems. Ann Arbor Science, Ann Arbor, MI. 
Poff, N. L., and J. D. Allan. 1995. Functional organization of stream fish assemblages in 
relation to hydrological variability. Ecology 76:606-627. 
Poff, N. L., and J. V. Ward. 1989. Implications of streamflow variability and 
predictability for lotic community structure: a regional analysis of streamflow 
patterns. Canadian Journal of Fisheries and Aquatic Sciences 46:1805-1817. 
 
 
9 
Sonneman, J. A., C. J. Walsh, P. F. Breen, and A. K. Sharpe. 2001. Effects of 
urbanization on streams of the Melbourne region, Victoria, Australia. II. Benthic 
diatom communities. Freshwater Biology 46:553-565. 
Walsh, C. J., A. H. Roy, J. W. Feminella, P. D. Cottingham, P. M. Groffman, and R. P. 
Morgan. 2005. The urban stream syndrome: current knowledge and the search for 
a cure. Journal of the North American Benthological Society 24:706-723. 
Wang, L., and J. Lyons. 2003a. Fish and benthic macroinvertebrate assemblages as 
indicators of stream degradation in urbanizing watersheds. Pages 227-249 in T. P. 
Simon, editor. Biological Response Signatures. CRC Press, New York. 
Wang, L., and J. Lyons. 2003b. Fish and benthic macroinvertebrate assemblages as 
indicators of stream dergadation in urbanizing watersheds. Pages 227-249 in T. P. 
Simon, editor. Biological response signatures: Indicator patterns using aquatic 
communities. CRC Press, New York. 
Wang, L., J. Lyons, P. Kanehl, and R. Bannerman. 2001. Impacts of urbanization on 
stream habitat and fish across multiple spatial scales. Environmental Management 
28:255-266. 
Wang, L., J. Lyons, P. Kanehl, R. Bannerman, and E. Emmons. 2000. Watershed 
urbanization and changes in fish communities in southeastern Wisconsin streams. 
Journal of the American Water Resources Association 36:1173-1189. 
Weaver, L. A., and G. C. Garman. 1994. Urbanization of a watershed and historical 
changes in a stream fish assemblage. Transactions of the American Fisheries 
Society 123:162-172. 
 
 
10 
Wernick, B. G., K. E. Cook, and H. Schreier. 1998. Land use and streamwater nitrate-N 
dynamics in an urban-rural fringe watershed. Journal of the American Water 
Resources Association 34:639-650. 
 
 11
CHAPTER 2.  Detection of biotic responses to urbanization using fish assemblages  
from small streams of western Georgia, USA. 
2.1  ABSTRACT 
I examined relationships between stream fish assemblages and land use alteration 
associated with urbanization in 15 lower Piedmont watersheds, along an urbanization 
gradient north of Columbus, western Georgia.  Based on land cover data from 2002 
Landsat 7 TM imagery aerial photos, streams drained watersheds that were largely urban, 
developing (suburban), agricultural (pasture), managed pine forest, and unmanaged 
mixed-forest.  I quantified fish seasonally from 3 run-pool segments in each stream, and 
used a variety of metrics as response variables in analyses of relationships between fish 
assemblage structure and land use and natural basin variation.  In general, Georgia-Index 
of Biotic Integrity (GA-IBI) values, Bray-Curtis faunal similarity of streams to mean 
conditions within reference streams, and proportions of fish as lithophilic spawners 
decreased while fish with eroded fins, lesions, tumors decreased with increasing 
urbanization.  Multiple regression indicated that assemblages were explained by a 
combination of land use and natural basin variables (basin size, average discharge, 
nearest distance to a larger downstream tributary [colonization source]), with land use 
variables being important predictors of summer assemblages and natural basin variables 
being more important in winter and spring assemblages.  Non-metric multidimensional 
 
 12
scaling (NMDS) ordinations revealed strong separation between assemblages in urban 
watersheds and forested watersheds, whereas assemblages in agricultural and developing 
watersheds were intermediate between those in urban and forested watersheds. Our data 
suggest that fish are reliable indicators of anthropogenic disturbance at the landscape 
scale, at least seasonally, and may be used to forecast the magnitude of landscape-level 
changes in stream structure and function associated with the conversion of forests to 
urban/suburban land in the Southeast.   
2.2  Introduction 
 Environmental quality and biotic composition of aquatic systems often are 
strongly affected by land use (Lenat and Crawford, 1994; Weaver and Garman, 1994; 
Wang et al., 2001).   One of the more pervasive and rapidly growing forms of land use 
change is the conversion of natural forested or vegetated agricultural land to urban 
environments (urban sprawl, US Census Bureau, 2001; Wang and Lyons, 2003).  
Inevitable landscape alteration associated with human population expansion has revealed 
dramatic effects on water quality and aquatic biota (Benke et al., 1981; Hirsch, 1990; 
Limburg and Schmidt, 1990; Weaver and Garman, 1994).  Fish are particularly useful in 
assessing long-term environmental impacts of urbanization because they integrate 
multiple trophic levels in aquatic communities, are relatively long-lived, and are easily 
sampled (Karr, 1987; Barbour et al., 1999).  In this context, increasing environmental 
stress associated with watershed urbanization often may decrease overall fish richness, 
abundance, and diversity, and cause shifts in assemblages from intolerant to tolerant 
and/or introduced species (Klein, 1979; Marsh and Minckley, 1982; Onorato et al., 1998; 
Walters et al., 2003a).   
 
 13
 Aquatic system response to urbanization often is linked to increased impervious 
surfaces within developing watersheds, which can alter stream water quality, quantity, 
and habitat availability (Hirsch, 1990; Herlihy et al., 1998; Paul and Meyer, 2001; Walsh 
et al., 2001). Such physicochemical alterations may manifest as increased flood 
magnitude and frequency (flashiness) and increased delivery of nutrients, metals, 
pesticides, and organic contaminants to receiving streams, all of which may degrade fish 
assemblages (Weaver and Garman, 1994; Lenat and Crawford, 1994; Wang et al., 2000; 
Paul and Meyer, 2001; Wang et al., 2003).  A particularly pervasive impact of forest land 
conversion on fish is through increased sedimentation (Tebo, 1955; Berkman and Rabeni, 
1987; Sutherland et al., 2002; Iwata et al., 2003).  Sedimentation in streams resulting 
from agriculture and logging are well known (Tebo, 1953; Cordone and Kelly, 1961; 
Waters, 1995; Rier and King, 1996), and emerging research suggests sedimentation in 
urbanized watersheds can reduce habitat quality and cause shifts in fish feeding and 
reproductive guilds toward generalist species (Berkman and Rabeni, 1987; Schleiger, 
2000; Walters et al., 2003a).   
 In high-gradient upland streams of northern Georgia, USA, a region with 
naturally high fish diversity and endemism, urbanization has been implicated in increased 
abundance of cosmopolitan species, or stream ?homogenization? (sensu Walters et al., 
2003a).  Fish assemblages in sandy, low-gradient streams within this same region are 
naturally less diverse and show comparatively lower endemism (Swift et al., 1986).  
Urbanization of lowland streams may, therefore, exhibit disparate influences on fishes 
than in upland systems.  I quantified fish assemblage structure in these lowland streams 
along an urbanization gradient, to assess the degree to which assemblages reflected 
 
 14
variation in land use from urbanization.  Specifically, I examined 1) the relative influence 
of watershed land use versus natural basin attributes (primarily physical habitat 
measures) on fish assemblage structure, and 2) if assemblage shifts varied predictably 
with increasing land use change attributable to urbanization in study watersheds.  
2.3 METHODS 
2.3.1  Study Area 
 I studied tributaries of the middle Chattahoochee River, western Georgia, USA, 
occurring in the Southern Outer Piedmont ecoregion.  The Piedmont biogeoclimatic 
province historically was developed for agriculture, although it now harbors many of the 
burgeoning metropolitan areas of the Southeastern US (Richmond, Raleigh/Durham, 
Greenville/Spartanburg, Atlanta), as well as extensive managed pine plantations.  The 
west Georgia landscape and the area surrounding the city of Columbus is a notable 
example of this trend (Lockaby et al., 2005).  Present-day development from Columbus is 
constrained by the Chattahoochee River to the west and a large military installation (Fort 
Benning) to the southeast, so most land conversion and urban expansion occurs mainly to 
the northeast.   
 I sampled fifteen 2
nd 
- to 3
rd
-order streams in 15 watersheds (4?25 km
2
) along an 
urbanization gradient stretching from the geologic fall line in the city of Columbus to an 
area 80 km northeast (Fig. 1).  Study streams were typical of those in the lower Piedmont, 
consisting of sandy-bottom channels and a run-pool morphology with infrequent riffles 
(Mulholland and Lenat, 1992). Average pool depth was ~0.25m and channel width ~10 
m.  Watersheds ranged in land use/cover from intense urbanization (up to 50% urban 
cover) and active suburban development to heavily forested (up to 95% forest cover).  
 
 15
This relatively large range in landscape character allowed us to compare across 
geomorphically similar streams that differed primarily in watershed-level land use and 
associated variation in streamwater physicochemical conditions (Schoonover et al., 
2008).   
2.3.2  Landscape Classification 
 Initially, land use/cover was quantified in the study watersheds using 2002 
Landsat 7 TM imagery (30-m resolution) and ArcView? software (Version 3.2a, 
Environmental Systems Research Institute, Inc., Redlands, California).  I derived 4 
primary land use categories (showing >30% of a given category) from this classification: 
% of the watershed occurring as urban (=% Urban), % of the watershed as coniferous 
forest (=% Conifer), % of the watershed as mixed coniferous-deciduous forest (=% 
Mixed), and % of the watershed as pasture (=% Pasture).  I also classified 3 forested 
watersheds that contained active or recent development into a separate category 
(Developing). In addition to quantifying % Conifer and % Mixed I also combined these 2 
classifications into a single forest category (=% Total Forest) for each watershed. 
Following analysis of Landsat data, the amount of impervious surface in study 
watersheds was determined by manually digitizing 1-m resolution, 3-band aerial 
photographs (taken March 2003). I determined watershed boundaries, size, average slope, 
and nearest distance to a larger downstream tributary (used as a conservative measure of 
potential source habitat for fishes) from USGS 30-m resolution digital elevation models 
(DEMs) and ArcView.  
 
 
 
 16
2.3.3  Field sampling 
 I quantified fish seasonally (summer/fall, winter, spring) from July 2002 to 
September 2003 from 3 run and pool habitats per stream along a representative 100-m 
reach. I sampled fish in most habitats to depletion with block nets and backpack 
electroshocker (Smith-Root LR-24), and seining.  I excluded juvenile fish (<20 mm total 
length, TL) from analyses because of inefficiency in their capture.  I identified fish to 
species, measured TL, and then examined them for eroded fins, lesions, tumors, and 
overall health (% DELT, Schleiger 2000). Except for voucher specimens, I returned all 
fish to the stream near the point of capture. Three streams (SB2, MU3, HC) could not be 
sampled in Summer 2002 because of stream drying, and 1 stream (MU1) contained no 
fish in Winter 2002, so I excluded these streams from analyses.   
 I classified species into breeding and feeding guilds, and assigned them to 
intolerant, moderately intolerant, tolerant, or pioneer tolerance classes (S. L. Schleiger, 
Georgia Department of Natural Resources, personal communication, Table 2).  I defined 
pioneers as those species that quickly reinvade a previously disturbed habitat (Smith 
1979).  I assigned fish to 5 breeding guilds (see Muncy et al., 1979; Berkeman and 
Rabeni 1987).  Fish were first classified into complex or simple breeders, based on the 
degree to which species prepare spawning sites, defend nests, and show pre-spawning 
social behavior (Pflieger, 1975; Trautman, 1981).  I further divided complex breeders 
into those species showing parental care (=P/C) and those that do not (= No P/C).  I 
divided simple breeders into spawners requiring clean, gravel substrate (lithophilous 
spawners, = Lithophils) and those capable of spawning on sand, silt, or vegetation 
(generalist spawners, = Simple Spawners).  I also calculated several community variables 
 
 17
(Table 2) including density, species richness, diversity (Shannon?s H?), and relative 
abundance for each stream and season.  
I used Bray-Curtis similarity index (Krebs 1999) to determine the pairwise 
similarity of fish assemblages at each stream based upon mean conditions from 4 
reference watersheds (MK, CB, BC, MU3, Table 1). I considered these sites to be close 
to or at the reference condition, based on 1) a high proportion of their watersheds in 
forest and correspondingly low silviculture, agriculture, or urbanization, and 2) their 
generally high apparent stream biotic integrity and physical condition.  I used tolerance 
values and breeding and feeding guilds in an Index of Biotic Integrity (IBI) developed for 
streams fishes in west-central Georgia (Schleiger, 2000).   
I also measured stream discharge monthly over the sampling period using the 
velocity-area method (Gore, 1996) with a Marsh-McBirney flow meter (Schoonover et 
al., 2005). I used mean discharge along with basin size, average slope, and downstream 
distance to the next largest stream as measures of variation in hydrologic and 
geomorphologic conditions among study watersheds, and as a suite of natural basin 
variables to contrast with land use variables in statistical analyses. 
2.3.4  Data Analyses 
  First, I used simple Pearson correlations (Zar 1999) to examine general 
relationships between fish variables and land use/natural basin variables. This procedure 
allowed us to assess general trends and identify potential multicollinearity among 
variables. Second, I used stepwise multiple regressions to develop season-specific models 
of relationships among fish and land use/natural basin variables (30-m resolution 
classification, Table 3) to assess the relative strength of land use vs. natural basin 
 
 18
variables on fish variables. I used variance inflation factors (VIF) to reduce the number of 
environmental variables in multiple regressions and avoid multicollinearity (variable 
removed if VIF >10, Myers, 1990).  Model selection was based on Mallow?s C(p), R
2
adj
, 
and parsimony (Myers, 1990).  Third, in addition to multiple regressions applied to 30-m 
spatial data, I used simple linear regressions between % impervious surface in the 
watershed, quantified from the 1-m resolution spatial data, against fish variables.  
Impervious surface is considered a useful landscape metric in studies characterizing 
urbanization impacts on streams (McMahon and Cuffney, 2000; Walsh et al., 2004), so I 
considered this analysis potentially useful in describing additional variation between fish 
and land use variables. Finally, I used non-metric multidimensional scaling (NMDS) to 
describe overall variation in fish assemblages (using species abundance data) among sites 
and seasons. NMDS is an ordination technique that uses pairwise similarity or 
dissimilarity matrices to determine positions of sites in terms of species space (Hawkins 
et al., 1997; McCune and Grace, 2002). Sites nearest to or furthest from each other on the 
ordination are those displaying highest and lowest faunal similarity, respectively.  I 
transformed proportional relative abundance data using arcsin-square root to satisfy 
assumptions of normality and homoscedasticity, and I excluded rare species (those in 
<10% of sites) for each season to reduce the influence of rare taxa on ordinations.  This 
step resulted in a 42 x 20 site by species matrix on which I based ordinations using a 
Sorenson distance measure.  I used SAS (version 8.2, SAS Institute Inc., Cary, North 
Carolina) for all correlation and regression analyses and PC-ORD (version 4, MjM 
Software Design, Glenden Beach, Oregon) for NMDS. 
 
 
 19
2.4  RESULTS 
 I collected 33 fish species (3772 individuals) from 7 families (Catastomidae, 
Centrarchidae, Cyprinidae, Ictaluridae, Percidae, Petromyzontidae, Poeciliidae) during 
the study. Centrarchidae and Cyprinidae were the most common families, composing 24 
and 55% of total fish collected, respectively.  Bandfin shiners (Luxilus zonistius), bluegill 
(Lepomis macrochirus), redbreast sunfish (L. auritus) and mosquitofish (Gambusia 
affinis) were the most abundant species.   Lepomis auritus was the most frequently 
collected species, occurring in 90% of samples, L. macrochirus was the 2
nd
-most frequent 
(74% of samples), and the silverjaw minnow (Notropis buccata) the 3
rd
-most frequent 
fish in collections (62%).   
 The number of fish collected ranged from 2 to 284 per stream, richness from 2 to 
16, H? from 0 to 2.24, and IBI score from 24 to 52 (Table 4).  Insectivores (~82% of total 
assemblage) and complex breeders with parental care (P/C, ~34%) were the most 
abundant feeding and breeding guilds, respectively (Table 4).  In general, IBI, Bray-
Curtis similarity to mean reference condition, and % of the assemblage as lithophilic 
spawners were negatively correlated with % Urban, whereas % DELT and % of the 
assemblage as herbivores (primarily Campostoma pauciradii) were positively correlated 
with % Urban, and in turn negatively correlated with % Total Forest (Table 5).  In 
particular, abundance of N. buccata, L. zonistius, Nocomis leptocephalus, N. longirostris, 
and Semotilus atromaculatus decreased with increasing % Urban. Proportion of the 
assemblage as tolerant species was positively correlated with % Pasture but was 
uncorrelated with % Urban (Table 5).  The % of the assemblage as lithophilic spawners, 
 
 20
IBI, and Bray-Curtis similarity to reference condition all were significantly correlated 
with % Conifer, % Mixed, and/or % Total Forest in study watersheds (Table 5). 
 As a result of multicollinearity determined by inspection of high VIF scores, I 
used only 5 independent variables in multiple regressions, including 2 land use variables 
(% Urban, % Pasture) and 3 natural basin variables (basin size, average discharge, nearest 
distance to a downstream source).  I observed significant relationships between the above 
independent variables and 10 fish variables, although the strength of relationships varied 
seasonally (Table 6).  Percent DELT and IBI were the only fish variables that were 
consistently explained by land use (vs natural basin) variables.  The % DELT was 
explained by % Urban in all 3 seasons (summer: R
2
adj
 = 0.755, p =0.0002; winter: R
2
adj
 = 
0.591, p = 0.0008; spring: R
2
adj
 = 0.582, p = 0.0006), whereas IBI were explained by % 
Urban and % Pasture, but only in summer (R
2
adj
 = 0.495, p = 0.0187, Table 6).   
 In contrast, several fish variables were best explained by natural basin variables.  
H? was explained by basin size and average discharge (R
2
adj
 = 0.342, p = 0.0321), but 
only in winter.  Proportion of the assemblage as cyprinid insectivores was explained by 
basin size and distance to a larger downstream source in summer (R
2
adj
 = 0.433, p = 
0.0314), and by basin size and discharge in winter (R
2
adj
 = 0.406, p = 0.0027).  Proportion 
of the assemblage as pioneer species and omnivores both were explained by the distance 
to a downstream source in some (but not all) seasons (Table 6).   However, variation in 
most fish metrics was explained by a mixture of land use and natural basin variables.  For 
example, richness and % tolerant species were explained by a combination of % Urban, 
% Pasture, and basin size (R
2
adj
 = 0.510 p = 0.0335; R
2
adj
 = 0.660 p = 0.0082, 
respectively) in summer.  In winter, IBI and proportion of the assemblage as herbivores 
 
 21
were explained by % Urban and basin size (R
2
adj
 = 0.529, p = 0.0064; R
2
adj
 = 0.681, p = 
0.0007, respectively).  In spring, % Urban, % Pasture, and basin size best explained IBI 
(R
2
adj
 = 0.638, p = 0.0024) and % Pasture and average discharge explained proportion of 
the assemblage as tolerant species (R
2
adj
 = 0.341, p = 0.0324, Table 6). 
 Use of finer scale (1-m resolution) % impervious surface data in simple 
regressions did little to improve relationships between and land use and fish assemblages 
(Table 6).  I found significant relationships for only 3 fish variables: % DELT increased 
with increasing impervious surface, and IBI decreased with increasing impervious 
surface (all seasons), whereas % herbivores increased with increasing impervious surface 
(summer and winter, Table 6). 
 NMDS revealed 2 axes that together accounted for 65.9% of the total variation in 
the fish assemblage among sites and seasons (Fig. 2).  Axes 1 and 2 accounted for 21.1 
and 44.8% of the total variation, respectively (number of dimensions = 3, number of 
iterations = 162, final stress = 14.21%).   Percent impervious surface (R
2
 = 0.34, p < 
0.0001) and distance from the study site to a larger downstream source (R
2
 = 0.20, p = 
0.0033) were both negatively related to Axis 1 (Table 7), whereas % Conifer (R
2
 = 0.43, 
p < 0.0001) and % Total Forest (R
2
 = 0.40, p < 0.0001) were positively related to Axis 2 
(Table 7, Fig. 2).  Streams in urban and developing watersheds tended to group to the left 
of the ordination, whereas streams draining watersheds with less urbanization, including 
managed (high % Conifer) and unmanaged (high % Mixed) streams, grouped to the right 
(Fig. 2). Streams within agricultural watersheds (high % Pasture) showed no clear 
grouping with respect to Axes 1 and 2 (Fig. 2). 
 
 22
 Axis 1 of the NMDS ordination was defined largely by abundance of pioneer 
species, primarily G. affinis and C. pauciradii, and 2 Lepomis species (Table 8, Fig. 2).  
In contrast, Axis 2 was defined mostly by abundance of lithophilic spawners, particularly 
N. leptocephalus and L. zonistius (Table 8, Fig. 2).  IBI and Bray-Curtis similarity values 
were positively associated with Axis 2 (R
2
 = 0.38, p < 0.001; R
2
 = 0.36, p < 0.0001, 
respectively, Table 7), whereas % DELT was negatively associated with this axis (R
2
 = 
0.22, p = 0.0023).  
2.5  DISCUSSION 
 Our results provide correlative evidence that watershed urbanization and the 
concomitant reduction in forest cover may exert strong negative impacts on stream fish 
assemblages.  These data corroborate earlier findings of stream fish studies in other 
urbanizing areas of the United States (e.g., Weaver and Garman, 1994; Wang et al., 2000; 
Walters et al., 2003a). In other studies, impacts of urbanization often manifest as 
decreased richness, diversity, sensitive species, and fish health as well as increased 
tolerant and introduced species (Roth et al., 1996; Wang et al. 2000).  I observed many of 
these same patterns in west Georgia streams. Fish health (as indicated by the % DELT), 
the proportion of fish in environmentally sensitive breeding guilds (% lithophilic 
spawners), and measures of fish biotic integrity (IBI values, Bray-Curtis similarity 
scores) all generally declined with increasing watershed urbanization (Table 5).  
Proportions of tolerant fish species increased with increasing watershed pastureland, 
although, somewhat surprisingly, this variable was unassociated with watershed 
urbanization. 
 
 
 23
2.5.1  Seasonality of fish response 
The purported mechanisms of declines in fish health and assemblage structure in 
urbanized streams often stem from altered physicochemical and hydrologic conditions 
(Booth and Jackson, 1997; Sutherland et al., 2002; Schoonover et al., 2005).  I suspect 
alterations of the hydrograph (i.e., increased flashiness and attendant increases in bed 
shear stress) in urbanized and developing watersheds, and increased sedimentation in the 
pasture-dominated watersheds, are important drivers of fish assemblage structure in west 
Georgia streams.   Several anticipated relationships were not observed, however, such as 
decreasing species richness and diversity with increasing urbanization.  Although 
difficult to reconcile, it is tempting to suggest that this disparity results from relatively 
high abundance of cosmopolitan fish species in our study streams, possibly resulting 
from a combination of stable biogeography and an extended history of human landscape 
alteration in the region (Smith, 1981; Hilliard, 1984; Swift et al., 1986; Feminella, 2000).    
 I observed considerable seasonal variation in the importance of watershed land 
use versus natural basin variables on fish assemblages.  Whereas 2 fish variables (% 
DELT, IBI) showed consistent relationships with land use in every season (Table 6), 
relationships between land use/natural basin attributes and virtually all other fish 
variables were inconsistent across seasons.  Perhaps most indicative of urban impact was 
the strong, seasonally invariant relationship between % Urban and the % of fish with 
eroded fins, lesions, and tumors. Moreover, land use both as high % urban cover and % 
pasture was particularly important in terms of low IBI values in summer, a time when 
streams are at baseflow, and dissolved nutrients, pathogenic bacteria, and other 
contaminants may reach highest annual concentrations (Schoonover et al., 2005).   
 
 24
In contrast, the higher relative importance of natural basin variables on fish 
assemblages in winter and spring may result from a combination of a decreased influence 
of factors linked with urbanization (e.g., reduced chemical toxicity during times when 
discharge is high and streamwater concentrations are low and fish are metabolically less 
active), and an attendant increase in importance of hydrological or geomorpholgical 
factors that structure fish assemblages by physical means (e.g., high shear stress during 
peak discharge, availability of refugia).  Precipitation and discharge is typically highest 
during winter and spring in our streams (B. Helms, unpublished data).  Hence, naturally 
larger basins or those with altered hydrographs from substantial landscape change, such 
as in high urban watersheds, may show higher peak discharges and greater physical 
disturbance to fish and other biota (Shaw, 1988; Booth et al., 2002). Taken together, our 
data suggest that in addition to quantifying remote landscape features it is necessary to 
consider importance of local geomorphic factors that may influence the degree to which 
changes in landscape conditions can exert strong effects on stream biota (Process Domain 
Concept, sensu Montgomery, 1999; see also Walters et al., 2003b).  
2.5.2  Impervious surface and fish response 
 Somewhat surprisingly, use of % impervious surface at a finer scale of resolution 
(i.e., 1-m, cf 30-m spatial data) did not improve relationships between urbanization and 
fish assemblage structure, as several variables were unassociated with this measure 
(Table 6).  Watershed imperviousness has been considered a good indicator of 
urbanization and has been implicated as a main driver in urban stream degradation 
(Schueler, 1994; McMahon and Cuffney, 2000; Walsh, 2000; Paul and Meyer, 2001; 
Walsh et al., 2004).  Degradation typically occurs at a low % impervious surface (10-
 
 25
15%, Schueler, 1994; Wang et al., 2000), a level close to average imperviousness in our 
watersheds (~8%, Table 3). At least 2 explanations for this equivocal pattern exist. First, 
whereas I selected sites that spanned a wide range of imperviousness overall (0 to 37%), 
impervious surface in all but 3 watersheds (BU1, BU2, RC, Table 1) was <10%, 
including our 3 developing sites (SB1, SB2, SB4, Table1). Thus it is possible that 
generally low correspondence between fish variables and impervious surface in simple 
regression models may have resulted, at least in part, from a lack of intermediate levels of 
impervious surface watersheds in our data set.  Developing watersheds were primarily 
selected based on active development, rather than proportion of impervious surface. 
Thus, from a spatial perspective and in terms of impervious surface and forest cover, 
developing watersheds appeared more like forested than urban watersheds.  Second, 
recent studies suggest that total impervious surface may not reflect an accurate 
hydrologic connection between watershed imperviousness and actual runoff delivered to 
streams (Brabec et al., 2002; Walsh et al., 2005).  If true for our watersheds, then 
imperviousness, while relatively simple to quantify from remotely sensed imagery, may 
be a less useful measure of the linkage between landscape alteration and stream structure 
and function. However, fish assemblages in developing watersheds, in general, were 
more similar to urban watersheds than forested watersheds (Fig. 2).  Taken together, 
these data suggest that impervious surface, while a reliably persistent component of the 
urban environment, is not the sole force governing fish assemblages in these watersheds.  
Our multiple regression models suggested that fishes in these watersheds are responding 
not to a single landscape factor or driver, but rather a complex suite of anthropogenic, 
basin, stream, and seasonality influences.  
 
 26
2.5.3  Urbanization and assemblage structure 
 There was a strong shift in fish assemblage structure along our urbanization 
gradient, a result reported from other systems (Weaver and Garman, 1994; Walters et al., 
2003a).  Clear separation existed between assemblages in urban/developing streams and 
those from forested watersheds, which was largely evident by differences in proportions 
of centrarchid species (Fig. 2).  Centrarchids in our streams, primarily including Lepomis 
auritus, L. macrochirus, L. cyanellus, and Micropterus salmoides, are generally tolerant 
species that often are numerical dominants in disturbed habitats (Karr, 1981; Weaver and 
Garman, 1994).  Also locally abundant in our urban sites and curiously scarce in other 
streams was the weed shiner, N. texanus.  This coastal plains minnow was rarely found 
outside of the urban streams, yet its low abundance in streams from forested watersheds 
likely resulted more from these watersheds being at the periphery of this species? range 
(Boschung and Mayden 2004).  The bluefin stoneroller (C. pauciradii) also was abundant 
in urban streams.  Species in Campostoma are herbivores as adults and can readily 
consume large turfs of filamentous algae (Power and Matthews, 1983). Thus, the 
predominance of C. pauciradii in urban sites may be a combination of its tolerance of 
physicochemical extremes and this species? capacity to consume abundant algae resulting 
from high NO
3
?N and NH
4
?N levels in these streams (Schoonover et al., 2005, B. Helms, 
unpublished data). 
 Abundance of lithophilic spawners was clearly related to urbanization, being 
prevalent in forested watersheds and comparatively scarce in urban watersheds.  This 
pattern was largely driven by abundances of the bluehead chub (N. leptocephalus), 
bandfin shiner (L. zonistius), and the longnose minnow (N. longirostris).  Other fish 
 
 27
associated with the less-urbanized sites were the rough shiner (N. baileyi), creek chub (S. 
atromaculatus), and the filter-feeding Southern brook lamprey (Ichthyomyzon gagei).  
Some of these species patterns may reflect interspecific associations as well as 
differential environment tolerance.  For example, during spawning male bluehead chubs 
create, maintain, and guard gravel-mound nests where females deposit eggs, and males 
tolerate nest associates such as bandfin shiners and rough shiners (Johnston and 
Birkhead, 1988).  Bluehead chub and bandfin shiner abundances were highly correlated 
in our study (r = 0.684, p < 0.0001). It is unknown if nest association between bluehead 
chubs and bandfin shiners is obligatory or facultative (Johnston and Birkhead, 1988).  If 
the association is facultative, then each species may respond to stresses of urbanization 
independently.  However, if nest association between species is obligatory, then absence 
of bluehead chubs from urbanized streams will likely govern bandfin shiner abundance.  
Further, subtle changes in stream abiotic conditions (e.g. initial changes associated with 
urbanization) could alter the reproductive behavior of these fishes and increase the 
currently benign impact of introduced nest associates like N. baileyi on L. zonistius 
(Herrington and Popp 2004).  Such indirect effects of urbanization on the dynamics of 
fish assemblages are largely unknown but may be potentially far-reaching.   
 In summary, our results suggest urban land use in general, but not solely in terms 
of the proportion of impervious surface, was a strong determinant of fish assemblage 
structure.  Moreover, fish assemblages showing clear signs of deteriorating health were 
consistently observed in urban and developing streams, so urban land use appears to exert 
a stronger influence on fish assemblages than watersheds predominantly in agricultural or 
silvicultural land use.  Our results also demonstrate that streams with relatively low levels 
 
 28
of species endemism also can display dramatic shifts in assemblages in response to 
urbanization similar to systems with high endemism (Walters et al., 2003a).  However, 
increases in the magnitude of urbanization are likely to increase dramatically in future 
(Cohen 2003), so to understand the responses of fish assemblages it may be necessary to 
investigate the complex interplay among several environmental factors including land 
use, seasonality, stream geomorphology and hydrology, and biotic interactions.   
 
 29
2.6  REFERENCES 
Barbour, M.T., Gerritsen, J., Snyder, B.D., Stribling, J.B. (1999)  Rapid Bioassessment 
Protocols for Use in Streams and Wadeable Rivers: Periphyton, Benthic 
Macroinvertebrates and Fish, 2
nd
 Edition.  EPA 841-B-99-002.  U.S. 
Environmental Protection Agency; Office of Water:  Washington D.C.  
Benke, A.C., Willeke, G.E, Parrish, F.K., and  Stites, D.L.  (1981)  Effects of 
urbanization on stream ecosystems.  Completion Report OWRT Project No. A-
055-GA.  US Department of Interior, Office of Water Research and Technology, 
Ft. Collins, CO. 
Berkman, H.E. and Rabeni, C.F.  (1987)  Effect of siltation on stream fish communities.  
Environmental Biology of Fishes 18, 285?294. 
Booth, D.B. and Jackson, C.R.  (1994)  Urbanization of aquatic systems:  degradation 
thresholds, stormwater detection, and the limits of mitigation.  Journal of the 
American Water Resources Association 33, 1077?1090. 
Booth, D. B., Hartley, D., and Jackson, R. (2002) Forest cover, impervious-surface area, 
and the mitigation of stormwater impacts. Journal of the American Water 
Resources Association 38, 835?845. 
Boschung, H. T. and Mayden, R. L.  (2004)  Fishes of Alabama.  The Smithsonian 
Institute.   Washington, D.C., USA. 
Brabec, E., Schulte, A., and Richards, P. L. (2002) Impervious surfaces and water 
quality: a review of current literature and its implications for watershed planning. 
Journal of Planning Literature 16:499?514. 
Cohen, J. E. (2003) Human population: the next half century. Science 302,1172?1175. 
 
 30
 
Cordone, A.J. and Kelley, D.W. (1961) The influences of inorganic sediment on the 
aquatic life of streams.  California Fish and Game 47, 189?228. 
Feminella, J.W. (2000) Correspondence between stream macroinvertebrate assemblages 
and 4 ecoregions of the southeastern USA. Journal of the North American 
Benthological Society 19, 442?461. 
Gore, J.A.  (1996)  Discharge measurements and streamflow analysis. In Methods in 
Stream Ecology (Hauer, F. R. and Lamberti, G. A. eds.), pp. 53?74. Academic 
Press, San Diego, California. 
Hawkins, C. P., Hogue, J. N., Decker, L. M., and Feminella, J. W. (1997) Channel 
morphology, water temperature, and assemblage structure of stream insects.  
Journal of the North American Benthological Society 16, 728?749. 
Herlihy, A.T., Stoddard, J.L., and Johnson, C.B.  (1998)  The relationship between stream 
chemisty and watershed land cover data in the Mid-Atlantic region, US.  Water, 
Air, and Soil Pollution 105, 377?386. 
Herrington, S. J., and  Popp, K. J. (2004)  Observations on the reproductive behavior of 
nonindigenous rough shiner, Notropis baileyi, in the Chattahoochee river system.  
Southeastern Naturalist 3, 267-276. 
Hilliard, S.B. (1984)  Atlas of Antebellum Southern Agriculture.  Louisiana State 
University Press, USA. 
 Hirsch, R.M., Walker, J.F., Day, J.C., and Kallio, R.  (1990)  The influence of man on 
hydrologic systems. In Surface Water Hydrology The Geology of America, Vol 0-
 
 31
1.  (M.G. Wolman and H.C. Riggs eds.). Boulder CO:  Geological Society of 
America. 
Iwata, T., Nakano, S., and Inoue, M. (2003)  Impacts of past riparian deforestation on 
stream communities in a tropical rain forest in Borneo.  Ecological Applications 
13, 461?473. 
Johnston, C.E. and Birkhead, W.B.  (1988)  Spawning in the bandfin shiner, Notropis 
zonistius (Pisces: Cyprinidae).  Journal of the Alabama Academy of Sciences 59, 
30?33. 
Karr, J.R. (1981)  Assessment of biotic integrity using fish communities.  Fisheries 6, 
21?27. 
Karr, J.R.  (1987)  Biological monitoring and environmental assessment:  a conceptual  
 framework. Environmental Management 11.  249-256. 
Klein, R.D. (1979)  Urbanization and stream quality impairment.  Water Resources 
Bulletin 15, 948?963. 
Krebs, C. J.  (1999)  Ecological Methodology.  Benjamin/Cummins. USA 
Lenat, D.R. and Crawford J.K. (1994)  Effects of land use on water quality and aquatic 
biota of three North Carolina Piedmont streams.  Hydrobiologia 294, 185?199. 
Limburg, K.E. and Schmidt R.E. (1990)  Patterns of fish spawning in Hudson River 
tributaries: response to an urban gradient?  Ecology 71, 1238?1245. 
Lockaby, B.G, Zhang, D. (2004) An integrated approach to urban development impacts 
on natural resources: the WestGA project. Urban Ecosystems. 
Marsh, P.C. and Minckley, W.L.  (1982)  Fishes on the Phoenix metropolitan area in 
central Arizona.  North American Journal of Fisheries Management 4, 395?402. 
 
 32
McCune B. and Grace, J.B.  (2002)  Analysis of Ecological Communities.  MjM Software 
Design.  Gleneden Beach, Oregon, USA. 
McMahon, G., and Cuffney, T.F.  (2000)  Quantifying urban intensity in drainage basins 
for assessing stream ecological conditions.  Journal of the American Water 
Resources Association 36, 1247?1262. 
Montgomery, D. M. (1999) Process domains and the river continuum. Journal of the 
American Water Resources Association 36, 397?410. 
Mulholland, P.J. and Lenat, D.R. (1992)  Streams of the southeastern Piedmont, Atlantic 
Drainage. In Biodiversity of the Souteastern United States ? Aquatic 
Communities, (C.T. Hackney, S.M. Adams, and W.A. Martin, eds.) pp. 193?232.  
John Wiley and Sons, New York. 
Muncy, R.J., Atchinson, G.J., Bulkley, R.V., Menzel, B.W., Perry, L.G., and Summerfelt, 
R.C.  (1979)  Effects of Suspended Solids and Sediment on Reproduction and 
Early Life of Warmwater Fishes: A Review.  EPA-600/3-79-0, Covallis 
Environmental Research Laboratory, Corvallis, 100 pp. 
Myers, R.H. (1990)  Classical and Modern Regression with Applications.  2
nd
 Edition.  
PWS-KENT Publishing Company, USA. 
Onorato, D., Marion, K.R., and Angus., R.A. (1998)  Longitudinal variations in the 
ichtyofaunal assemblages of the upper Cahaba River: possible effects of 
urbanization in a watershed.  Journal of Freshwater Ecology 13, 139?154. 
Paul, M.J. and Meyer, J.L. (2001)  Streams in the urban landscape.  Annual Review of 
Ecology and Systematics 32, 333?365. 
 
 33
Pflieger, W.L. (1975) The Fishes of Missouri. Missouri Department of Conservation and 
Western Publishing Company, Columbia, Missouri. 
Power, M. E., and Matthews, W. J. (1983)  Algae-grazing minnows (Campostoma 
anomalum), piscivorous bass (Micropterus spp.), and the distribution of attached 
algae in a small prairie-margin stream. Oecologia 60:328-332. 
Rier, S.T. and King, D.K. (1996)  Effects of sedimentation and riparian clearing on 
benthic community metabolism in an agriculturally-disturbed stream.  
Hydrobiologia 339, 111?121. 
Roth, N.E., Allan, J.D., Erickson., D.L. (1997)  Landscape influences on stream biotic 
integrity assessed at multiple spatial scales.  Landscape Ecology 1, 141?156. 
Schleiger, S.L. (2000)  Use of an index of biotic integrity to detect effects of land uses on 
stream fish communities in west-central Georgia.  Transactions of the American 
Fisheries Society 129, 118?1133. 
Schoonover, J.E., Lockaby, B. G., and Pan, S. (2004) Changes in chemical and physical 
properties of stream water across an urban-rural gradient in western Georgia.  
Urban Ecosystems.  
Schueler, T. R. (1994) The importance of imperviousness. Watershed Protection 
Technology 1, 100-111. 
Shaw, E. M. (1988) Hydrology in Practice. 2nd edition. T. J. Press Ltd, Padstow, 
Cornwall. 
Smith, P.W. (1979)  The Fishes of Illinois.  University of Illinois Press, Urbana. 
Smith, G.R. (1981) Late Cenozoic Freshwater Fishes of North America. Annual Review 
of Ecology and Systematics 12, 163?193. 
 
 34
Swift, C.C., Gilbert, C.R., Bortone, S.A., Burgess, G.H., and Yerger, R.W. (1986) 
Zoogeogrpahy of the freshwater fishes of the southeastern United States:  
Savannah River to Lake Pontchartrain.  In Zoogeography of North American 
Freshwater Fishes (C.H. Hocutt and E.O. Wiley, eds.), pp. 213?255. John Wiley 
and Sons, New York. 
Sutherland, A.B., Meyer, J.L., and Gardiner, E.P. (2002)  Effects of land cover on 
sediment regime and fish assemblage structure in four southern Appalachian 
streams.  Freshwater Biology 47, 1791?1805. 
Trautman, M.B. (1981) The Fishes of Ohio, 2
nd
 Edition. Ohio State University Press, 
Columbus, Ohio. 
Tebo, L.B. Jr. (1955)  Effects of siltation, resulting from improper logging, on the bottom 
fauna of a small trout stream in the southern Appalachians.  Progressive Fish-
Culturist 17, 64?70. 
Walsh, C.J.  (2000)  Urban impacts on the ecology of receiving waters: a framework for 
assessment, conservation and restoration.  Hydrobiologia 431, 107?114. 
Walsh, C. J., Sharpe, A. K., Breen, P. F., and Sonneman, J. E. (2001) Effects of 
urbanization on streams of the Melbourne region, Victoria, Australia. I. Benthic 
macroinvertebrate communities.  Freshwater Biology 46, 535?551. 
Walsh, C.J., Papas, P.J., Crowther, D., Sim, P.T., and Yoo, J. (2004) Stormwater drainage 
pipes as a threat to a stream-dwelling amphipod of conservation significance, 
Austrogammarus australis, in south-eastern Australia. Biodiversity and 
Conservation 13, 781?793.  
 
 35
Walsh, C. J., Fletcher, T. D., and Ladson, A. R. 2005. Stream restoration of urban 
catchments through re-designing stormwater systems: looking to the catchment to 
save the stream. Journal of the North American Benthological Society 24, 690?
705.  
Walters, D.M., Leigh, D.S., and Bearden, A.B (2003a) Urbanization, sedimentation, and 
the homogenization of fish assemblages in the Etowah River Basin, USA. 
Hydrobiologia 494, 5?10.  
Walters, D.M., Leigh, D.S, Freeman, M.C., Freeman, B.J., and Pringle, C.M. (2003b) 
Geomorphology and fish assemblages in a Piedmont river basin, U.S.A. 
Freshwater Biology 48, 1950?1970. 
Wang, L., Lyons, J., Kanehl, P., Bannerman, R., and Emmons, E.  (2000) Watershed 
urbanization and changes in fish communities in southeastern Wisconsin streams.  
Journal of the American Water Resources Association 36, 1173?1189.   
Wang, L., Lyons, J., Kanehl, P., and Bannerman, R. (2001) Impacts of urbanization on 
stream and fish across multiple spatial scales.  Environmental Management 28, 
255?266. 
Wang, L. and Lyons, J. (2003) Fish and benthic macroinvertebrate assemblages as 
indicators of stream degradation in urbanizing watersheds. In Biological Response 
Signatures ? Indicator Patterns Using Aquatic Communities (T.P. Simon, ed.), 
pp. 227?249. 
Waters, T. F. (1995)  Sediment in Streams. American Fisheries Society Monograph 7. 
American Fisheries Society, Bethesda, Maryland. 
 
 36
Weaver, L.A. and Garman, G.C. (1994) Urbanization of a watershed and historical 
changes in a stream fish assemblage.  Transactions of the American Fisheries 
Society 123, 162?172. 
Zar, J.  (1999)  Biostatistical Analysis. 4
th
 Edition.  Prentice-Hall, Inc.  Upper Saddle 
River, New Jersey, USA. 
 
 
 
 37
Table 1.  Characteristics of the 15 study watersheds.  LU/LC refers to the dominant land 
use/land cover in the watershed based on GIS.  A watershed was classified only if it 
contained >30% of a given land use category, except for Developing, which showed 
<30% urban cover but contained active residential development.  UTM coordinates were 
taken at the sampling site furthest downstream. Trib. = tributary.  
 
  
          
Site Stream 
Watershed size  
(km
2
) 
LU/LC UTM North UTM East 
SB 1 Schley Creek 20.09 Developing 0685312 3608722 
SB 2 Standing Boy Creek Trib. 6.34 Developing 0693082 3614122 
SB 4 Standing Boy Creek 26.59 Developing 0696898 3612214 
HC  House Creek Trib. 6.55 Coniferous 0678280 3630695 
MU 2 Mulberry Creek Trib. 6.06 Coniferous 0709195 3621084 
SC Sand Creek 8.96 Coniferous 0680325 3635890 
BC Beech Creek 6.47 Mixed 0704322 3657675 
MK Flat Creek Trib. 6.63 Mixed 0703887 3668333 
CB Cline's Branch 8.97 Mixed 0681196 3623522 
MU 3 Turntime Branch 10.44 Mixed 0701795 3619093 
FS 1 Wildcat Creek 24.20 Pasture 0684280 3641319 
MU1 Ossahatchie Creek Trib. 11.95 Pasture 0712764 3615524 
BU 1 Lindsey Creek 25.47 Urban 0693619 3593874 
BU 2 Cooper Creek 24.69 Urban 0695357 3596969 
RC Roaring Branch Creek 3.67 Urban 0691329 3602142 
 
 
 38
Table 2.  Fish metrics quantified by stream and season in the 15 study watersheds.   
Complex breeders that lack parental care were simply classified ?Complex?. P/C = 
complex breeders showing parental care, No P/C = complex breeders showing no 
parental care. See text for definitions of breeding guilds and tolerance values. 
 
        
Diversity/similarity Feeding guild Breeding guild Tolerance value 
Shannon?s H? Filter-feeder Complex Pioneer 
Bray-Curtis similarity Insectivore Complex P/C Tolerant 
 Piscivore Complex No P/C Moderate 
 Herbivore Simple Spawners Intolerant
 Omnivore Simple Lithophils  
 Generalist  
 
 
 
 
 39
 
Table 3. Initial land use and natural basin variables quantified in the 15 study watersheds.  
Asterisks indicate those variables used in final analyses. 
 
 
Variable Mean Range 
Land use    
Urban (%)* 11.1 0.1?48.9 
Impervious surface (%)* 8.66 1.22?37.11
Pasture (%)* 24.6 4.0?52.7
Mixed Forest (%) 37.2 22.5?55.2 
Conifer (%) 26.0 8.4?69.6 
Total Forest (%) 63.2 30.9?94.8 
Natural basin    
Average discharge (m
3
/s)* 0.282 0.003?0.910 
Basin size (ha)* 1352 366?2659 
Average basin slope (%) 13.6 5.3?146.1 
Average channel slope (%) 0.05 0.02?0.08 
Distance to downstream source (km)* 2.9 0.2?10.0 
Link magnitude (count) 4.3 1.0?12.0 
 
 
 
 
 
 
 
 40
 
Table 4.  Summary fish data for the 15 study watersheds.  Values are cumulative for the 
sites (seasons combined).  IBI= GA Index of Biotic Integrity (see Schleiger 2000). % 
DELT = proportion of fish in the sample with eroded fins, lesions, or tumors. P/C = 
proportion of complex breeders showing parental care, No P/C = proportion of complex 
breeders showing no parental care. See text for further explanation of variables. 
 
 Mean Range 
Assemblage variable   
Number collected 91.9 2?284 
H' 1.47 0.17?2.24 
Species richness 8.4 2.0?16.0 
Bray-Curtis similarity 31.05 3?68 
IBI 41.37 24?52 
Tolerant species (%) 16.34 0?98.47 
Pioneer species (%) 28.56 0?82.76 
% DELT 
6.61 0?85.7 
Feeding guild variable   
Piscivore (%) 1.99 0?14 
Herbivore (%) 5.53 0?50 
Omnivore (%) 4.2 0?28 
Insectivore (%) 81.85 49?100 
Filter feeder (%) 2.71 0?44 
Generalist feeder (%) 3.64 0?23 
Breeding guild variable   
No P/C (%) 25.15 0?5.26 
P/C (%) 34.11 0?96.95 
Simple spawner (%) 17.02 0?100 
Lithophilic spawner (%) 23.49 0?78.57 
 
 
 
 
 
 
 
 
 
 
 
 
 
 41
 
Table 5.  Pearson product-moment correlation coefficients between selected fish 
variables and land use variables (based on 30-m resolution Landsat data, see text).  % 
Herbivore = proportion of herbivorous species in the sample, % Simple spawner = 
proportion of species in the sample showing no complex spawning behavior, % 
Lithophils = proportion of species in the sample requiring clean, gravel substrate for 
spawning, %Tolerant = proportion of tolerant species in the sample, % DELT = 
proportion of fish in sample with eroded fins, lesions, or tumors in the sample, Bray-
Curtis = stream faunal similarity to average reference condition, IBI= Georgia Index of 
Biotic Integrity (see Schleiger 2000). *p = 0.05, ** p = 0.01, *** p < 0.001. ns = 
nonsignificant. 
 
      
Fish variable % Urban % Pasture % Conifer % Mixed 
% Total 
Forest 
    
% Herbivore 0.4804*** ns ns ns ?0.3136* 
    
% Simple spawner ns ns ns ns ns 
    
% Lithophilic 
spawner 
?0.4454** ns 0.453** ns 0.4829*** 
    
% Tolerant ns 0.4238** ns ns ns 
    
% DELT 0.7222*** ?0.3075* ?0.371* ?0.3405* ?0.5244*** 
     
Taxa richness ns ns ns ns ns
     
H? ns ns ns ns ns
     
Bray-Curtis 
similarity 
?0.3795* ns 0.3117* 0.3001* 0.4486** 
    
IBI ?0.6012*** ns 0.4714** 0.3803** 0.6365*** 
      
 
 
 
 
 42
Table 6.  Seasonal multiple regression models for select fish metrics and simple linear 
regressions of fish metrics with 1-m resolution % impervious surface. % Herbivore 
=proportion of herbivorous species in the sample, % Omnivore =proportion of 
omnivorous species in the sample, % Cyprinid insectivore =proportion of insectivorous 
species in the family Cyprinidae, % Pioneer = proportion of pioneer species in the 
sample, % DELT = proportion of fish in sample with eroded fins, lesions, or tumors in 
the sample, Bray-Curtis = stream faunal similarity to reference stream, and IBI= Georgia 
Index of Biotic Integrity (see Schleiger 2000).  Direction of relationships for simple 
regressions are denoted by (+) and (?) in the R
2
 column when appropriate. 
  
Variables in multiple 
regression model 
  Impervious
surface 
 
   
R
2
adj p R
2 
p 
Summer   
  
 
 
H'  ---- 
---- ---- ---- ---- 
Species richness  urban, pasture, basin size 
0.4760 0.0432 ---- ---- 
Tolerant species  urban, pasture, basin size 
0.672 0.0071 ---- ---- 
% Herbivore  pasture, source distance 
0.461 0.0252 0.456 (+) 0.0160 
% Omnivore  ---- 
---- ---- ---- ---- 
% cyprinid insectivore  basin size, source distance 
0.433 0.0314 ---- ---- 
% Pioneer  source distance 
0.340 0.0274 ---- ---- 
% DELT  Urban 
0.755 0.0002 0.871 (+) <0.0001 
Bray-Curtis similarity  ---- 
---- ---- ---- ---- 
IBI  urban, pasture
0.495 0.0187 0.376 (?) 0.0341 
    
Winter   
H?  basin size, discharge 
0.342 0.0321 ---- ---- 
Species richness  ---- 
---- ---- ---- ---- 
Tolerant species  ---- 
---- ---- ---- ---- 
% Herbivore  urban, discharge 
0.681 0.0007 0.636 (+) 0.0006 
% Omnivore  ---- 
---- ---- ---- ---- 
%Cyprinid insectivore  basin size, discharge 
0.4060 0.0027 ---- ---- 
% Pioneer  ---- 
---- ---- ---- ---- 
% DELT  urban  
0.591 0.0008 0.733 (+) <0.0001 
Bray-Curtis similarity  ---- 
---- ---- ---- ---- 
IBI  urban, discharge 
0.529 0.0064 0.414 (?) 0.0130 
  
    
Spring   
H' ---- 
---- ---- ---- ---- 
Species richness  ---- 
---- ---- ---- ---- 
Tolerant species  pasture, discharge 
0.341 0.0324 ---- ---- 
% Herbivore  ---- 
---- ---- ---- ---- 
% Omnivore  source distance 
0.8183 <0.0001 ---- ---- 
% Cyprinid insectivore  ---- 
---- ---- ---- ---- 
% Pioneer  source distance 
---- ---- ---- ---- 
% DELT  Urban 
0.582 0.0006 0.720 (+) <0.0001 
Bray-Curtis similarity  pasture, discharge 
0.398 0.0190 ---- ---- 
IBI  urban, pasture, basin size 
0.6381 0.0024 0.369 (?) 0.0163 
 
 43
 
Table 7.  Watershed and fish variables regressed against Axes 1 and 2 in non-metric 
multidimensional scaling (NMDS) space based on abundance of 20 most common fish 
species in the 15 study watersheds.  Values are R
2
 that were significant (p < 0.05). 
 
 
NMDS  
Axis 1 
NMDS  
Axis 2 
Watershed variables   
% Impervious surface  0.3375 
% Pasture   
% Conifer  0.4305 
% Total Forest  0.3975 
Basin size  0.0963 
Distance from source 0.2004  
Fish variables   
% Pioneer 0.4504 0.2288 
% Omnivore 0.3218  
% Lithophilic Spawner  0.4956 
% DELT  0.2145 
Bray-Curtis similarity  0.3561 
IBI  0.3829 
 
 
 44
Table 8.  Pearson product-moment correlation coefficients and associated p values of fish 
species associations with Axes 1 and 2 of a non-metric multidimensional scaling 
(NMDS) ordination based on abundance of 20 most common fish species in the 15 study 
watersheds. Significant correlations and associated p values are shown in bold. 
 
 
  
 
Species 
Species 
Code 
NMDS 
Axis 1
p  
NMDS 
Axis 2 
p  
    
Ameirus natalis A_NATA 0.211 0.1858 ?0.175 0.2733 
Campostoma pauciradii C_PAUC ?0.599 <0.0001 0.196 0.2202 
Ericymba buccata E_BUCC 0.196 0.2190 ?0.241 0.1283 
Gambusia affinis G_AFFI 0.836 <0.0001 0.093 0.5640 
Hybopsis sp. H_SPEC ?0.262 0.0979 0.293 0.0628 
Ichthyomyzon gagei I_GAGE ?0.180 0.2588 ?0.542 0.0003 
Lepomis auritus L_AURI ?0.470 0.0002 0.430 0.0052 
Lepomis cyanellus L_CYAN ?0.416 0.0069 0.059 0.7157 
Lepomis gulosus L_GULO 0.153 0.3385 0.294 0.0624 
Lepomis macrochirus L_MACR 0.024 0.8785 0.659 <0.0001 
Luxilus zonistius L_ZONI 0.016 0.9187 ?0.820 <0.0001 
Micropterus salmoides M_SALM ?0.200 0.2089 0.210 0.1886 
Minytrema melanops M_MELA ?0.100 0.5345 0.195 0.2228 
Nocomis leptocephalus N_LEPT ?0.227 0.1530 ?0.700 <0.0001 
Notropis baileyi N_BAIL ?0.184 0.2482 ?0.319 0.0418 
Notropis longirostris N_LONG 0.099 0.5389 ?0.508 0.0007 
Notropis texanus N_TEXA ?0.203 0.2023 0.390 0.0160 
Percina nigrofasciata P_NIGF ?0.036 0.8216 ?0.271 0.0865 
Pomoxis nigromaculatus P_NIGM 0.098 0.5425 0.371 0.1710 
Semotilus atromaculatus S_ATRO ?0.098 0.5404 ?0.543 0.0002 
 
 
 45
Figure. 1.  Map of the 15 study watersheds (shaded regions) in the middle Chattahoochee 
River Basin in Muscogee, Harris, Meriwether, and Troup counties, west Georgia.  
20 0 20 Kilometers
HARRIS
MERIWETHER 
MUSCOGEE
TROUP
GA
Columbus
C
h
a
t
t
a
h
o
o
c
h
e
e
 
R
i
v
e
r
#
#
Atlanta
MK
BC
FS1
SC
HC
CB
SB1
SB2
SB4
MU3
MU2
MU1
BU2
BU1
RC
 
 
 
 46
Figure 2.  Non-metric multidimensional scaling (NMDS) ordination of study sites in fish 
species space.  Axes are scaled proportionate to the longest axis (% of Max). Symbols are 
the 15 study streams in different seasons, and fish species are labeled according to 
abbreviations defined in Table 8. Land use classifications were based on the predominant 
land use in the watershed (>30% of a given land use category), except for ?Developing? 
sites, which showed <30% urban cover but contained active residential development. 
Land use and natural basin variables significantly correlated with NMDS axes were the 
proportions of watershed impervious surface (=% Impervious Surface) and total forest 
cover (=% Forest Cover) on Axis 1, and nearest distance from the study site to a larger 
downstream tributary (=Distance to Source) on Axis 2. Arrows on x and y axes show 
direction of increasing values.  
A_NATA
C_PAUC
E_BUCC
G_AFFI
H_SPEC
I_GAGE
L_AURI
L_CYAN
L_GULO
L_MACR
L_ZONI
M_SALM
M_MELA
N_LEPT
N_BAIL
N_LONG
N_TEXA
P_NIGR
P_NIGM
S_ATRO
0
0
20 40 60 80 100
20
40
60
80
100
Axis 2 (% of max)
Ax
is 1 (%
 o
f
 m
ax)
Land use
Mixed
Managed
Pasture
Developing
Urban
 
 
 
 
 
% Forest  
% Impervious  
 
 47
CHAPTER 3.  Assessing influences of hydrology, physicochemistry, and habitat on 
stream fish assemblages across a changing landscape 
3.1 ABSTRACT 
I evaluated the indirect impact of land use/land cover on fish assemblages by 
examining relationships between stream hydrology, physicochemistry, and instream 
habitat and their association with fish responses over a 12-mo period of record (POR) in 
streams draining 18 watersheds of the Lower Piedmont of western Georgia, USA.   
Several important relationships between watershed land use and physicochemical, 
hydrological, and habitat parameters were observed, particularly increased frequency of 
spate flows and decreased dissolved oxygen with increasing impervious surface cover, 
increased habitat quality with forest cover, and increased suspended solids with 
increasing pasture cover.  Fish assemblages however were largely explained by 
physicochemical and hydrological rather than habitat variables.  Specifically, fish species 
diversity, richness, and biotic integrity were lower in streams that received high 
frequency of spate flows.  Also, overall fish assemblage structure (as determined by non-
metric multidimensional scaling) was best described by dissolved oxygen (DO), with low 
DO streams containing sunfish-based assemblages and higher DO streams containing 
minnow-based assemblages.  Our results suggest that altered hydrological and 
physicochemical conditions, induced largely by impervious surface, may be a strong 
 
 48
driver of fish assemblage structure in these lowland streams.   These analyses allow for a 
more mechanistic understanding of how land use ultimately affects these systems. 
3.2 INTRODUCTION 
  Biotic patterns in stream communities often are attributable to the combined 
influences of broad-scale environmental factors, regional species pool, watershed-specific 
processes, and local conditions (Frissell et al.,1986; Poff, 1997; Fausch et al., 2002).  
Human activities at the landscape level can affect these filters and thus have dramatic 
effects on stream community structure and function.  In particular, watershed land use 
and/or land cover (LU/LC) can alter local conditions by directly affecting water 
physicochemistry, hydrology, and instream habitat, which, separately or in combination, 
in turn can influence biotic composition and ecological integrity (Lenat and Crawford, 
1994; Clements, 2000; Paul and Meyer, 2001; Allan, 2004; Schoonover et al., 2006).  
 Increased levels of agriculture and urbanization in watersheds can lead to several 
significant changes to many stream features.  In general, agricultural and urbanized land 
has been implicated in increased streamwater pollutants, decreased riparian cover, 
elevated water temperatures, altered hydrology, increased storm flows and sedimentation, 
and overall reduced habitat quantity/quality (Paul and Meyer, 2001; Allan, 2004).   All of 
these impacts have been shown to decrease biotic integrity, such as reducing species 
richness/diversity, increasing physiological stress, and causing assemblage shifts (Scott et 
al., 1986; Schleiger, 2000; Walters et al., 2003; Weaver and Garman, 1994; Helms et al., 
2005; Roy et al., 2005a).   
 Of the multiple direct abiotic consequences watershed land use has on receiving 
streams, hydrological alteration is one of the more obvious and pervasive (Booth and 
 
 49
Jackson, 1997; Groffman et al., 2003; Wang and Lyons, 2003; Walsh et al., 2005).  As a 
result of high levels of watershed imperviousness and connectedness, streams draining 
urban and urbanizing watersheds often display flashy hydrographs with multiple peak 
flows and reduced base flows (Ferguson and Suckling, 1990; Rose and Peters, 2001; 
Schoonover et al., 2006).  Storm flows often increase in magnitude and frequency in 
agricultural settings because of the use of drainage ditches, loss of wetlands, and soil 
compaction (Peterson and Kwak, 1999; Allan, 2004).  Such hydrological alteration can 
have far-reaching effects on instream conditions.  Increased peak flows can accelerate 
geomorphic changes in stream channels, leading to increased sedimentation, scour, and 
channelization, the combination of which may reduce biotic habitat quality and quantity 
(Wolman, 1967; Hammer, 1972; Bledsoe and Watson, 2001).  In addition, stormwater 
runoff in urbanized watersheds often elevates chemical pollutants, including nutrients, 
metals, pesticides, and pharmaceuticals (Paul and Meyer, 2001; Kolpin et al., 2002; 
Wang and Lyons, 2003), and water temperature. Changes in temperature may cause 
thermal pulses and altered thermal regimes in receiving waters, which can increase 
mortality of sensitive species and skew assemblages towards tolerant species (Galli, 
1991; Wang et al., 2000; Krause et al., 2004).   
Thus conceptually, watershed LU/LC can directly alter hydrological regimes 
which in turn lead to degradations in physicochemical and geomorphic conditions.  
Watershed LU/LC also can directly influence physicochemical (e.g. point-source 
pollution) and geomorphic conditions (e.g. livestock trampling). Altered hydrological 
regimes can then directly influence biota and/or indirectly through alterations in 
physicochemical and geomorphic conditions (Figure 1).  
 
 50
 In a previous study, I described the relationships between urbanization and fish 
assemblage structure in streams of western Georgia, USA (Helms et al., 2005).  There I 
reported that declines in biotic integrity and assemblage shifts were associated with 
watershed land use as well as broad physical/environmental features in high flow 
seasons.  As a result, I suspected that differences in hydrographs across these watersheds 
were important in explaining fish assemblages.  Our present study therefore was designed 
to investigate in more detail the association of land use and altered hydrology on stream 
fish assemblage structure in this area in a subsequent year.  Specifically, I examined the 
direct relationships among 1) watershed land use and the physical instream factors 
hydrology, habitat, and water physicochemistry; and 2) these physical instream factors 
with fish assemblages.  Our objective was to determine relative explanatory power of 
hydrology, physicochemistry, and habitat variables associated with LU/LC change on 
variation in fish assemblages. 
3.3 STUDY AREA 
 I studied stream reaches from tributaries of the middle Chattahoochee River, 
western Georgia, USA, occurring in the Southern Outer Piedmont ecoregion (Griffith et 
al., 2001).   Currently, conversion of pasture and forests to urbanized areas is occurring 
rapidly northeast of the city of Columbus, Muscogee County.  Therefore, I sampled 2
nd
- 
to 3
rd
-order streams within 18 watersheds (4?25 km
2
) along this urbanization gradient 
from the geologic fall line in the city of Columbus to an area 80 km northeast (Muscogee, 
Harris, Troup, and Meriwether counties, Figure 1).  All study reaches were typical of the 
lower Piedmont, consisting of sandy-bottom channels with run-pool morphologies 
(Mulholland and Lenat, 1992).  Watersheds ranged in land cover from intense 
 
 51
urbanization and active suburban development to pasture to heavily forested areas.  This 
relatively large range in landscape character allowed comparison across geomorphically 
similar streams that differed primarily in watershed-level LU/LC and associated variation 
in streamwater physicochemical conditions.   
3.4 METHODS 
3.4.1 Land cover analysis 
 I determined watershed boundaries and size from USGS 30-m resolution digital 
elevation models and ArcView 3.2a software (Environmental Research Systems Institute, 
Inc., Redwoods, Calfornia).   True color (3-band) aerial photographs of study watersheds 
were taken in March 2003 during leaf-off to determine LU/LC.  Impervious surface (IS) 
and water bodies were manually digitized and the remaining land cover was classified 
using a hybrid unsupervised/supervised classification scheme, a modification of the 
Anderson Classification Scheme (Myeong et al., 2001; Lockaby et al., 2005).  
Watersheds were ground-truthed to verify LU/LC classes, and the overall classification 
accuracy (all land covers combined) was 91% (see Lockaby et al., 2005 for method).  I 
used % of each watershed as IS, pasture, and forest (deciduous + evergreen) for analyses.   
I also assigned each watershed to 1 of 4 broad LU/LC categories (Urban, Developing, 
Pasture, and Forest) to aid in describing any perceived differences among watersheds 
(Table 1).  These categories were based on the dominant land cover in the watershed (% 
IS, pasture, and/or forest) from land cover analysis except for the Developing, which 
were predominantly forested but contained active or recent residential development. 
 
 
 
 52
3.4.2 Hydrology measures 
 I quantified continuous stream discharge (Q) from July 2003 to July 2004 using a 
Mini-Troll? pressure transducer data logger (In-Situ Inc., Ft. Collins, Colorado) housed 
in PVC pipe and installed near the outflow point of each watershed.  I set data loggers to 
record a stage reading at 15-min intervals (0.01-m depth resolution) and then, by 
correlating these stage readings with discharge (either directly measured or calculated at 
various stages (Gordon et al., 2004)), I developed stage?Q rating curves for each 
watershed to estimate continuous Q (Schoonover et al., 2006).   I characterized 5 separate 
elements of Q from each hydrograph (Table 2): 1) Baseflow (groundwater contribution), 
2) Predictability/Flashiness (rate/amount of change in Q), 3) Duration (length of time 
associated with a particular high or low Q event), 4) Magnitude (amount of Q for a given 
interval), and 5) Frequency (number of occurrences of a given magnitude of Q)  (Poff 
and Ward, 1989; Richter et al.,1996; Poff et al., 1997; McMahon et al., 2003).   
 I predicted baseflow for each watershed using a 5-d smoothed minima technique 
(Gustard, 1992; Schoonover et al., 2006), calculated by dividing the Q data into non-
overlapping 5-d blocks and determining the minimum flow in each block.  The minimum 
value in a given block was compared to the minimum values of the previous and 
subsequent 5-d blocks (Gustard, 1992). If the minimum value was less than these 
adjacent values, it was considered an estimate of baseflow for that period.   Then, I used 
linear interpolation between each baseflow estimate to predict baseflow for each 
observed flow measure for the entire data set.  I then developed a baseflow index measure 
of overland flow (BI) as: 
   ? predicted baseflow / ? observed flow. 
 
 53
 BI values can range from 1, when 100% of observed Q was from baseflow (low overland 
contribution) to 0 when 0% of observed Q was from baseflow (high overland 
contribution, see Gustard et al., 1992; Schoonover et al., 2006).   Ultimately, I calculated 
29 hydrological variables considered important in describing stream biotic parameters 
(Richter et al., 1996; Poff et al., 1997; Roy et al., 2005a) for each watershed (Table 2).   
3.4.3 Physicochemistry measures 
 I measured several physicochemical variables over the hydrological period of 
record (Table 3).  Stream temperature was measured continuously with HOBO? Temp 
data loggers placed near the pressure transducers, DO and pH were measured seasonally 
in areas where fish were sampled while total dissolved solids (TDS) and total suspended 
solids (TSS) were determined monthly as detailed in Schoonover and Lockaby (2006). 
3.4.4 Fish and habitat sampling 
 I sampled stream fish assemblages in June 2004 from 3 run and 3 pool habitats 
per stream along a representative 100-m reach.  I sampled fish in each habitat to depletion 
with block nets, a backpack electroshocker (Smith-Root LR-24) and seines.  I field-
identified and measured total length (TL) of all fish captured and returned them near the 
point of collection, except for voucher specimens of each species, which were deposited 
in the Auburn University Museum Fish Collection.   
I assigned species to feeding and breeding guilds (Muncy et al., 1979; Berkeman 
and Rabeni, 1987), as they have been shown by others to be reliable indicators of biotic 
integrity in Georgia piedmont streams (Schleiger, 2000; Helms et al., 2005). For feeding 
guilds, fish were classified as piscivores, insectivores, herbivores, omnivores, or filter 
feeders.  For breeding guilds, I initially classified fish as complex or simple breeders, 
 
 54
based on the degree to which species prepare spawning sites, defend nests, and show pre-
spawning social behavior (Pflieger, 1975; Trautman, 1981).  Complex breeders were then 
further classified into those species that show parental care (P/C) and those that do not 
(No P/C) while simple breeders divided into spawners requiring clean, gravel substrate 
(lithophilic spawners, = Lithophils) and those capable of spawning on sand, silt, or 
vegetation (generalist spawners, = Simple Spawners).  I used these classifications to 
assess whether there were functional changes in assemblages associated with different 
stream conditions. 
 I calculated species richness and species diversity (Shannon?s H?) for each stream.  
Richness and H? are commonly used metrics for comparing fish assemblages; however, 
human disturbance may cause only nominal changes in H? or species richness but major 
changes in composition (Scott and Helfman, 2001; Walters et al., 2005).  Therefore, I 
also calculated a stream-specific Index of Biotic Integrity (IBI) modified for Georgia 
piedmont streams (Schleiger, 2000) and relative abundance for use in a non-metric 
multidimensional scaling (NMDS) ordination to describe overall fish variation among 
watersheds. 
To assess available habitat quality/quantity, I used a comprehensive multimetric 
habitat assessment from the Georgia Environmental Protection Division designed for use 
in fish biomonitoring (GA DNR, 2005).  This Habitat Index included visual estimates of 
available cover (number and frequency of habitats), substrate characterization (type and 
condition), pool morphology (shape and frequency), channel alteration (frequency of 
riprap, dredging, etc), channel sinuosity (run-to-bend ratio), sediment deposition (particle, 
point bar, and island size), flow status (degree to which channel is filled with water), 
 
 55
bank condition (erosion potential and vegetation cover) and riparian condition (vegetation 
cover/quality) (GA DNR, 2005).  This assessment involves taking the average of 3 
individuals? summed scores (1-10 or 1-20, depending upon parameter) of the different 
habitat parameters to obtain an overall habitat quality value for the representative reach, 
with high average score indicating high habitat quality.  I used the same 3 observers at all 
sites.  
In addition to the GA DNR habitat assessment, I assessed stream habitat by 
estimating habitat volume (mean depth x width x length), benthic organic matter 
abundance (BOM), substrate particle sizes, and benthic shear stress (T
G
 ) in each habitat 
unit at the time of fish sampling.  I estimated BOM and substrate size by sampling 
transitional areas between the runs and pools where fish were sampled to standardize our 
efforts and avoid error associated with scour in the runs and deposition in the pools.  I 
sampled BOM by determining the ash-free dry mass (AFDM) of 9 replicate 2.5 x 10 cm 
benthic cores.  For substrate particle size, I collected 3 benthic samples per stream (near 
where BOM was sampled) using a 76.2-mm diameter PVC substrate core to a depth of 10 
cm.  I dried samples and separated particles into 5 size classes: gravel-cobble (>2mm), 
very coarse sand (1 ? 2 mm), coarse to medium sand (0.25 ? 1 mm), fine sand (0.1 ? 0.25 
mm), very fine sand (0.05 ? 0.1mm), and silt/clay (< 0.05mm) to determine median 
substrate size by weight (USDA 1951).   I estimated T
G
   close to where pressure 
transducers were located using  
    T
G
 = pgRS 
where p is the density of water, g is gravitational acceleration, R is hydraulic radius, and 
S is energy slope (Gore, 1996). 
 
 56
3.4.5 Data analyses 
 I used simple Pearson correlations to examine relationships between 
environmental variables (hydrology, habitat, physicochemical) and LU/LC.  All 
continuous variables were log-transformed and percent variables were arcsine-squareroot 
transformed as needed to meet assumptions of normality (Zar, 1998).    
 Second, I used non-metric multidimensional scaling (NMDS) to describe overall 
variation in fish assemblages among sites. NMDS is an ordination technique that handles 
data with many zeroes and nonnormal data (both common in species data) better than 
other ordination techniques such as Principal Components Analysis (PCA, McCune and 
Grace, 2002) but, unlike PCA, the order of the axes in the resultant ordination does not 
necessarily imply the order of greatest variation explained.  I transformed proportional 
relative abundance data using arcsin-square root and excluded rare species (those in 
<10% of sites) to reduce the influence of rare taxa on ordinations.  This step resulted in 
an 18 x 23 site?species matrix on which I based ordinations using a Sorenson distance 
measure (McCune and Grace, 2002).  I correlated all environmental variables and fish 
assemblage variables to the resulting NMDS ordination to assess changes over the 
landscape.  
 Last, I used stepwise multiple regression analyses (p = 0.05 to enter and leave the 
model) to determine which environmental variables, directly or indirectly related to 
LU/LC, had the most explanatory power in regards to fish assemblages.  To conform to 
the conceptual model in Figure 1, any hydrologic variable not related to some aspect of 
LU/LC was dropped from multiple regression as were any physicochemical or habitat 
variable not related to either LU/LC or the selected hydrologic variables.  In order to 
 
 57
avoid multicollinearity among predictors, all parameters in the final models had a 
variance inflation factor (VIF) < 10 (Myers 1990).   
3.5 RESULTS 
3.5.1 Stream hydrology 
 In general, the hydrographs of the urban and developing watersheds were flashier 
and less stable than hydrographs in other watersheds (Table 2).  There was a strong 
relationship between urbanization and the frequency and predictability of hydrological 
events as evidenced by several parameters being positively correlated with IS, notably the 
measures 5XMed, 7XMed, 9XMed and N>99
th
 and the variables Inc1h100 and Inc3h100 
(Table 2).   Median Q also was positively correlated and Inc1h5000 was negatively 
correlated with proportion of watershed as pasture.  There was no significant relationship 
between any hydrological parameter and proportion of watershed as forest (Table 2).   I 
therefore used these 8 hydrologic variables in subsequent multiple regressions since they 
showed significant relationships with some aspect of watershed LU/LC.  
3.5.2 Physicochemistry and habitat 
 Physicochemical and habitat parameters were variable across the watersheds with 
all parameters associated with either watershed LU/LC and/or hydrology (Tables 3 & 4).  
In general, higher stream temperatures and lower DO levels were associated with 
increased watershed IS and spate flows (5xMed, 7xMed, and 9xMed, Tables 3 & 4).   
Mean and minimum DO levels were highly correlated with each other (r = 0.884, p < 
0.001), and since mean DO was a continuous measure and minimum DO was a single 
measurement, I used mean DO for statistical analyses.   Further, TDS was positively 
associated with IS cover and spate frequency measures, TSS was positively correlated 
 
 58
with % pasture and the flashiness variable Inc1h5000, and pH was positively correlated 
with Inc1h100 and Inc3h100 (Table 3 & 4).   Of the habitat variables considered, habitat 
volume and median substrate size were positively correlated with IS while OM and the 
Habitat Index were positively correlated with forest cover (Table 3).  Substrate size was 
also positively correlated with spate frequency variables, tractive force increased with 
median discharge, and the Habitat Index increased with Inc1h5000.  All physicochemical 
variables were correlated with either a hydrologic or LU/LC variable, therefore they were 
all included in multiple regressions analysis. 
3.5.3 Fish assemblage structure 
 I collected 27 fish species (1152 individuals) in 7 families during the study, with 
Cyprinidae (minnows) and Centrarchidae (sunfishes) being the most common and 
abundant families (Table 5).  Of the breeding and feeding guilds, proportion of lithophilic 
spawners declined with increasing IS (r = -0.66, p < 0.01) and increased with total forest 
cover (r = 0.51, p < 0.05) while proportion of insectivores increased with IS (r = 0.57, p < 
0.05) and decreased with forest cover (r = -0.51, p < 0.05).   Species richness ranged from 
2 to 13 species, with highest measures in pasture and forested watersheds, and was 
negatively correlated with IS (r = -0.66, p < 0.01) (Table 6).  H? ranged from 1.07 to 2.95, 
with the highest values in forested and pasture dominated watersheds, and was negatively 
correlated with IS (r = ?0.85, p < 0.01) and positively correlated with forest cover (r = 
0.52, p = 0.03), while proportion of sunfish increased with urban cover (Table 6).  IBI 
ranged from 22 to 42, with highest values in pasture and forested watersheds; however, 
IBI was not significantly correlated with any land cover.   
 
 59
 NMDS axes 1 and 3 (NMDS
1
 and NMDS
3
) were the two axes describing the most 
variation in fish assemblages among streams (48.7 and 22.8%, respectively; stress = 13.8, 
instability = 0.00001, iterations = 91; Figure 3).  Streams in Urban and Developing 
watersheds generally grouped to the left of the ordination whereas streams from Forest 
and Pasture watersheds grouped mostly to the right (Figure 3).   Number of fish collected, 
H?, proportion of lithophils, and IBI all were positively correlated with NMDS
1
 while the 
proportion of sunfish, insectivores, and fish showing no parental care, were all negatively 
correlated with NMDS
1
 (Figure 3).  The proportion of the assemblage showing parental 
care was significantly correlated with NMDS
3
.  There were several environmental 
parameters associated with these shifts in fish assemblage structure across the landscape, 
including stream DO levels, water temperature, TDS, substrate size, frequency of spate 
flows, and habitat volume (Figure 3).    
3.5.4 Influence of environmental variables  
 Multiple regression analyses revealed that hydrologic and physicochemical 
variables were good predictors of fish assemblages (Table 6).  Models describing 
richness, diversity, and the IBI all contained measures of spate frequency (5xMed, 
7xMed, and N>99).  Stream water temperature was the best predictor of number of fish 
collected while temperature and 5xMed best described taxa richness (Table 6).  TDS was 
also prominent in models as a strong predictor of diversity and % sunfish (Table 6).  
Interestingly, no habitat variables were included as parameters in any best models.     
3.6 DISCUSSION 
 Environmental controls of stream fish assemblages are varied, often interactive, 
and frequently associated with landscape disturbance (Roth et al., 1996, Matthews, 1998; 
 
 60
Lammert and Allan, 1999).  Our results provide additional empirical evidence that the 
indirect effects of land use on the integrity of fish assemblages can occur through the 
alteration of instream environmental conditions, particularly alterations in hydrological 
and physicochemical conditions. 
3.6.1 Hydrology and fish assemblages 
Stream flow is often considered the master variable limiting aquatic biota by its 
effects on instream physicochemistry, geomorphology, and habitat diversity (Poff and 
Allan, 1995; Poff et al., 1997).  In our study streams, hydrology appeared to have a 
strong effect on differences in fish assemblages among watersheds.  Richness, H?, and 
IBI all were lower in streams experiencing numerous high-magnitude flows, whereas 
more taxonomically rich and diverse assemblages were associated with streams 
experiencing fewer high-magnitude flows.   Fish patterns were also strongly associated 
with the number of events that exceeded 5 , 7, and 9 times median flow, moderate events 
that, on average, were all less than 33% of bankfull Q in these watersheds.  Small, 
frequent spates have been suggested to be more important than infrequent larger events in 
causing ecological impacts (Walsh et al., 2005).   Moreover, baseflow index (BI) was a 
strong correlate of NMDS axis 1, suggesting that increases in overland flow events and 
associated spate frequency, and not necessarily alterations in the duration, predictability, 
or magnitude of flows, are strong hydrological drivers of fish assemblages in these 
watersheds.  These findings support other studies on fish reporting increased proportions 
of habitat generalist species with increasing frequency of hydrological disturbance, and 
taken together, underscore the far-reaching effects of hydrology on stream ecological 
 
 61
integrity (Resh et al., 1988; Poff et al., 1997; Freeman et al., 2001; Roy et al., 2005a).   If 
true for other piedmont watersheds, then such frequency-based hydrological variables 
should be taken into consideration by resource managers to identify flow-related impacts 
to fish in developed/developing watersheds. 
As shown by others, fish assemblages overall show lower diversity and biotic 
integrity in developing and highly urbanized watersheds compared with less-developed 
watersheds (Koel and Petarka, 2003; Walters et al., 2003; Helms et al., 2005; Roy et al., 
2005a).  As determined from the NMDS ordination, this pattern appeared to correspond 
to the combined changes in stream hydrological and physicochemical conditions 
associated with increasing impervious surface and connectedness in the Urban and 
Developing watersheds.  Interestingly, BI was not significantly correlated to any LU/LC 
variable, likely offset by the relatively high % Forest in the Developing streams, 
demonstrating the pervasive influence of even a small percentage of watershed 
urbanization.  In a related study of these same sites (Schoonover et al., 2006) measures of 
BI suggest that overland flow (vs. baseflow inputs) contributed up to 90% of Q reaching 
Urban streams and 65 to 90% of Q reaching Developing streams.  High overland flow 
and associated spates are not only likely to contribute significant physical impacts on 
fishes (i.e., through downstream displacement of individuals, habitat alterations, etc.), but 
can also act to transport pollutants, elevate water temperatures and bacteria and nutrient 
concentrations from terrestrial sources and also resuspend materials in the stream bed 
(Casey and Farr, 1982; Paul and Meyer, 2001).   
In our study streams TDS and temperature were generally elevated and DO 
decreased in urbanized systems and were associated with decreased fish diversity.  Many 
 
 62
other studies have also observed elevated TDS (or specific conductivity) with increased 
urban area / impervious surface (Dow and Zampella 2004) as well as with decreased 
biotic integrity (Walsh et al., 2001; Roy et al. 2003).  However, TDS concentrations were 
not necessarily at biologically significant levels, as most aquatic systems with biota can 
withstand TDS levels up to 1000 mgL
-1
 (Boyd 2000). Therefore the strong association of 
TDS with urbanized/urbanizing watersheds suggests that it is a likely indicator of 
increased non-point pollution associated with efficient runoff, thus an ?anthropogenic 
marker? in these streams.     
 Water temperature and DO are major regulators of fish distribution, growth, 
migration, and survival (Fry, 1947; Regeir et al., 1990; Smale and Rabeni, 1995; Krause 
et al., 2004), and levels of each of these parameters are important predictors of fish 
assemblages in streams of western Georgia.  Warmwater sunfish species were ubiquitous 
in our study streams, but they were far more abundant in streams with higher water 
temperatures and lower DO than in streams without these stressors.  This pattern suggests 
that elevated stream temperature and low DO, particularly in urban and developing 
streams with thermally enhanced overland flow and/or reduced riparian cover (Van 
Buren, 2000; Roy et al., 2005b), may negatively affect presence or abundance of fishes in 
general.  It seems unlikely though that elevated temperature directly affected fishes as the 
maximum water temperature I observed (25.6
O
C) was well below most physiological 
thresholds and habitat requirements for most native fish of this region (Aho et al., 1986; 
Brown, 1974; Krause et al., 2004).  However, besides causing mass fish kills (Gafny et 
al., 2000), low DO may produce important sublethal effects, leading to habitat and 
behavioral shifts in populations and, ultimately, altered local assemblage structure 
 
 63
(Kramer,1987; Matthews, 1987;1998). Specifically, critical DO levels for similar fish 
assemblages in warm-water streams range 0.49 to 1.49 mgL
-1 
and strong effects of 
hypoxia on fish habitat use and species composition have been implicated when water 
DO minima fall below 4 to 5 mgL
-1
 (Smale and Rabeni, 1995).  DO levels in our streams 
occasionally reached these lethal levels and frequently reached those reported levels that 
could influence species distribution over the period of record.     
Altered land use can induce physical changes in stream channels, influencing the 
dynamics and spatial arrangement of channel features and instream habitat (Allan, 2004).  
Our analyses indicated that some instream fish habitat conditions were related to 
watershed LU/LC.  However, habitat variables in general were weak predictors of fish 
assemblage structure.  There was a relationship between assemblage structure and 
substrate size as evidenced by NMDS axis 1, with more tolerant assemblages being 
associated with streams with large substrate size.  This habitat feature is likely the result 
of bed coarsening and flushing, and is common in urban and other hydrologically altered 
watersheds (Finkenbine et al., 2000; Walsh et al., 2005).  However, considering the size 
range (0.5 ? 1.8 cm), and that lithophilic spawners were negatively associated with 
substrate size, substrate composition was unlikely to be an important driver of 
assemblages in these streams.  Taken together, our results suggest that hydrological 
regimes may influence instream habitat conditions in these watersheds, but local habitat 
per se is not a strong driver of the observed differences in fish assemblages (cf. Poff et 
al., 1997; Sutherland et al., 2002).  It should be noted, however, that the weak link 
between fish assemblages and habitat variables, compared with hydrological and 
physicochemical measures, may reflect a disparity of 1) scale between our measures of 
 
 64
habitat (100-m study reaches) and potential longitudinal movement of fish (>100 m), 2) 
sampling, given that I measured hydrology and physicochemistry multiple times (and, for 
some measures, continuously) over the study, and habitat variables were measured just 
once, and 3) the biotic composition of fishes in general in lower Piedmont watersheds of 
the Chattahoochee drainage, which have a natural predominance of widespread species 
(Hilliard, 1984; Swift et al., 1986).  
3.6.2 Conclusions 
Our results suggest that LU/LC induced changes in hydrology and streamwater 
physicochemistry, particularly in urban/urbanizing watersheds, influence stream fish 
assemblages more so than alterations in physical habitat; however, there is undoubtedly 
high complexity in the functional interrelationships of environmental variables in these 
streams.  Physicochemical conditions are closely linked to hydrology and land use, and 
teasing the relative importance of each is a daunting task.  The use of multivariate 
analyses effectively allowed us to identify important correlates of fish assemblage 
structure that, as a result of the nature of the suburban landscape, were not strongly 
associated with measured LU/LC values (e.g., BI).  However, many physicochemical 
conditions are correlated, so it may only be necessary to identify a single 
physicochemical or hydrological group of variables for certain management and/or 
restoration goals (e.g. spate frequency, temperature/DO).  Further, in these lower 
Piedmont systems with a history of land use degradation, an assemblage-based response 
(such as the NMDS ordination) may be better suited to evaluating the impacts of human 
induced change than traditional metrics (e.g. H?) because of the high abundance of 
 
 65
widely distributed species and relatively few endemics, often hindering useful 
comparisons to degraded systems.   
Human population expansion and the inevitable landscape alteration caused by 
such growth have produced dramatic impacts on stream ecosystems.  By attempting to 
identify the specific hydrological and physicochemical driver(s) of biotic composition 
resulting from these perturbations, I can better address management and/or restoration 
needs designed to protect or minimize changes in stream biotic integrity. 
 
 66
3.7 REFERENCES 
Aho, J.M., Anderson, C.S., and Terrell, J.W.  1986.  Habitat Suitability Index Models and 
Instream Flow Suitability Curves: Redbreast Sunfish.  U.S. Fish and Wildlife 
Service Biological Report 82 (10.119), 23 pp. 
Allan, J. D.  2004.  Landscapes and Riverscapes:  the Influence of Land Use on Stream 
Ecosystems.  Annual Review of Ecology and Systematics 35: 257-284. 
Berkman, H. E. and Rabeni, C. F. 1987.  Effect of Siltation on Stream Fish Communities.  
Experimental Biology of Fishes 18: 285-294. 
Bledsoe, B.P. and Watson, C.C.  2001.  Effects of Urbanization on Channel Instability.  
Journal of the American Water Resources Association 37: 255-270. 
Booth, D.B. and Jackson, C.R.  1994.  Urbanization of Aquatic Systems:  Degradation 
Thresholds, Stormwater Detection, and the Limits of Mitigation.  Journal of the 
American Water Resources Association 33: 1077-1090. 
Boyd, C. E. 2000. Water Quality: An Introduction. Kluwer Academic Publishers, 
Norwell, Massachusetts, USA 
Brown, H.W.  1974.  Handbook of the Effects of Temperature on Some North American 
Fishes.  American Electric Power Corporation, Canton, Ohio. 
Casey, H., and I.S. Farr.  1982.  The Influence of Within Stream Disturbance on 
Dissolved Nutrient Levels During Spates.  Hydrobiologia 92: 447 ? 462.  
Clements, W.H., Carlisle, D.M., Lazorchak, J.M., and Johnson, P.C.  2000.  Heavy 
Metals Structure Benthic Communities in Colorado Mountain Streams.  
Ecological Applications 10: 626-638. 
 
 67
Fausch, K.D., Torgersen, C.E., Baxter, C.V., and Li, H.W.  2002.  Landscapes to 
Riverscapes: Bridging the Gap Between Research and Conservation of Stream 
Fishes.  BioScience 52: 483-498. 
Ferguson, B.K. and Suckling, P.W.  1990.  Changing Rainfall-runoff Relationships in the 
Urbanizing Peachtree Creek watershed, Atlanta, Georgia.  Water Resource 
Bulletin  26:  313-322. 
Finkenbine, J.K., Atwater, J.W., and Mavinic, D.S.  2000.  Stream Health After 
Urbanization Journal of the American Water Resources Association 36:  1149-
1160. 
Fisher, R.F. and Binkley, D.  2000.  Ecology and Management of Forest Soils.  Third ed.  
John Wiley & Sons, NY. 
Freeman, M.C., Bowen, Z.H., Bovee, K.D. and Irwin, E.R.  2001.  Flow and Habitat 
Effects of Juvenile Fish Abundance in Natural and Altered Flow Regimes.  
Ecological Applications 11: 179-190. 
Frissell, C.A., Liss, W.J., Warren, C.E. and Hurley, M.D.  1986.  A Hierarchical 
Framework for Stream Habitat Classification: Viewing Streams in a Watershed 
Context.  Environmental Management 12: 199-214. 
Fry, F.E.J. 1947.  Effects of the Environment on Animal Activity.  University of Toronto 
Studies, Biological Series 55.  (Publication of the Ontario Fisheries Research 
Laboratory 68: 1-62). 
Gafney, S., M. Goren, and A. Gasith.  2000.  Habitat Condition and Fish Assemblage 
Structure in a Coastal Mediterranean Stream (Yargon, Israel) Receiving Domestic 
Influence.  Hydrobiologia 422/423: 319 ? 330.   
 
 68
Galli, F.J.  1991.  Thermal Impacts Associated with Urbanization and Stormwater 
Management. Best Management Practices.  Washington, DC: Washington 
Council of Governments. 
Georgia Department of Natural Resources, Environmental Protection Division. 2005.  
Standard Operating Procedures for Conducting Biomonitoring on Fish 
Communities in Wadeable Streams in Georgia.  Water Protection Branch, 
Atlanta, Georgia, USA. 
Gholz, H.L., K.C. Ewel, R.O. Teskey.  1990.  Water and Forest Productivity.  Forest 
Ecology and Management 30: 1 ? 18. 
Gordon, N.D., T.A. McMahon, B.L. Finlayson, C.J. Gippel, and R.J. Nathan.  2004.  
Stream Hydrology: An Introduction for Ecologists ? 2
nd
 edition.  John Wiley & 
Sons Ltd.  Hoboken, New Jersey, USA.  
Groffman, P.M., Bain, D.J., Band, L.E., Belt, K.T., Brush, G.S., Grove, J.M., Pouyat, 
R.V., Yesilonis, I.C. and Zipperer, W.C.  2003.  Down by the Riverside: Urban 
Riparian Ecology.  Frontiers in Ecology and the Environment 1: 315-321. 
Groffman, P.M., Baron, J.S., Blett, T., Gold, A.J., Goodman, I., Gunderson, L.H., 
Levinson, B.M., Palmer, M.A., Paerl, H.W., Peterson, G.D., Poff, N.L., Rejeski, 
D.W., Reynolds, J.F., Turner, M.G., Weathers, K.C., and WIens, J.  2006.  
Ecological Thresholds:  The Key to Successful Environmental Management or an 
Important Concept with No Practical Application?  Ecosystems 9: 1-13. 
Griffith, G.E., J.M. Omernik, J.A. Comstock, S. Lawrence, and T. Foster.  2001.  Level 
III and IV Ecoregions of Georgia (CWCS color poster with map and descriptive 
 
 69
text).  Wildlife Resources Division, Georgia Department of Natural Resources.  
Social Circle, Georgia, USA. 
Gustard, A., A. Bullock, and J.M. Dixon.  1992.  Low Flow Estimation in the United 
Kingdom, Report 108. Institute of Hydrology, Wallingford, England.  pp 19-25.   
Hammer, T.R.  1972.  Stream Channel Enlargement Due to Urbanization.  Water 
Resources Research 8: 139-167. 
Harding, J. S., Benfield, E. F., Bolstad, P. V., Helfman, G. S., and Jones, E. B. D. 1998. 
Stream Biodiversity: The Ghost of Landuse Past. Proceedings of the National 
Academy of Sciences  95:14843-14847. 
Hawkins, C.P., Hogue, J.N. Decker, L.M. and Feminella, J.W.  1997.  Channel 
Morphology, Water Temperature, and Assemblage Structure of Stream Insects.  
Journal of the North American Benthological Society 16: 728-749. 
Helms, B.S., Feminella, J.W. and Pan, S.  2005.  Detection of Biotic Responses to 
Urbanization Using Fish Assemblages from Small Streams of Western GA, USA. 
Urban Ecosystems 8: 39-57. 
Hilliard, S. B. 1984. Atlas of Antebellum Southern Agriculture. Louisiana State 
University Press, Baton Rouge, Louisiana. 
Holling, C.S.  1973.  Resilience and Stability of Ecological Systems.  Annual Review of 
Ecology and Systematics 4: 1-23. 
Koel, T.M. and Peterka, J.J.  2003.  Stream Fish Communities and Environmental 
Correlates in the Red River of the North, Minnesota and North Dakota.  
Environmental Biology of Fishes 67: 137-155. 
 
 70
Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zaugg, S.D., Barber, L.B., 
and Buxton, H.T.  2002.  Pharmaceuticals, Hormones, and Other Organic 
Wastewater Contaminants in US Streams, 1999-2000: A National 
Reconnaissance.  Environmental Science and Technology 36: 1202-1211. 
Kramer, D.L.  1987.  Dissolved-oxygen and Fish Behavior.  Environmental Biology of 
Fishes 18: 81 ? 92. 
Krause, C.W., Lockard, B., Newcomb, T.J., Kibler D., Lohani V., and Orth, D.J.  2004.  
Predicting Influences of Urban Development on Thermal Habitat in a Warm 
Water Stream.  Journal of the American Water Resources Association 40: 1645-
1658. 
Lammert, M. and Allan, J.D.  1999.  Assessing Biotic Integrity of Streams: Effects of 
Scale in Measuring the Influence of Land Use/Cover and Habitat Structure on 
Fish and Macroinvertebrates.  Environmental Management 23: 257-270. 
Lenat, D.R., and Crawford, J.K.  1994.  Effects of Land-use on Water-quality and 
Aquatic Biota of Three North Carolina Piedmont Streams.  Hydrobiologia 294: 
185-199. 
Lockaby, B.G., Zhang, D., McDaniels, J. Tian, H., and Pan, S.  2005.  
 Interdisciplinary Research at the Urban-rural Interface: The WestGa 
 Project. Urban Ecosystems 8: 7-21. 
Loewenstein, N.J., and Loewenstein, E.F.  2005.  Non-native Plants in the Understory of 
Riparian Forests Across a Land Use Gradient in the Southeast.  Urban 
Ecosystems 8: 79-91. 
 
 71
Matthews, W. J. 1987. Physicochemical Tolerances and Selectivity of Stream Fishes as 
Related to Their Geographic Ranges and Local Distributions. In Community and 
Evolutionary Ecology of North American Stream Fishes. Edited by W.J. 
Matthews and D. H. Heins. University of Oklahoma Press, Norman, Oklahoma, 
USA, pp 111-120. 
 Matthews, W.J.  1998.  Patterns in Freshwater Fish Ecology.  Chapman and Hall.  New 
York. 
McCune, B. and Grace, J.B.  2002.  Analysis of Ecological Communities.  MjM Software 
Design, Glenedon Beach, Oregon, USA. 
McMahon, G., Bales, J.D., Coles, J.F.,  Giddings, E.M.P., and Zappia, H.  2003.  Use of 
Stage Data to Characterize Hydrologic Conditions in an Urbanizing Environment.  
Journal of the American Water Resources Association 39: 1529-1546. 
Mulholland, P.J. and Lenat, D.R.  1992.  Streams of the Southeastern Piedmont, Atlantic 
Drainage.  In Biodiversity of the Southeastern United States ? Aquatic 
Communities. Edited by C.T. Hackney, S.M. Adams, and W.A. Martin. New 
York: Wiley. pp 193-232.   
Myeong, S., Nowak, D.J., Hopkins, P.F., and Brock, R.H.  2001.  Urban Cover Mapping 
Using Digital, High-spatial Resolution Aerial Imagery.  Urban Ecosystems 5: 
243-256. 
Muncy, R.J., Atchinson, G.J., Bulkley, R.V. Menzel, B.W., Perry, L.G., and Summerfelt, 
R.C.  1979.  Effects of Suspended Solids and Sediment on Reproduction and 
Early Life of Warmwater Fishes: A Review.  EPA-600/3-79-0, Covallis 
Environmental Research Laboratory, Corvallis, 100 pp. 
 
 72
Paul, M.J. and Meyer, J.L..  2001.  Streams in the Urban Landscape.  Annual Review of 
Ecology and Systematics 32: 333-363. 
Peterson, J.T., and Kwak, T.J.  1999.  Modeling the Effects of Land Use and Climate 
Change on Riverine Smallmouth Bass.  Ecological Applications. 9: 1391-1404. 
Pflieger, W.L.  1975.  The Fishes of Missouri.  Missouri Department of Conservation and 
Western Publishing Company, Columbia, Missouri. 
Poff, N.L., and J.D. Allan.  1995.  Functional Organization of Stream Fish Assemblages 
in Relation to Hydrological Variability.  Ecology 76: 606 ? 627. 
Poff, N.L., and Ward, J. V.  1989.  Implications of Streamflow Variability and 
Predictability for Lotic Community Structure: A Regional Analysis of Streamflow 
Patterns.  Canadian Journal of Fisheries and Aquatic Sciences 46: 1805-1818. 
Poff, N.L., Allan, J.D., Bain, M.B. Karr, J.R. Prestegaard, K.L. Richter, B.D. Sparks, 
R.E., and Stromberg, J.C.. 1997.  The Natural Flow Regime: a Paradigm for River 
Conservation and Restoration.  BioScience 47: 769-784. 
Regier, H.A., Holmes J.A., and Pauly, D.  1990.  Influence of Temperature Changes on 
Aquatic Ecosystems: An Interpretation of Empirical Data.  Transactions of the 
American Fisheries Society 119:  374-389. 
Resh, V.H., Brown, A.V., Covich, A.P., Gurtz, M.E., Li, H.W., Minshall, G.W., Reice, 
S.R., Sheldon, A.L., Wallace, J.B., and Wissmar, R.C.  1988.  The Role of 
Disturbance in Stream Ecology.  Journal of the North American Benthological 
Society 7: 433-455. 
Richter, B.D., Baumgartner, J.V., Powell, J., Braun, D.P.  1996.  A Method for Assessing 
Hydrologic Alteration within Ecosystems.  Conservation Biology 10: 1163-1174. 
 
 73
Rose S. and Peters, N.E.  2001.  Effects of Urbanization on Streamflow in the Atlanta 
Area (Georgia, USA):  A Comparative Hydrological Approach.  Hydrological 
Processes 15:  1441-1457. 
Roth, N.E., Allan, J.D., and Erickson, D.L.  1996.  Landscape Influences on Stream 
Biotic Integrity Assessed at Multiple Spatial Scales.  Landscape Ecology  11: 
141-156. 
Roy, A.H., Freeman, M.C.,  Freeman B.J., Wenger, S.J., Ensign, W.E., and Meyer, J.L.  
2005a.  Investigating Hydrologic Alteration as a Mechanism of Fish Assemblage 
Shifts in Urbanizing Streams.  Journal of the North American Benthological 
Society  24: 656-678. 
Roy, A.H., Faust, C.L., Freeman, M.C., and Meyer, J.L..  2005b.  Reach Scale Effects of 
Riparian Forest Cover in Urban Stream Ecosystems.  Canadian Journal of 
Fisheries and Aquatic Sciences 62: 2312-2329. 
Schleiger, S. L.  2000.  Use of an Index of Biotic Integrity to Detect Effects of Land Uses 
on Stream Fish Communities in West-central Georgia.  Transactions of the 
American Fisheries Society 129: 1118-1133. 
Schoonover, J.E. and Lockaby, B.G.  2006.  Changes in Nutrient and Bacteriological 
Properties of Stream Water Across an Urban - Rural Gradient in Western 
Georgia.  Journal of Hydrology 331: 371 ? 382.. 
Schoonover, J.E., Lockaby, B.G., and Helms, B.S.  2006.  Impacts of Land Cover on 
Stream Hydrology in the West Georgia Piedmont.  Journal of Environmental 
Quality 35: 2123 ? 2131. 
 
 74
Scott, J.B., Steward, C.R., and Stober, Q.J.  1986.  Effects of Urban Development on Fish 
Population Dynamics in Kelsey Creek, Washington.  Transactions of the North 
American Fisheries Society 115: 555-567. 
Scott, M.C. and Helfman, G.S.  2001.  Native Invasions, Homogenization, and the 
Mismeasure of Integrity of Fish Assemblages.  Fisheries.  26: 6 ? 15. 
Smale, M.A. and Rabeni, C.F.  1995.  Influences of Hypoxia and Hyperthermia on Fish 
Species Composition in Headwater Streams.  Transactions of the American 
Fisheries Society 124: 711-725. 
Sutherland, A.B., Meyer, J.L., and Gardiner, E.P.  2002.  Effects of Land Cover on 
Sediment Regime and Fish Assemblage Structure in Four Southern Appalachian 
Streams.  Freshwater Biology 47: 1791:1805. 
Swift, C. C., Gilbert, C. R., Bortone, S. A., Burgess, G. H., and Yerger, R. W.. 1986. 
Zoogeography of the Freshwater Fishes of the Southeastern United States: 
Savannah River to Lake Pontchartrain. In Zoogeography of North American 
Fishes. Edited by C. H. Hocutt and E. O. Wiley. New York. pp. 213-255 
Trautman, M.B.  1981.  The Fishes of Ohio, 2
nd
 Edition.  Ohio State University Press, 
Columbus, Ohio. 
Van Buren, M.A., Watt, W.E.,  Marsalek, J., and Anderson, B.C.  2000.  Thermal 
Enhancement of Stormwater Runoff by Paved Surfaces.  Water Resources  34: 
1359-1371. 
Walsh, C. J., Roy, A.H., Feminella, J.W., Cottingham, P.D., Groffman, P.M., and 
Morgan, R.P. 2005. The Urban Stream Syndrome: Current Knowledge and the 
 
 75
Search for a Cure. Journal of the North American Benthological Society 24: 706-
723.              
Walters, D.M., Leigh, D.S., and Bearden, A.B. 2003. Urbanization, Sedimentation, and 
the Homogenization of Fish Assemblages in the Etowah River Basin, USA.  
Hydrobiologia  494: 5?10 
Walters, D.M., Freeman, M.C., Leigh, D.S., Freeman, B.J., and Pringle, C.M.  2005.  
Urbanization Effects on Fishes and Habitat Quality in a Southern Piedmont River 
Basin.  In  Effects of Urbanization on Stream Ecosystems. Edited by L.R. Brown, 
R.H. Gray, R.H. Hughes, and M.R. Meador.    American Fisheries Society, 
Symposium 47, Bethesda, Maryland.  pp. 69-86 
Wang, L., Lyons, J.,  Kanehl, P., Bannerman, R., and Emmons, E.  2000.  Watershed 
Urbanization and Changes in Fish Communities in Southeastern Wisconsin 
Streams.  Journal of the American Water Resources Association 36:  1173-1189.   
Wang, L. and Lyons, J.  2003.  Fish and Benthic Macroinvertebrate Assemblages as 
Indicators of Stream Degradation in Urbanizing watersheds. In: Biological 
Response Signatures: Indicator Patterns Using Aquatic Communities.  Edited by 
T. Simon.  CRC Press, Boca Raton, USA, pp. 227-249. 
Weaver, L. A. and Garman, G. C.  1994.  Urbanization of a Watershed and Historical 
Changes in a Stream Fish Assemblage.  Transactions of the American Fisheries 
Society 123: 162-172. 
Whitehead, P.G. and Robinson, M.  1993.  Experimental Basin Studies ? An International 
and Historical Perspective of Forest Impacts.  Journal of Hydrology  145: 217 ? 
230. 
 
 76
Wolman, M.G.  1967.  A Cycle of Sedimentation and Erosion in Urban River Channels.  
Geografiska Annaler 49A: 385-395. 
Zar, J.  1999  Biostatistical Analysis.  4
th
 edition.  Prentice-Hall, Inc. Upper Saddle River, 
New Jersey, USA 
  T
a
bl
e 1.
  L
a
n
d
 co
v
e
r
 an
d phy
s
i
ca
l
 c
h
ar
act
er
s
 
o
f
 s
t
ud
y
 wat
e
r
s
h
e
d
s
.
  
IS
 = 
%
 i
m
p
e
r
v
i
o
u
s
 s
u
r
f
ace co
v
e
r
,
 
Pa
st
u
r
e
 
=
 %
 p
a
s
t
ur
e
 co
ve
r
,
 a
n
d
 
Mi
x
e
d
 F
o
r
e
s
t
 =  %
 m
a
naged a
n
d unm
a
n
aged 
f
o
r
e
s
t
 c
o
v
e
r
,
 an
d 
LU/
L
C
 =
 d
o
mina
n
t
 
l
a
nd
 u
s
e
/
la
nd
 c
o
ve
r
 
in w
a
t
e
r
s
he
d
,
 a
n
d
 
Spate
 
f
 
= 
s
p
at
e f
r
eque
n
c
y
 cat
ego
r
y
 as
s
i
g
n
ed eac
h
 wat
e
r
s
hed (
s
ee Res
u
l
t
s
)
.
   
 
 
    
ID Si
t
e
 
St
r
e
a
m
 
W
a
t
e
r
she
d
 si
z
e
 
(k
m
2
)
 I
S
 
P
a
s
t
u
re 
F
o
res
t
 
L
U
/L
C
 
1 S
B
1 
S
c
h
l
ey
 C
r
eek
 
2
0
.
1
 
2 20 
73 
D
e
ve
l
o
pi
n
g
 
2 S
B
2 
S
t
an
d
i
n
g
 B
o
y
 C
r
eek
 Tr
i
b
.
 
6
.
3
 
3 20 
73 
D
e
ve
l
o
pi
n
g
 
3 S
B
4 
S
t
an
d
i
n
g
 B
o
y
 C
r
eek
 
2
6
.
6
 
3 28 
64 
D
e
ve
l
o
pi
n
g
 
4 H
C
 
H
o
us
e
 Cr
e
e
k
 T
r
i
b
.
 
6.
6 
1 20 
75 
Fo
r
e
st
 
5 M
U
2 
M
u
l
b
er
r
y
 C
r
eek
 Tr
i
b
.
 
6
.
1
 
1 9 
82 
Fo
r
e
st
 
6 S
C
 
S
a
n
d
 C
r
e
e
k
 
9.
0 
1 21 
74 
Fo
r
e
st
 
7 B
C
 
B
e
ech
 C
r
eek
 
6
.
5
 
2 13 
81 
Fo
r
e
st
 
8 B
L
N
 
B
l
a
n
t
o
n C
r
e
e
k
 
3.
6 
1 19 
76 
Fo
r
e
st
 
9 M
K
 
F
l
at
 C
r
e
e
k
 T
r
ib
. 
6
.
6
 
2 20 
74 
Fo
r
e
st
 
10 M
O
 
C
l
in
e'
s
 B
r
an
c
h
 
9
.
0
 
2 13 
81 
Fo
r
e
st
 
11 M
U
3 
T
u
rn
t
i
m
e
 B
r
a
n
c
h
 
10.
4 
2 15 
78 
Fo
r
e
st
 
12 F
S
2
 
Wi
l
d
ca
t C
r
eek
 Tr
i
b
.
 
1
4
.
5
 
3 36 
59 
Pa
st
u
r
e
 
13 F
S
3
 
Wi
l
d
ca
t C
r
eek
 Tr
i
b
.
 
3
.
0
 
3 34 
62 
Pa
st
u
r
e
 
14 H
C
2 
H
o
us
e
 Cr
e
e
k
 
14.
1 
2 44 
52 
Pa
st
u
r
e
 
15 M
U
1 
O
s
s
a
ha
t
c
h
i
e
 C
r
e
e
k
 
T
r
i
b
.
 
12.
0 
4 37 
53 
Pa
st
u
r
e
 
16 BU
1 
L
i
n
d
s
e
y
 Cr
e
e
k
 
25.
5 
40 23 
34 
Ur
b
a
n
 
17 BU
2 
C
o
o
p
er
 C
r
eek
 
2
4
.
7
 
25 25 
46 
Ur
b
a
n
 
18 RB
 
Ro
a
r
i
ng B
r
a
n
c
h
 
3.
7 
30 27 
39 
Ur
b
a
n
 
77 
  T
a
bl
e 2.
  L
i
s
t
 
o
f
 hy
dr
o
l
o
g
i
ca
l
 
var
i
a
b
l
e
s
 us
ed 
i
n
 a
n
a
l
y
s
es
 a
n
d t
h
e
i
r
 r
a
n
g
e o
f
 
v
a
l
u
e
s
,
 P
ear
s
o
n
 cor
r
e
l
a
t
i
o
n
s
 t
o
 pr
ed
o
m
i
n
a
n
t
 l
a
n
d
 c
ov
e
r
 
c
l
a
s
s
e
s
 
i
n
 t
h
e 18 wat
e
r
s
h
e
ds
,
 an
d 
l
o
ad
i
ngs
 o
n
 Pr
i
n
c
i
pa
l
 
C
o
m
p
o
n
e
n
t
s
 An
a
l
y
s
i
s
 (
P
C
A
)
 axes
.
  
IS
 
=
 p
r
op
or
ti
on
 of
 
i
m
p
e
r
v
i
ou
s
 s
u
r
f
ac
e
 
co
v
e
r
,
 
Fo
r
e
st
 = pr
o
p
or
t
i
o
n
 o
f
 
f
o
r
e
s
t
 (
m
a
n
aged + u
n
m
a
n
a
ged)
 co
v
e
r
,
 
Pa
st
u
r
e
 =
 p
r
op
or
ti
on
 of
 p
a
s
t
u
r
e
 
c
o
v
e
r
,
 
P
CA1
 an
d 
P
CA2 
= 
Pr
i
n
c
i
p
a
l
 C
o
m
p
o
n
e
n
t
s
 Ana
l
y
s
i
s
 a
x
e
s
 1 an
d 2,
 r
e
s
p
ect
i
v
e
l
y
.
  *  p
 < 0
.
05,
 ** p
 < 0.
01
.
 
 
V
a
r
i
a
b
l
e
 ID
 
D
e
sc
r
i
p
t
i
o
n
 
R
a
n
g
e
  
IS
  
Fo
r
e
st
 
Pa
st
u
r
e
 
M
agnitude 
 
  
 
Me
d
Q
 
Me
d
i
a
n
 d
i
sc
h
a
r
g
e
 (
L
 s
-1
)
 
0.
01 ?
 0.
90 
 
 
0.
49* 
Ma
x
Q
 
Ma
x
i
m
u
m
 d
i
sc
h
a
r
g
e
 (
L
 s
-1
)
 
0.
54 ?
 21.
98 
 
 
 
Mi
n
Q
 
Mi
n
i
m
u
m
 d
i
sc
h
a
rg
e
 (L
 s
-1
)
 
0 ? 0
.
38 
 
 
 
Fr
equ
e
n
cy (
#
 t
i
m
es
 exceeded t
h
r
e
s
h
ol
d)
 
  
 
 
3x
M
e
d 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 3
x
 
m
e
d
i
a
n
 
f
l
o
w
 
5 ? 116 
 
 
 
5x
M
e
d 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 5
x
 
m
e
d
i
a
n
 
f
l
o
w
 
1 ? 70 
0.
50* 
 
 
7x
M
e
d 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 7
x
 
m
e
d
i
a
n
 
f
l
o
w
 
1 ? 64 
0.
56* 
 
 
9x
M
e
d
 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 9
x
 
m
e
d
i
a
n
 
f
l
o
w
 
0 ? 58 
0.
58* 
 
 
>75
th
 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 75t
h
 per
c
e
n
t
i
l
e
 
25 ? 115 
 
 
 
>95
th
 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 95t
h
 per
c
e
n
t
i
l
e
 
12 ? 66 
 
 
 
>99
th
 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 99t
h
 per
c
e
n
t
i
l
e
 
2 ? 35 
0.
62** 
 
 
D
u
ra
t
i
o
n
 (
#
 o
f
 h
o
u
rs sp
e
n
t
 a
b
o
v
e
 t
h
re
sh
o
l
d
)
 
  
 
 
>3xM
ed_d 
Ho
ur
s
 di
s
c
h
a
r
g
e was
 >3
x
 
m
e
d
i
a
n
 
f
l
o
w
 
36.
5 ?
 3026 
 
 
 
>5xM
ed_d 
Ho
ur
s
 di
s
c
h
a
r
g
e was
 >5
x
 
m
e
d
i
a
n
 
f
l
o
w
 
6 ? 2518 
 
 
 
>7xM
ed_d 
Ho
ur
s
 di
s
c
h
a
r
g
e was
 >7
x
 
m
e
d
i
a
n
 
f
l
o
w
 
1.
5 ?
 2412 
 
 
 
>9xM
ed_d 
Ho
ur
s
 di
s
c
h
a
r
g
e was
 >9
x
 
m
e
d
i
a
n
 
f
l
o
w
 
0 ? 17 
 
 
 
P
r
e
d
ic
tability
 and F
l
as
hine
s
s
 
 
  
 
 
 
C
.
V.
 
%
 C
o
ef
f
i
c
i
e
n
t
 o
f
 Var
i
at
i
o
n
 
42 ? 402 
 
 
 
I
n
c1
h
100 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 100%
 w
i
t
h
i
n
 1 
h
r
 
7 ? 109 
0.
55* 
 
 
I
n
c1
h
1000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 1000%
 w
i
t
h
i
n
 1
 h
r
 
0 ? 31 
 
 
 
I
n
c1
h
5000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 5000%
 w
i
t
h
i
n
 1
 h
r
 
0 ? 17 
 
 
-
0
.
47* 
78 
  
I
n
c3
h
100 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 100%
 w
i
t
h
i
n
 3 
h
r
 
11 ? 122 
0.
53* 
 
 
I
n
c3
h
1000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 1000%
 w
i
t
h
i
n
 3
 h
r
 
0 ? 44 
 
 
 
I
n
c3
h
5000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 5000%
 w
i
t
h
i
n
 3
 h
r
 
0 ? 19 
 
 
 
Dec1h
100 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 100%
 w
i
t
h
i
n
 1 h
r
 
0 ? 67 
 
 
 
Dec1h
1000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 1000%
 w
i
t
h
i
n
 1
 h
r
 
0 ? 28 
 
 
 
Dec1h
5000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 5000%
 w
i
t
h
i
n
 1
 h
r
 
0 ? 12 
 
 
 
Dec3h
100 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 100%
 w
i
t
h
i
n
 3 h
r
 
1 ? 92 
 
 
 
Dec3h
1000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 1000%
 w
i
t
h
i
n
 3
 h
r
 
0 ? 36 
 
 
 
Dec3h
5000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 5000%
 w
i
t
h
i
n
 3
 h
r
 
0 ? 18 
 
 
 
B
a
s
e
flo
w
 (L s
-1
) 
 
  
 
M
e
d_
bf
 
Me
d
i
a
n
 b
a
se
f
l
o
w
  (
L
 s
-1
)
 
0.
04 ?
 730 
 
 
 
BI
 
B
a
se
f
l
ow
 inde
x (
?
 pr
e
d
ic
t
e
d
 ba
se
f
l
ow
/ 
?
o
b
s
e
r
v
ed
 fl
o
w
) 
0.
03 ?
 0.
82 
 
 
 
  
 
 
  
   
79 
    T
a
b
l
e
 3
.
  P
h
y
s
i
c
oc
h
e
m
i
cal
 an
d
 
h
a
b
i
t
a
t
 (
e
n
v
i
r
on
m
e
n
t
al
)
 v
a
r
i
ab
l
e
s
 u
s
e
d
 
i
n
 an
al
y
s
es
,
 t
h
e
 r
a
n
g
e
 of
 v
a
l
u
es
,
 a
n
d
 t
h
ei
r
 P
e
a
r
s
o
n
 c
o
r
r
el
at
i
o
n
s
 
t
o
 pr
ed
o
m
i
n
a
n
t
 l
a
n
d
 co
v
e
r
 c
l
as
s
e
s
 
i
n
 t
h
e 18 wat
e
r
s
h
e
d
s
.
  
IS
 
= p
r
o
p
o
r
t
i
o
n
 o
f
 
i
m
p
er
vi
o
u
s
 s
u
r
f
ace co
v
e
r
,
 
Fo
r
e
st
 =
 p
r
op
o
r
ti
on
 of
 
f
o
r
e
s
t
 
co
v
e
r
 (
m
a
n
ag
ed + unm
a
n
aged)
,
 
Pa
st
u
r
e
 = pr
o
p
or
ti
o
n
 o
f
 pas
t
ur
e co
v
e
r
.
  *
  p
 < 0.
05
,
 **
 p <
 0.
01
,
 p
 < 
0.
001***.
 
 
  
  
 
 
V
a
r
i
a
b
l
e
 ID
 
D
e
sc
r
i
p
t
i
o
n
 
R
a
n
g
e
  
IS
  
Fo
r
e
st
 
Pa
st
u
r
e
 
T
e
m
p
 
M
e
d
i
a
n
 wat
e
r
 
t
e
m
p
er
at
ur
e (
o
C
)
 
13.
3 ?
 15.
6 
0.
67** 
-
0
.
54* 
 
D
O
 
Me
a
n
 d
i
sso
l
v
e
d
 O
2
 (
m
g 
L
-1
)
 
8.
6 ?
 14.
5 
-
0
.
50* 
 
 
mi
nDO 
M
i
nim
u
m
 d
i
s
s
o
l
v
e
d O
2
 (
m
g L
-1
)
 
0.
2 ?
 8.
3 
 
 
 
pH 
M
e
a
n
 pH 
5.
7 ?
 6.
9
  
 
 
 
TD
S
 
Me
a
n
 t
o
t
a
l
 d
i
sso
l
v
e
d
 so
l
i
d
s c
o
n
c
e
n
t
r
a
t
i
o
n
 (
m
g
 L
-1
)
 
19.
9 ?
 58.
6 
0.
58* 
 
 
T
S
S
 
Me
a
n
 t
o
t
a
l
 su
sp
e
n
d
e
d
 so
l
i
d
s c
o
n
c
e
n
t
r
a
t
i
o
n
 (
m
g
 L
-1
)
 
2.
1 ?
 8.
1 
 
 
0.
54* 
V
o
l
u
me
 
M
e
a
n
 d
e
p
t
h x w
i
d
t
h 
x 
l
e
ng
t
h
 o
f
 
ha
b
i
t
a
t
 s
a
m
p
le
d
 (
m
3
)
 
0.
6 ?
 6.
9 
0.
76*** 
-
0
.
65** 
 
OM
  
B
e
n
t
hi
c o
r
gani
c 
m
a
t
t
e
r
 
(
g
)
 
  0
.
3 ?
 1.
2 
 
0.
47* 
 
Sub
s
t
r
at
e 
M
e
d
i
a
n
 s
u
b
s
t
r
at
e
 s
i
ze (
c
m
)
 
  0
.
5 ?
 1.
8 
  0
.
52* 
 
 
T
g
 
T
r
ac
t
i
ve 
f
o
r
ce 
  2
.
7? 73
.
9
 
 
 
 
Habi
t
a
t
 I
n
dex
 
Habi
t
a
t
 as
s
e
s
s
m
e
n
t
 i
n
de
x
 s
c
o
r
e  
  
  54
.
2
 ?
 125.
5 
 
0.
48* 
 
80 
 
 81
 
Table 4.  Associations between select hydrologic variables and environmental variables.  
Values are Pearson correlation coefficients, and only those with a significant correlation 
are shown.  *  p < 0.05, ** p < 0.01, p < 0.001***. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
                
 MedQ 5xMed 7xMed 9xMed Inc1h100 Inc1h5000 Inc3h100
Temp  0.55* 0.66** 0.69** 0.48*   
DO  -0.59* -0.69** -0.72**    
pH     0.76***  0.73** 
TDS  0.78*** 0.84*** 0.85***    
TSS      -0.69**  
Substrate  0.72** 0.74*** 0.73** 0.53*  0.48* 
Tg 0.52*       
Habitat 
Index           0.55*   
 
 82
 
Table 5.  Fish species collected and Pearson correlation coefficients to non-metric 
multidimensional scaling (NMDS) axes 1 and 3.  Rare species excluded from NMDS 
analyses are denoted by dashes. Breeding guild abbreviations: S = Simple miscellaneous, 
L = Simple lithophil, Cpc = Complex with parental care, Cnc = Complex with no parental 
care.  Feeding guild abbreviations: P = Predator, I = Insectivore, H = Herbivore, O = 
Omnivore, F = Filterer.  
 
 
 
 
    
  
Family Species Common name Breed Feed NMDS
1
NMDS
3
Catastomidae Erimyzon oblongus Creek chubsucker S I 0.38 0.18 
 Hypentelium etowanum
Alabama hog 
sucker 
L I 0.12 0.22 
 Minytrema melanops Spotted sucker L I - - 
Centrarchidae Lepomis auritus Redbreast sunfish Cpc I -0.37 0.72 
 L. cyanellus Green sunfish Cpc I -0.39 0.19 
 L. gulosus Warmouth Cpc P -0.29 -0.34 
 L. macrochirus Bluegill Cpc I -0.52 -0.55 
 L. megalotis Longear sunfish Cpc I -0.39 -0.43 
 L. miniatus 
Redspotted 
sunfish 
Cpc I 0.06 0.22 
 Micropterus salmoides Largemouth bass Cpc P -0.29 0.13 
Cyprinidae Campostoma pauciradii Bluefin stoneroller Cnc H 0.04 0.06 
 Luxilus zonistius Bandfin shiner L I 0.81 0.24 
 Nocomis leptocephalus Bluehead chub Cnc O 0.82 0.17 
 
Notemigonus 
crysoleucas 
Golden shiner S I -0.35 -0.30 
 Notropis baileyi Rough shiner L I 0.65 -0.19 
 N. buccatus Silverjaw minnow S I 0.19 -0.37 
 N. longirostris Longnose shiner L L 0.48 -0.11 
 N. texanus Weed shiner S I -0.43 -0.12 
 Semotilus atromaculatus Creek chub Cnc O 0.61 0.07 
Ictaluridae Ameirus natalis Yellow bullhead Cpc O -0.40 0.16 
 A. nebulosus Brown bullhead Cpc O 0.07 0.37 
 Ictalurus punctatus Channel catfish Cpc O - - 
 Noturus leptocanthus Speckled madtom Cpc I - - 
Percidae Percina nigrofasciata 
Blackbanded 
darter 
L I -0.17 -0.29 
Petromyzontidae Ichthyomyzon gagei Brook lamprey Cnc F 0.38 0.37 
Poeciliidae Gambusia affinis 
Western 
mosquitofish 
Cnc I -0.43 0.14 
 
 83
 
Table 6  Fish metrics and their correlations watershed LU/LC and with best NMDS axes.   
Only significant correlations are shown.  IS = proportion of impervious surface cover, 
Forest = proportion of forest cover, Pasture = proportion of pasture cover, NMDS
1
 and 
NMDS
2
 are NMDS axes 1 and 2, respectively.  Fish metrics are as described  in text.  *  
p < 0.05, ** p < 0.01. 
 
   
  
Metric IS Forest Pasture NMDS
1
 NMDS
3
 
Breeding      
Parent Care     0.78*** 
No Parent Care    -0.61**  
Simple    
Lithophilic -0.66** 0.51*  0.87***  
Feeding   
Piscivore      
Insectivore 0.57* -0.51*  -0.57*  
Herbivore    
Omnivore  0.61**  
Filterer   0.50* 
Assemblage    
Number    0.50*  
Richness -0.66**   0.46*  
Diversity -0.85*** 0.52*  0.55*  
% Sunfish 0.58*   0.75***  
IBI   0.48* 
 
 
 
 
 
 
 
 
 
 
 84
Table 7.  Best multiple regression models for fish assemblage variables with standardized 
regression coefficients and R
2
adj
.  Fish metric definitions are as in text, Predictor 
definitions are as in Tables 2 and 3.  *  p < 0.05, ** p < 0.01, *** p < 0.001. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
        
Standardized 
Metric Predictors 
Estimate 
R
2
adj
 
   
N Temp -0.58 0.29** 
    
Richness 5xMed -0.68 0.43** 
 Temp -0.40  
   
Diversity TDS -0.54 
0.54**
* 
 N>99 -0.40  
   
%Sunfish TDS 0.73 0.48** 
 MedQ 0.43  
   
IBI 7xMed -0.52 0.22* 
 
 85
 
Figure 1.  Conceptual model of the potential causal relationships between land use/land 
cover, abiotic conditions, and biotic assemblages of streams.  
 
 86
  
 
Figure 2.  The study area included 18 small watersheds (shaded) of the Chattahoochee 
River basin in 4 counties in the Lower Southern Piedmont ecoregion.  The city of 
Watershed 
LU/LC 
Hydrology 
 
Stream biotic 
assemblages 
Physico-
chemistry 
Geo-
morphology 
 
 87
Columbus is located in western Muscogee County.  County names are in capital letters. 
Numbers refer to watersheds and correspond to codes in Table 1. 
 
 
 
 
Figure 3.   Non-metric multidimensional scaling ordination (NMDS) of sites in ordination 
space.  Axes are scaled proportionate to the longest axis (% of max).  Symbols are the 18 
study sites coded by land use classifications as described in text with vectors that show 
Columbus
2002 0 Ki l om ete rs
GA
?
?
Atlanta
9
3
10
8
4
2
6
14 
1
12
13
7
11
5
18
16
15
17
MUSCOGEE 
HARRIS 
TROUP MERIWETHER 
 
 88
relative direction and strength of correlated environmental variables.  Arrows on x and y 
axes show direction of correlated fish assemblage values.  Axes 1 and 3 explain 57.1 and 
25.3% of the total variation, respectively.   
 
 
 
 
Substrate
Temp
DO
TDS
5med
7med
9med 
Volume 
-1.5
-1.0
-1.0 -0.5 0.0 0.5 1.0 1.5
-0.5
0.0
0.5
1.0
1.5
Axis 1
LU/LC
Forest
Pasture
Developing
Urban
Axi
s
 3
 
BI 
%Sunfish 
Diversity, N, %Lithophils, IBI 
 
 89
CHAPTER 4.  Land use impacts on stream macroinvertebrate assemblages: temporal 
variability and the interplay of hydrology, physicochemistry, and geomorphology 
4.1  ABSTRACT 
I examined the influence of land use / land cover (LU/LC) on macroinvertebrate 
assemblages over a 12-mo period of record in streams draining 20 small watersheds in 
the Lower Piedmont of Georgia, USA.  Specifically, I quantified seasonal relationships 
between LU/LC (as % impervious surface [IS], pasture, silviculture, and mixed forest) 
and several macroinvertebrate metrics as well as the relative influence of hydrology, 
physicochemistry, and geomorphology on annual values of macroinvertebrate metrics.  
There were seasonal differences in all macroinvertebrate metrics, but most revealed no 
seasonal interaction with LU/LC.  High-IS watersheds generally had more biologically 
impaired streams (with degraded conditions generally appearing at ~10% IS) than 
watersheds containing intense pasture, silviculture, and suburban development, compared 
with forested reference watersheds.   Several elements of hydrology, geomorphology, and 
physicochemistry were related to LU/LC, including measures of spate flow frequency, 
benthic habitat conditions, stream temperature, sulfate, nitrate, and fecal coliform levels, 
although only physicochemical and geomorphological variables were associated with 
macroinvertebrate variables.  Nonmetric multidimensional scaling revealed that 
macroinvertebrate assemblages were associated with benthic detritus cover, dissolved 
oxygen levels, stream water temperature, and total dissolved solid concentration.  Our
 
 90
results suggest that the influence of LU/LC on aquatic biota remains invariant throughout 
the year and that macroinvertebrate assemblages are more directly influenced by changes 
in physicochemistry and benthic habitat conditions than by hydrological alterations 
associated with LU/LC in the watersheds of this region. 
4.2  INTRODUCTION 
 Human activities are responsible for much of the alteration in land-use/land cover 
(LU/LC) worldwide, and rivers and streams are among the most affected ecosystems by 
these changes (Hammett 1992, Schnaiberg et al. 2002).  LU/LC within a watershed can 
influence the overall quality of stream ecosystems by directly altering hydrology, 
physicochemical conditions, and geomorphology, which, in turn, determine biotic 
composition  (Bormann et al. 1999, Peterson and Kwak 1999, Clements et al. 2000, 
Bledsoe and Watson 2001, Riis and Sand-Jensen 2001, Maloney et al. 2005).  Ultimately, 
even slight changes in LU/LC can dramatically alter stream biotic integrity by increasing 
physiological stress of individuals, reducing population sizes, and causing alterations in 
species diversity and other measures of community structure (Lenat and Crawford 1994, 
Paul and Meyer 2001, Roy et al. 2003, Walters et al. 2003, Allan 2004, Maloney and 
Feminella 2006). 
 In the southeastern United States, silviculture, agriculture, and urbanization are 
common types of land use, and disturbance from each type may apply its own unique 
suite of pressures on receiving streams (Brown et al. 2005). Timber harvest and 
agricultural practices often degrade receiving waters through the alteration/removal of 
riparian and watershed vegetation, which can increase stream bank failure and 
sedimentation, nutrient and pesticide concentrations and loads, light levels, water 
 
 91
temperature, and productivity (Allan 2004, Fortino et al. 2004).  Disturbance from 
silviculture is generally not thought to be as damaging as row crop agriculture since post-
harvest vegetation regrowth often approaches predisturbance levels, whereas it does not 
with row crop agriculture (Campbell and Doeg 1989, Fortino et al. 2004); however the 
impacts of both are often minimized with implementation of Best Management Practices 
(BMP) guidelines (Prud'homme and Greis 2002, Wang et al. 2002, Fortino et al. 2004).  
Neither silviculture nor agriculture is as damaging to receiving waters as are the 
physical and chemical impacts from urbanization.  The substantial influence of watershed 
urbanization on streams are often considered a direct result of, or mediated by, altered 
hydrologic regimes (Paul and Meyer 2001, Allan 2004).  Increases in impervious surface 
runoff and stormwater conveyance can dramatically increase magnitude and frequency of 
storm flows, accelerating nutrient and pollutant transport, altering thermal and 
decomposition dynamics, eroding streambeds, and displacing organisms (Jones and Clark 
1987, Walsh et al. 2001, Walters et al. 2003, Roy et al. 2005, Chadwick et al. 2006).  The 
predictable biotic impacts from watershed disturbance has led to the identification of a 
generalized ecosystem distress syndrome (or specifically ?urban stream syndrome? in 
urbanized watersheds (sensu (Meyer et al. 2005), which includes reduced biodiversity, 
altered productivity, and increased abundance of tolerant and exotic species (Rapport et 
al. 1985, Walsh et al. 2005b).    
 Effects of LU/LC on streams often are regional and episodic, varying 
geographically and seasonally, usually becoming more pronounced with increased 
precipitation and discharge (Weaver and Garman 1994, Helms et al. 2005, Walsh et al. 
2005b).   In addition, the shape of biotic response (e.g., linear, threshold, subsidy-stress) 
 
 92
to LU/LC disturbance depends upon the species sensitivity and the stressor?s mode of 
action (e.g. toxicity, sub-lethal stress), properties that also can vary spatially and 
temporally (Allan 2004).  Further, specific mechanisms of LU/LC-associated impacts on 
aquatic systems have been elusive because of their interrelated nature (Paul and Meyer 
2001, Allan 2004).  A more comprehensive understanding of temporal and spatial 
influences of LU/LC on receiving waters at the regional scale is essential for effective 
monitoring, assessment, and management of aquatic resources in streams draining 
degraded watersheds.  Our study was designed to further examine these LU/LC impacts 
on stream macroinvertebrates in the Southern Outer Piedmont of the US, and examine 
potential mechanisms associated with any observed biotic changes.  Specifically, I 
quantified 1) the relative seasonal impact of various LU/LC conditions on stream 
macroinvertebrate assemblages, and 2) the relative influence of putative hydrological, 
chemical, and geomorphological stressors associated with LU/LC disturbance.  This 
information will allow for a more time-integrated, mechanistic understanding of LU/LC 
impacts on stream biota in attempts to minimize anthropogenic impacts on freshwater 
systems. 
4.3  METHODS 
4.3.1  Study area 
 I studied stream reaches from tributaries of the middle Chattahoochee River, 
western Georgia, USA, occurring in the Southern Outer Piedmont ecoregion, 45b 
(Griffith et al. 2001).   Conversion of pasture and forests to urbanized areas is occurring 
rapidly northeast of the city of Columbus, Muscogee County.  Therefore, I sampled 2
nd
- 
to 3
rd
-order streams within 20 watersheds (4?25 km
2
) along an urbanization gradient from 
 
 93
the geologic fall line in the city of Columbus to an area 80 km northeast (Muscogee, 
Harris, Troup and Meriwether counties, Figure 1).  Stream study reaches were ~100 m 
and typical of the lower Piedmont, consisting of sandy-bottom channels and run-pool 
morphologies.    The relatively large range in landscape character in the region allowed 
comparison across geomorphically similar streams that differed primarily in watershed-
level LU/LC and associated variation in streamwater physicochemical conditions.   
4.3.2 Land cover / land use analysis 
 I determined watershed boundaries and size from US Geological Survey 30-m 
resolution digital elevation models and ArcView 3.2a software (Environmental Research 
Systems Institute, Inc., Redwoods, California).   In March 2003, true-color 1-m resolution 
aerial photographs of the study watersheds were taken coinciding with leaf-off.  
Impervious surfaces (IS) were manually digitized and remaining land cover was 
classified using a hybrid unsupervised/supervised technique, resulting in a land cover 
classification similar to the Anderson Classification Scheme (Myeong et al. 2001).  For 
each watershed I determined % IS, % pasture, % evergreen forest, and % deciduous 
forest for analyses.  Detailed image processing methods for the study watersheds are 
presented elsewhere (Lockaby et al. 2005).  I then inferred land use by grouping each 
watershed into the following categories based on predominant land cover and assumed 
influential land use assessed in the field: Forest (mixed evergreen/deciduous forests); 
Silviculture (heavily managed pine forests); Pasture (cattle grazing and/or forage 
production); Developing (forested but with current or recent residential development); 
and Urban (downtown Columbus) (Table 1).    
 
 
 94
4.3.3 Stream hydrology variables 
 I quantified stream discharge (Q) from June 2003 to June 2004 using a Mini-
Troll? pressure-transducer data logger (In-Situ Inc., Ft. Collins, Colorado) installed in 
each study stream.  Data loggers were set to measure water depth every 15 min (0.01-m 
depth resolution), and I then developed depth?Q rating curves to estimate continuous Q 
(Schoonover et al. 2006).   I characterized 5 separate elements of Q from each watershed 
hydrograph: 1) magnitude (amount of Q for a given interval), 2) frequency (number of 
occurrences of a given magnitude of Q, 3) duration (length of time associated with a 
particular high or low Q event, 4) predictability/flashiness (coefficient of variation and 
rate of change of Q), and 5) baseflow (predicted baseflow contribution to Q and baseflow 
index, BI) (Richter et al. 1996, Poff et al. 1997, McMahon et al. 2003).   I predicted 
baseflow for each watershed using a 5-d smoothed minima technique (Gustard et al. 
1992) and calculated the BI as: 
? predicted baseflow / ? observed flow 
 where 1 = no overland contribution and 0 = complete overland contribution to observed 
flow (Gustard et al. 1992, Schoonover et al. 2006).  Ultimately, I calculated 28 
hydrologic variables considered important in determining aquatic biota (Richter et al. 
1996, Poff et al. 1997) (Table 2). 
4.3.4 Physicochemical variables 
 I measured several stream-specific physical and chemical variables thought to be 
important in determining biotic assemblages over the hydrological period of record 
(Table 3).  I measured stream temperature continuously with HOBO? Temp data loggers 
placed near pressure transducers used for Q.  Dissolved oxygen (DO) and streamwater 
 
 95
pH were measured seasonally.  In addition, I quantified several nutrient and sediment 
parameters, including total suspended solids (TSS), total dissolved solids (TDS), fecal 
coliform colonies (FC), total phosphorus (P), dissolved organic carbon (DOC), SO
4
2-
 and 
NO
3
?
 concentrations.  I measured each of the above parameters by taking with 
streamwater grab samples collected monthly for each watershed (see methods in 
(Schoonover and Lockaby 2006). 
4.3.5 Macroinvertebrates 
 I quantified benthic macroinvertebrates in 3 pools and 3 runs per stream reach 
during September (summer), February (winter), and April ? May (spring) in 2003 ? 2004 
using a Surber sampler (250-?m mesh; 0.093 m
2
 sampling area).  I consolidated multiple 
Surber collections in each habitat (3 for pools, 4 for runs), resulting in a 0.27- and  0.36-
m
2
 sample from each pool and run, respectively (1.89 m
2
 total area sampled for each 
study reach per stream and season).  I elutriated excess sediment and preserved all 
samples with 95% EtOH in the field.  In the laboratory, I removed all organisms >2mm 
with the unaided eye, and subsampled the remainder under a dissecting microscope (? 
300 organisms, (Vinson and Hawkins 1996).  I counted, measured length (nearest mm), 
and identified macroinvertebrates to the lowest possible taxonomic level (usually genus 
or morphospecies) using keys in (Merritt and Cummins 1996), (Wiggins 1996), and 
(Epler 2001) and also assigned macroinvertebrates to functional feeding, habit, and 
tolerance groups (Merritt and Cummins 1996). 
 I calculated several benthic macroinvertebrate metrics selected from standard US 
Environmental Protection Agency (USEPA) bioassessment protocols (Barbour et al. 
1999) spanning 4 broad categories of metrics (Table 5): (1) Taxonomic metrics, which 
 
 96
included the proportion and/or taxa count of particular taxonomic groups, such as 
Ephemeroptera, Plecoptera, Trichoptera, and other putatively sensitive groups; (2) 
Functional metrics, which included the proportion and/or taxa count of 
macroinvertebrates in various functional feeding groups (FFGs), habitat uses, and 
tolerance classes; (3) Community variables, which included diversity (as Shannon?s H?), 
taxa richness, total invertebrate biomass using published length-mass relationships 
(Benke et al. 1999), and density; and (4) a single Composite variable, which was the 
Georgia Benthic Macroinvertebrate Index (GA-BMI, (GADNR 2007).  The GA-BMI is 
an ecoregion-specific multimetric index designed for streams in Georgia, with 6 metrics 
calculated for the Southern Outer Piedmont, where scores ? 84 = very good, 56 ? 83 = 
good, 32 ? 55 = Fair, 17 -31 = Poor, and < 17 = Very Poor (GADNR 2007).  For seasonal 
analyses, I used seasonal macroinvertebrate values from each watershed, whereas for 
annual analyses I averaged seasonal values to obtain one value per watershed.  
I characterized stream microhabitat when macroinvertebrates were sampled by 
quantifying mean depth, width, current velocity (Marsh-McBirney model 2000 flow 
meter) and benthic substrate composition at each run and each pool habitat in each season 
(Table 4).   Benthic substrate characterization consisted of estimating the % cover of clay, 
silt, sand, gravel, cobble, wood, detritus, and wood in 10% increments in each habitat 
sampled.  All substrate estimates were done by a single observer (BSH) to reduce 
observational error.  In addition to the measures taken seasonally, I quantified mean 
benthic organic matter (BOM), substrate size, and tractive force (shear stress, T
G 
(Gordon 
et al. 1992))  for each watershed once over the study.  I estimated BOM and substrate size 
by sampling in transitional areas between runs and pools in all streams to minimize the 
 
 97
influence of deposition and scour on our estimates. I quantified BOM by determining the 
ash-free dry mass (AFDM) of 9 replicate 2.5 x 10 cm benthic cores.  For substrate 
particle size I collected 3 benthic samples per stream (near where BOM was sampled) 
using a 76.2-mm diameter PVC core inserted into the substrate to a depth of 10 cm.  I 
dried samples and separated particles into 5 size classes: gravel-cobble (>2mm), very 
coarse sand (1 ? 2 mm), coarse to medium sand (0.25 ? 1 mm), fine sand (0.1 ? 0.25 
mm), very fine sand (0.05 ? 0.1mm), and silt/clay (< 0.05mm) to determine median 
substrate size by mass.   I estimated T
G
 close to where pressure transducers using the 
formula 
    T
G
 = pgRS 
where p is the density of water, g is gravitational acceleration, R is hydraulic radius, and 
S is slope of the energy line (Newbury 1996). 
4.3.6  Statistical analyses 
I used general linear models (GLMs) to describe broad differences in 
macroinvertebrate community and composition metrics among LU/LC categories, 
seasons, and potential LU/LC?season interactions.  Following GLM, I assessed any 
differences between sites and seasons with Tukey HSD multiple comparison tests (Zar 
1999).  Two sites (FR and BR) were only sampled for macroinvertebrates once (Spring) 
and thus had no accompanying hydrographs; therefore, I excluded these watersheds from 
all subsequent multivariate analyses. 
Simple Pearson correlation analyses were used to relate hydrologic, habitat, and 
chemical variables to LU/LC variables.  I then conducted 3 separate Principal Component 
Analyses (PCA) to describe overall variation in hydrologic, physicochemical, and habitat 
 
 98
conditions, respectively, in study watersheds.  Annual values of all environmental 
variables for each particular group (hydrology = 28 variables, physicochemistry = 11, 
habitat = 12) were used in the respective PCA ordinations, resulting in 3 site ? parameter 
matrices, upon which ordinations were constructed (18 x 28 hydrology, 18 x 11 
physicochemistry, and 18 x 12 habitat).  I transformed all variables as needed to meet 
assumptions of normality and based ordinations on the correlation cross-products matrix.  
The resulting ordinations were used to isolate groups of variables explaining a large 
proportion of the environmental variance, with first 2 PC axes from each ordination used 
as predictor variables in subsequent multiple regression analyses between community and 
composition macroinvertebrate variables.      
I used nonmetric multidimensional scaling (NMDS) to describe overall variation 
in macroinvertebrate assemblages among sites. NMDS, an ordination technique that uses 
species data to create pairwise similarity matrices to determine positions of sites in 
species space, handles nonnormal data (e.g., with many zero counts) better than other 
ordination techniques (Hawkins et al. 1997, McCune and Grace 2002).  NMDS 
ordinations were based upon annual presence/absence of macroinvertebrates, with rare 
taxa (those occurring in only 1 site) being excluded.  This step resulted in an 18 x 113 
site-species matrix on which ordinations were based using a Sorenson distance measure 
(McCune and Grace 2002).  I then correlated environmental and macroinvertebrate 
responses to the 2 most explanatory axes of the NMDS.   
 I used simple regressions to determine the best LU/LC predictor for the 
community and composite macroinvertebrate metrics.  Then, using the first 2 axes from 
each PCA as predictors and the composite/community metrics and the best NMDS axis 
 
 99
as response variables, I used stepwise multiple regression (p = 0.05 to enter and leave the 
model) to determine the environmental parameter groups with the most explanatory 
power regarding annual values of macroinvertebrate metrics.   Parameters in final models 
had a variance inflation factor (VIF) <10; thus, multicollinearity was avoided (Myers 
1990).   
 I used SAS (version 8.2, SAS Institute Inc., Cary, North Carolina) for all 
correlation and regression analyses and PC-ORD (version 4, MjM Software Design, 
Glenden Beach, Oregon) for PCA and NMDS.  For all analyses, ? = 0.05.  
4.4 Results 
4.4.1 Environmental variables and LU/LC 
 Many environmental variables showed strong associations with LU/LC.  In 
general, spate flow frequency and overall flashiness increased with % IS, as evidenced by 
the positive relationships between IS and 5xMed, 7xMed, 9xMed, N>99, and Inc3h100 
(Table 2).   Also, substrate and habitat conditions differed across sites given contrasting 
LU/LC, with substrate size increasing and benthic detritus decreasing with % IS, benthic 
detritus increasing with % deciduous forest, and OM increasing with % evergreen forest 
(Table 3).  Chemical conditions also were related to LU/LC.  Temp, TDS, SO
4
 and FC all 
increased with increasing IS and decreased with increasing deciduous cover (Table 4).  
Further, DO increased with % deciduous cover and decreased with % IS, whereas NO
3
 
was positively associated with % pasture and negatively associated with % evergreen 
forest (Table 4; see (Schoonover and Lockaby 2006, Schoonover et al. 2006) for details 
about hydrologic and chemical associations with LU/LC). 
 
 
 100
4.4.2 Macroinvertebrate assemblages 
 I collected a total of 148 unique taxa in 52 families, with Chironomidae (Diptera), 
Hydropsychidae (Trichoptera), Tubificidae (Haplotaxida) and Simuliidae (Diptera) being 
the 4 most common groups.  Gatherers and Filterers generally were the most abundant 
FFGs, whereas Clingers and Scrapers were the least abundant (Table 5).  Filterers had the 
highest single occurrence (64% of total abundance; BU2) whereas Shredders were 
virtually absent from several streams, primarily in winter and spring.  Numerical 
dominance of FFGs varied by season, with Shredders and Filterers being dominant in 
summer, Gatherers in winter, and Filterers and Gatherers in spring (Table 5).  Clingers, 
Filterers, Gatherers, and Predators occurred at all sites.  
 H? ranged from 0.79 to 2.90, taxa richness ranged from 14 to 64, biomass ranged 
from 1 to 759 g / m
2
, and density ranged from 2504 to 68,429 individuals / m
2
.  GLM 
indicated strong seasonal and LU/LC differences in macroinvertebrate metrics.  H? was 
highest in summer (F
2, 55
 = 3.47, p = 0.04), richness (F
2, 55
 = 6.21, p = 0.0004) and 
biomass (F
2, 55
 = 5.33, p = 0.006) were highest in spring, and density (F
2, 55
 = 4.2, p = 
0.022) was highest in winter (Table 6, Fig. 2).   GA-BMI ranged from 3 to 75 and was 
highest in summer (F
2, 55
 = 27.93, p < 0.001) (Table 6, Fig. 2).  Forest watersheds 
generally showed higher richness (F
4, 55
 = 24.96, p< 0.001), H? (F
4, 55
 = 5.89, p = 0.001), 
and GA-BMI (F
4, 55
 = 18.12, p < 0.001) than Urban watersheds, with Silviculture, 
Pasture, and Developing watersheds generally being intermediate between Forest and 
Urban watersheds (Fig. 2).  In contrast, mean biomass was higher (F
4, 55
 = 3.96, p = 
0.011) in Urban than Forest, Silviculture, and Pasture watersheds (Fig. 2). 
 
 101
 Across seasons, the best LU/LC predictors of community and composite 
macroinvertebrate metrics were % deciduous forest and IS (Table 7).  % deciduous cover 
best described seasonal averages for richness (R
2
 = 0.73, p < 0.0001), H? (R
2
 = 0.42, p = 
0.002), and GA-BMI (R
2
 = 0.67, p < 0.0001) whereas % IS best described seasonal 
means for biomass (R
2
 = 0.39, p = 0.003).  GA-BMI also was inversely related to IS, with 
metric scores decreasing as a curvilinear function of increased % IS (Fig. 3). 
4.4.3 Multivariate relationships 
 In the hydrology PCA, the first 2 axes (H-PC
1
 and H-PC
2
) explained 69.8% of the 
variation (45.3 and 19.4%, respectively) among sites.  In general, median discharge 
(MedQ), median baseflow (MedBF), and groundwater contribution to baseflow (BI) 
decreased whereas spate frequency (as 5xMed, 7xMed) and flashiness (as Inc1h1000, 
Inc3h1000, Inc3h5000, Dec1h100, Dec3h100) increased with H-PC
1
 (Table 8).  
Maximum discharge (MaxQ) and spate frequency (as 3, 5, 7xMed) decreased whereas BI 
increased with H-PC
2
. 
 For substrate/habitat PCA, the first 2 PCA axes (S-PC
1
 and S-PC
2
) explained 
57.8% of the variation (35.1 and 22.7%, respectively).  Substrate size decreased along S-
PC
1
, with proportion of sand and wood being positively correlated and proportion of 
gravel, cobble, and bedrock negatively so with S-PC
1
 (Table 8).   Substrate size increased 
whereas organic matter (OM) and proportion of detritus decreased along S-PC
2
 (Table 8).   
For the water chemistry PCA, axis 1 (C-PC
1
) explained 50.1% and axis 2 (C-PC
2
) 
explained 20.9% of the variation (71% total).  Many chemical parameters loaded strongly 
on C-PC
1
, with temperature, pH, TDS, SO
4
, NH
4
, DOC, and FC all decreasing and DO 
 
 102
increasing along this axis.  TSS and NH
4
 increased and NO
3
 decreased along C-PC
2
 
(Table 8). 
 NMDS axes 1 and 2 described the most variation in macroinvertebrate 
assemblages among streams (79.3 and 9.8%, respectively; stress = 13.4, instability = 
0.00001, iterations = 66; Fig. 4).  Streams of similar watershed LU/LC generally grouped 
together, although streams in Urban watersheds grouped far to the left of the other groups 
(Fig. 4); thus, I considered NMDS
1
 to be a general urbanization axis, and used it as an 
invertebrate response variable in multiple regressions.   
Based on NMDS ordination, Diptera and EPT richness and % scrapers were more 
associated with Forest and Developing watersheds whereas % tolerant taxa and % 
Oligochaetes were higher in Urban than other sites.  Regarding composite and 
community metrics, GA-BMI, H?, and richness all were positively correlated with 
NMDS
1
 and negatively so with NMDS
2
 (Table 9)
.  
 In addition, biomass was negatively 
correlated with NMDS
1
 and density negatively so with NMDS
2
 (Table 8).  Last, spate 
flow frequency (5xMed, 7xMed, 9xMed), spate magnitude (N>99), substrate size, water 
temperature (MeanT, MaxT), DO, and SO
4
 all were negatively associated with NMDS
1
 
(Table 9).    
All multiple regression models based on the 1
st
 and 2
nd
 axes of the 3 PCAs for 
composite and community macroinvertebrates were significant, most with R
2
adj
 >0.75 
(Table 7).  Generally, water chemistry was the overall best predictor of composite and 
community macroinvertebrate responses, as evidenced by the presence and large 
standardized estimate of either C-PC
1
 or C-PC
2 
(or both) in all best multiple regression 
models (e.g., 11 of 15 significant factors were C-PC
1 
or C-PC
2
, Table 7).  In contrast, 
 
 103
hydrologic axes (i.e., H-PC
1 
and H-PC
2 
) were entirely absent as predictors from all best 
models.   
4.5 DISCUSSION 
 Environmental factors shaping benthic macroinvertebrate distributions in streams 
are numerous, often interactive, and frequently associated with landscape disturbance 
(Allan 2004).  Like other studies, our results show that LU/LC is particularly influential 
on macroinvertebrate assemblage structure and its effects persist across different seasons.  
In our study, streams with any level of recent human watershed activity (as agriculture, 
silviculture, urban development) had generally fewer species, more tolerant taxa, and 
overall lower biotic integrity scores than those streams draining primarily mixed forest.  
These impacts undoubtedly occur via several pathways and mechanisms, but LU/LC-
induced changes in stream physicochemical conditions appears to be paramount in 
influencing assemblage structure in these systems.  
4.5.1  Relative impacts of different LU/LC 
 Urbanization has been identified as one of the most pervasive and damaging types 
of landscape disturbance on aquatic systems (Paul and Meyer 2001).  Streams in our 
study within heavily urbanized watersheds had consistently lower biological integrity, 
and generally different macroinvertebrate assemblages altogether, than other watersheds, 
suggesting an alternative urban stable state or dynamic regime (Scheffer and Carpenter 
2003).  Urban sites were characterized by increased proportions of a suite of tolerant taxa 
(e.g. oligochaete worms, physid snails, Corbicula, hydropsychid caddisflies) and a 
concomitant low abundance of many otherwise common macroinvertebrates.  This 
pattern is likely a function of increased spate flow frequencies, chemical concentrations, 
 
 104
water temperature and decreased DO and benthic habitat observed in these streams, 
together considered part of the urban stream syndrome (Walsh et al. 2005b).   
Overall, % IS or Deciduous were the best LU/LC predictors of macroinvertebrate 
metrics.  IS cover in these watersheds spanned a broad range, yet most watersheds had 
<4% and were largely indistinguishable below this level.  In contrast, deciduous cover 
across the study watersheds was much more even (8 ? 38%) and, thus, could be 
considered the inverse of general human disturbance from silviculture, pasture, and/or 
urbanization.  Therefore, macroinvertebrate responses best described by % IS (i.e., 
biomass and NMDS
1
) ostensibly were driven by the small proportion of urban sites in our 
study, and likely best captured the degradation associated with this land use.  Therefore, 
the differences between urban and non-urban sites likely account for most of the variation 
in macroinvertebrate assemblage structure and biomass.  Richness, H?, and GA-BMI 
were best described inversely by % Deciduous, suggesting these metrics discerned 
variability associated with general, rather than urban, LU/LC disturbance.   Interestingly, 
GA-BMI scores were better predicted by % deciduous forest, but they responded non-
linearally to % IS. This pattern suggests a threshold of urban degradation ~10% 
watershed imperviousness, a level similar to what other studies have reported (Booth and 
Jackson 1997, Wang et al. 2000, Walsh et al. 2005b).   Although these 2 land cover 
classes were largely complementary in urban watersheds, the fact some metrics were best 
predicted by deciduous forest cover and some by IS cover suggests nuances in metric 
sensitivities to the various conditions associated with different land covers.   
Streams draining managed pine forests, agriculture, and residential development 
generally showed intermediate levels of impairment, with no discernable differences 
 
 105
among them in richness, H?, or biotic integrity.  It is generally considered that 
disturbance from timber harvest is less intense than that from agriculture (Fortino et al. 
2004).  It has also been suggested that although food web structure can change somewhat, 
overall stream biotic integrity may not be strongly affected by timber harvest in low-
gradient streams of the southeastern US Piedmont (Goodman et al. 2006) similar to those 
of our study.  Further, prior studies examining fish in these streams reveal similar 
assemblage structure and hydrology between the Urban and Developing watersheds 
(Helms et al. 2005, Schoonover et al. 2006).  It was therefore surprising that silviculture, 
pasture, and developing watersheds were so similar to each other in their 
macroinvertebrate assemblages.  However, the agriculture in these watersheds is 
primarily pasture, and the influence of pasture land use is often less than that of intensive 
cultivation (Strayer et al. 2003), which could explain unexpectedly high 
macroinvertebrate integrity in these watersheds.  In addition, although all silviculture 
watersheds had intact riparian zones and operations ostensibly followed BMP guidelines, 
timber harvesting in 2 of these watersheds was a recent activity (within 2 years prior to 
sampling), which, could explain the lower than expected response.  Last, although 
Developing and Urban streams have much flashier hydrographs than the Forest streams 
(Schoonover et al. 2006), the proportion of benthic detritus was similarly high in streams 
of Developing and Forest watersheds (as opposed to streams in Urban watersheds), 
emphasizing the important role of benthic habitat quality for macroinvertebrates (Roy et 
al. 2003).  
4.5.2 Macroinvertebrate seasonality 
 
 106
 Stream macroinvertebrate assemblages show dramatic seasonal variation in 
species presence/absence, richness, diversity, and calculated biotic indices (Minshall 
1981, Rosillon 1985, Linke et al. 1999).  There was no seasonal interaction with LU/LC 
in our study, although macroinvertebrate richness, H?, biomass, and density all were 
highest in spring than other seasons.  The GA-BMI was highest overall in summer, as 
were %Shredders and %Scrapers, 2 components of this index.  However, trends between 
LU/LC categories remain constant throughout the year and since there were no 
interactions between seasons and LU/LC category the metrics considered, these results 
suggest that these metrics are robust enough to use for relative comparative purposes 
year-round. 
4.5.3 Potential assemblage drivers  
 As evidenced by the NMDS ordination, the most diverse and highest integrity 
assemblages were largely associated with streams containing relatively cool water, low 
SO
4
?
 and TDS, and high DO, benthic detritus, and current velocity.  Further, multiple 
regressions suggest that water chemistry and, to a lesser degree, stream geomorphology, 
were the best predictors of macroinvertebrate assemblages.  Interestingly, stream 
hydrology did not appear in any of the multiple regression models nor were hydrological 
variables one of the stronger correlates of NMDS
1
; taken together, these results suggest 
that gross hydrologic changes associated with LU/LC alteration may not directly 
influence macroinvertebrates in these streams.  This result is counter to other studies that 
have shown strong relationships between altered hydrology and aquatic biota, particularly 
those involving fishes (Freeman et al. 2001, Roy et al. 2005).  As a result of their close 
association with the benthos, small size and short generation time, benthic 
 
 107
macroinvertebrates may be more susceptible to the indirect effects of altered hydrology 
(e.g. reduced water and habitat quality from increased runoff and spate frequency) rather 
than direct effects (e.g. individual or egg displacement) that are often implicated in larger, 
more nektonic organisms like fish (Freeman et al. 2001).   
As in our study streams, many studies have observed elevated TDS (or specific 
conductivity) with increased urban area / impervious surface (Zampella 1994, Dow and 
Zampella 2000) as well as with decreased biotic integrity (Walsh et al. 2001, Roy et al. 
2003).  However, TDS concentrations were not necessarily at biologically significant 
levels, as most aquatic systems with biota can withstand TDS levels up to 1000 mgL
-1
 
(Boyd 2000). Therefore, strong association of TDS and other dissolved solids, such as 
SO
4
?,
 within urbanized watersheds suggest that TDS is likely a surrogate for increased 
non-point pollution associated with efficient runoff, thus anthropogenic markers in these 
streams.     
Temperature and DO were 2 of the strongest correlates with macroinvertebrate 
assemblages.  Temperature is one of the most important environmental variables 
regulating aquatic biota, affecting metabolic rates and feeding requirements as well as 
growth and development (Ward and Stanford 1982).   Even small changes in temperature 
(1 ? 5
O
C) can alter hatching, emergence, growth, and recruitment of aquatic organisms 
(Beacham and Murray 1990, Sweeney 1993).  Associated with increasing water 
temperature is decreasing DO concentrations, and several studies have demonstrated 
elevated water temperature associated with increased biochemical oxygen demand 
(BOD) in streams, especially those draining urbanized watersheds (Sponseller et al. 2001, 
Walsh et al. 2001, Krause et al. 2004).  This pattern frequently results from increased 
 
 108
discharge, which may scour, widen and/or straighten channels, effectively causing 
decreased riparian shading and increased transport of thermally heated stormwater to the 
channel (Trimble 1997, Krause et al. 2004).  In our study, more diverse benthic 
assemblages occurred in streams within predominately forested watersheds, even those 
with high levels of residential development, which showed on average cooler, more 
oxygenated water.  In marked contrast, Oligochaetes and other tolerant organisms were 
strongly associated with streams containing higher temperature and lower DO.  Mean 
temperatures across the sites showed a range of ~2.5
O
C whereas mean DO showed a 
range of ~5.9 mgL
-1
.  These contrasting conditions collectively are well within the realm 
of biological significance as potential direct drivers of macroinvertebrate assemblages, 
possibly acting as life history constraints, influencing food resources, predators or all of 
these factors (Hynes 1970, Ward and Stanford 1982, Allan 1995). 
4.5.4 Conclusion and Implications 
 Use of a categorical approach to investigate land use and a correlative approach to 
investigate land cover allowed us to quantify and compare putative impacts of watershed 
LU/LC change on stream benthic assemblages; our analyses suggested that impacts are 
far-reaching and that some level of biological impairment is accompanied with all the 
common LU/LC disturbances in this area.  Intense urbanization appears to be the 
predominant land use influencing assemblages in these watersheds when considered 
together, and significantly more so than residential development, pasture, and silviculture 
land uses.  And, although the interrelated influences of watershed disturbance are difficult 
to separate, macroinvertebrate responses in these systems were most strongly associated 
 
 109
with alterations in water chemistry parameters and benthic habitat such that hydrologic 
alterations appeared to play an indirect role structuring benthic assemblages.   
By attempting to identify specific environmental drivers of biotic composition 
resulting from LU/LC perturbations operating at watershed scales, I can better address 
management and/or restoration needs designed to protect or minimize changes in stream 
biotic integrity.  Therefore, management and restoration efforts in disturbed systems, and 
particularly urban watersheds, should be directed at both the reach- and watershed-scale 
(Rabeni and Sowa 1996, Walsh et al. 2005a).  Improved riparian and instream conditions, 
such as BMP implementation, riparian reconstruction, and habitat augmentations, are 
essential to increasing biotic integrity in disturbed streams, as has been shown in other 
studies (Gore et al. 1998, Northington and Hershey 2006, Yates et al. 2007).  However, 
without addressing hydrological influences associated with human alterations it is 
unlikely that riparian or benthic habitat quality improvements will have significant long-
term effects, particularly in urban systems (Gore et al. 2001, Walsh et al. 2005a) .   
Reducing  overland flow and pollutant inputs in urbanized watersheds as a restoration 
approach has been shown to enhance instream restoration efforts (Charbonneau and Resh 
1992).  Thus, habitat and physicochemical conditions (ostensibly influenced by 
hydrological alterations) in disturbed watersheds will unlikely be influenced by reach-
scale restoration efforts alone. As such, implementation of biologically realistic 
management strategies in human-influenced systems may rely upon a multi-scale and 
hierarchical approach to assessment and abatement. 
 
 110
4.6 REFERENCES 
Allan, J. D. 1995. Stream ecology: structure and function of running waters. Chapman & 
Hall, New York. 
Allan, J. D. 2004. Landscapes and Riverscapes: The influence of land use on stream 
ecosystems. Annual Review of Ecology and Systematics 35:257-284. 
Barbour, M. T., J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. Rapid bioassessment 
protocols for use in streams and wadeable rivers: periphyton, benthic 
macroinvertebrates and fish, second edition. EPA 841-B-99-002, U.S. 
Environmental Protection Agency; Office of Water, Washington D.C. 
Beacham, T. D., and C. B. Murray. 1990. Temperature, egg size, and development of 
embryos and alevins of 5 species of Pacific Salmon - a comparative analysis. 
Transactions of the American Fisheries Society 119:927-945. 
Benke, A. C., A. D. Huryn, L. A. Smock, and J. B. Wallace. 1999. Length-mass 
relationships for freshwater macroinvertebrates in North America with particular 
reference to the southeastern United States. Journal of the North American 
Benthological Society 18:309-343. 
Bledsoe, B. P., and C. C. Watson. 2001. Effects of urbanization of channel instability. 
Journal of the American Water Resources Association 37:255-270. 
Booth, D. B., and C. R. Jackson. 1997. Urbanization of aquatic systems: degradation 
thresholds, stormwater detection, and the limits of mitigation. Journal of the 
American Water Resources Association 33:1077-1090. 
 
 111
Bormann, H., B. Diekkruger, and M. Hauschild. 1999. Impacts of landscape management 
on the hydrological behaviour of small agricultural catchments. Physics and 
Chemistry of the Earth Part B-Hydrology Oceans and Atmosphere 24:291-296. 
Boyd, C. E. 2000. Water Quality: An Introduction. Kluwer Academic Publishers, 
Norwell, Massachusetts, USA. 
Brown, D. G., K. M. Johnson, T. R. Loveland, and D. M. Theobald. 2005. Rural land-use 
trends in the conterminous United States, 1950-2000. Ecological Applications 
15:1851-1863. 
Campbell, I. C., and T. J. Doeg. 1989. Impact of timber harvesting and production on 
streams - a review. Australian Journal of Marine and Freshwater Research 
40:519-539. 
Chadwick, M. A., D. R. Dobberfuhl, A. C. Benke, A. D. Huryn, K. Suberkropp, and J. E. 
Thiele. 2006. Urbanization affects stream ecosystem function by altering 
hydrology, chemistry, and biotic richness. Ecological Applications 16:1796-1807. 
Charbonneau, R., and V. H. Resh. 1992. Strawberry Creek on the University-of-
California, Berkeley Campus - a case history of urban stream restoration. Aquatic 
Conservation-Marine and Freshwater Ecosystems 2:293-307. 
Clements, W. H., D. M. Carlisle, J. M. Lazorchak, and P. C. Johnson. 2000. Heavy 
metals structure benthic communities in Colorado mountain streams. Ecological 
Applications 10:626-638. 
Dow, C. L., and R. A. Zampella. 2000. Specific conductance and pH as indicators of 
watershed disturbance in streams of the New Jersey Pinelands, USA. 
Environmental Management 26:437-445. 
 
 112
Epler, J. H. 2001. Identification Manual for the Chironomidae (Diptera) of North and 
South Carolina. . North Carolina Department of Environment and Natural 
Resources, Division of Water Quality, Raleigh, North Carolina. 
Fortino, K., A. E. Hershey, and K. J. Goodman. 2004. Utility of biological monitoring for 
detection of timber harvest effects on streams and evaluation of Best Management 
Practices: a review. Journal of the North American Benthological Society 23:634-
646. 
Freeman, M. C., Z. H. Bowen, K. D. Bovee, and E. R. Irwin. 2001. Flow and habitat 
effects on juvenile fish abundance in natural and altered flow regimes. Ecological 
Applications 11:179-180. 
Georgia Department of Natural Resources. 2007. Macroinvertebrate Biological 
Assessment of Wadeable Streams in Georgia:  Standard Operating Procedures. 
Watershed Protection Branch, Environmental Protection Division, Georgia 
Department of Natural Resources, Atlanta, GA. 
Goodman, K. J., A. E. Hershey, and K. Fortino. 2006. The effect of forest type on benthic 
macroinvertebrate structure and ecological function in a pine plantation in the 
North Carolina Piedmont. Hydrobiologia 559:305-318. 
Gordon, N. D., T. A. McMahon, and B. L. Finlayson. 1992. Stream hydrology: An 
introduction for ecologists. John Wiley & Sons, New York. 
Gore, J. A., D. J. Crawford, and D. S. Addison. 1998. An analysis of artificial riffles and 
enhancement of benthic community diversity by physical habitat simulation 
(PHABSIM) and direct observation. Regulated Rivers-Research & Management 
14:69-77. 
 
 113
Gore, J. A., J. B. Layzer, and J. Mead. 2001. Macroinvertebrate instream flow studies 
after 20 years: a role in stream management and restoration. Regulated Rivers: 
Research and Management 17:527-542. 
Griffith, G. E., J. M. Omernik, J. A. Comstock, S. Lawrence, and T. Foster. 2001. Level 
III and IV Ecoregions of Georgia.  United States Geological Survey, Reston, VA. 
Gustard, A., A. Bullock, and J. M. Dixon. 1992. Low flow estimation in the United 
Kingdom. 108, Institute of Hydrology, Wallingford, U.K. 
Hammett, J. E. 1992. The shapes of adaptation - historical ecology of anthropogenic 
landscapes in the southeastern United-States. Landscape Ecology 7:121-135. 
Hawkins, C. P., J. N. Hogue, L. M. Decker, and J. W. Feminella. 1997. Channel 
morphology, water temperature, and assemblage structure of stream insects. 
Journal of the North American Benthological Society 16:728-749. 
Helms, B. S., J. W. Feminella, and S. Pan. 2005. Detection of biotic responses to 
urbanization using fish assemblages from small streams of western Georgia, USA. 
Urban Ecosystems 8:39-57. 
Hynes, H. B. N. 1970. The ecology of running waters. University of Toronto Press. 
Jones, R. C., and C. C. Clark. 1987. Impact of watershed urbanization on stream insect 
communities. Water Resouces Bulletin 23:1047-1056. 
Krause, C. W., B. Lockard, T. J. Newcomb, D. Kibler, V. Lohani, and D. J. Orth. 2004. 
Predicting influences of urban development on thermal habitat in a warm water 
stream. Journal of the American Water Resources Association 40:1645-1658. 
Lenat, D. R., and J. K. Crawford. 1994. Effects of land use on water quality and aquatic 
biota of three North Carolina Piedmont streams. Hydrobiologia 294:185-199. 
 
 114
Linke, S., R. C. Bailey, and J. Schwindt. 1999. Temporal variability of stream 
bioassessments using benthic macroinvertebrates. Freshwater Biology 42:575-
584. 
Lockaby, B. G., D. Zhang, J. McDaniel, H. Tian, and S. Pan. 2005. Interdisciplinary 
research at the urban-rural interface: The WestGa project. Urban Ecosystems 8:7-
21. 
Maloney, K. O., and J. W. Feminella. 2006. Evaluation of single- and multi-metric 
benthic macroinvertebrate indicators of catchment disturbance over time at the 
Fort Benning Military Installation, Georgia, USA. Ecological Indicators 6:469-
484. 
Maloney, K. O., P. J. Mulholland, and J. W. Feminella. 2005. Influence of catchment-
scale military land use on physical and organic matter conditions in small 
Southeastern Plains streams (USA). Environmental Management 35:677-691. 
McCune, B., and J. B. Grace. 2002. Analysis of ecological communities. MjM Software 
Design, Gleneden Beach, Oregon. 
McMahon, G., J. D. Bales, J. Coles, E. M. P. Giddings, and H. Zappia. 2003. Use of stage 
data to characterize hydrologic conditions in an urbanizing environment. Journal 
of the American Water Resources Association 39:1529-1546. 
Merritt, R. W., and K. W. Cummins. 1996. An Introduction to the Aquatic Insects of 
North America. Kendall/Hunt Publishing Company Dubuque, IA. 
Meyer, J. L., M. J. Paul, and W. K. Taulbee. 2005. Stream ecosystem function in 
urbanizing landscapes. Journal of the North American Benthological Society 
24:602-612. 
 
 115
Minshall, G. W. 1981. Structure and Temporal Variations of the Benthic 
Macroinvertebrate Community Inhabiting Mink Creek, Idaho, USA, a 3rd Order 
Rocky-Mountain Stream. Journal of Freshwater Ecology 1:13-26. 
Myeong, S., D. J. Nowak, P. F. Hopkins, and R. H. Brock. 2001. Urban cover mapping 
using digital, high-spatial resolution aerial imagery. Urban Ecosystems 5:243 - 
256. 
Myers, R. H. 1990. Classical and modern regression with applications, Second edition. 
Duxbury Press, Belmont, California. 
Newbury, R. W. 1996. Dynamics of flow. Pages 75-92 in F. R. Hauer and G. A. 
Lamberti, editors. Methods in stream ecology. Academic Press, New York. 
Northington, R. M., and A. E. Hershey. 2006. Effects of stream restoration and 
wastewater treatment plant effluent on fish communities in urban streams. 
Freshwater Biology 51:1959-1973. 
Paul, M. J., and J. L. Meyer. 2001. Streams in the urban landscape. Annual Review of 
Ecology and Systematics 32:333-365. 
Peterson, J. T., and T. J. Kwak. 1999. Modeling the effects of land use and climate 
change on riverine smallmouth bass. Ecological Applications 9:1391-1404. 
Poff, N. L., J. D. Allan, M. B. Bain, J. R. Karr, K. L. Prestegaard, B. D. Richter, R. E. 
Sparks, and J. C. Stromberg. 1997. The natural flow regime: a paradigm for river 
conservation and restoration. BioScience 47:769-784. 
Prud'homme, B. A., and J. G. Greis. 2002. Best management practices in the South. in D. 
N. Wear and J. G. Greis, editors. Southern forest resource assessment.  General 
 
 116
Technical Report, Southern Research Station, Forest Service, US Department of 
Agriculture, Asheville, North Carolina. 
Rabeni, C. F., and S. P. Sowa. 1996. Integrating biological realism into habitat restoration 
and conservation strategies for small streams. Canadian Journal of Fisheries and 
Aquatic Sciences 53:252-259. 
Rapport, D. J., H. A. Regier, and T. C. Hutchinson. 1985. Ecosystem behavior under 
stress. American Naturalist 125:617-640. 
Richter, B. D., J. V. Baumgartner, J. Powell, and D. P. Braun. 1996. A method for 
assessing hydrologic alteration within ecosystems. Conservation Biology 
10:1163-1174. 
Riis, T., and K. Sand-Jensen. 2001. Historical changes in species composition and 
richness accompanying perturbation and eutrophication of Danish lowland 
streams over 100 years. Freshwater Biology 46:269-280. 
Rosillon, D. 1985. Seasonal-variations in the benthos of a chalk trout stream, the River 
Samson, Belgium. Hydrobiologia 126:253-262. 
Roy, A. H., M. C. Freeman, B. J. Freeman, S. J. Wenger, W. E. Ensign, and J. L. Meyer. 
2005. Investigating hydrologic alteration as a mechanism of fish assemblage 
shifts in urbanizing streams. Journal of the North American Benthological Society 
24:656-678. 
Roy, A. H., A. D. Rosemond, M. J. Paul, D. S. Leigh, and J. B. Wallace. 2003. Stream 
macroinvertebrate response to catchment urbanisation (Georgia, U.S.A.). 
Freshwater Biology 48:329-346. 
 
 117
Scheffer, M., and S. R. Carpenter. 2003. Catastrophic regime shifts in ecosystems: 
linking theory to observation. Trends in Ecology and Evolution 18:648-656. 
Schnaiberg, J., J. Riera, M. G. Turner, and P. R. Voss. 2002. Explaining human 
settlement patterns in a recreational lake district: Vilas County, Wisconsin, USA. 
Environmental Management 30:24-34. 
Schoonover, J. E., and B. G. Lockaby. 2006. Land cover impacts on stream nutrients and 
fecal coliform in the lower Piedmont of West Georgia. Journal of Hydrology 
331:371-382. 
Schoonover, J. E., B. G. Lockaby, and B. S. Helms. 2006. Impacts of land cover on 
stream hydrology in the west Georgia piedmont, USA. Journal of Environmental 
Quality 35:2123-2131. 
Sponseller, R. A., E. F. Benfield, and H. M. Valett. 2001. Relationships between land 
use, spatial scale and stream macroinvertebrate communities. Freshwater Biology 
46:1409-1424. 
Strayer, D. L., R. E. Beighley, L. C. Thompson, S. Brooks, C. Nilsson, G. Pinay, and R. 
J. Naiman. 2003. Effects of land cover on stream ecosystems: roles of empirical 
models and scaling issues. Ecosystems 6:407 - 423. 
Sweeney, B. W. 1993. Effects of streamside vegetation on macroinvertebrate 
communities of White Clay Creek in Eastern North America. Proceedings of the 
Academy of Natural Sciences of Philadelphia 144:291-340. 
Trimble, S. W. 1997. Contribution of stream channel erosion to sediment yield from an 
urbanizing watershed. Science 278:1442-1444. 
 
 118
Vinson, M. R., and C. P. Hawkins. 1996. Effects of sampling area and subsampling 
procedure on comparisons of taxa richness among streams. Journal of the North 
American Benthological Society 15:392-399. 
Walsh, C. J., T. D. Fletcher, and A. R. Ladson. 2005a. Stream restoration in urban 
catchments through redesigning stormwater systems: looking to the catchment to 
save the stream. Journal of the North American Benthological Society 24:690-
705. 
Walsh, C. J., A. H. Roy, J. W. Feminella, P. D. Cottingham, P. M. Groffman, and R. P. 
Morgan. 2005b. The urban stream syndrome: current knowledge and the search 
for a cure. Journal of the North American Benthological Society 24:706-723. 
Walsh, C. J., A. K. Sharpe, P. F. Breen, and J. A. Sonneman. 2001. Effects of 
urbanization on streams of the Melbourne region, Victoria, Australia. I. Benthic 
macroinvertebrate communities. Freshwater Biology 46:535-551. 
Walters, D. M., D. S. Leigh, and A. B. Bearden. 2003. Urbanization, sedimentation, and 
the homogenization of fish assemblages in the Etowah River Basin, USA. 
Hydrobiologia 494:5-10. 
Wang, L., J. Lyons, P. Kanehl, R. Bannerman, and E. Emmons. 2000. Watershed 
urbanization and changes in fish communities in southeastern Wisconsin streams. 
Journal of the American Water Resources Association 36:1173-1189. 
Wang, L. Z., J. Lyons, and P. Kanehl. 2002. Effects of watershed best management 
practices on habitat and fish in Wisconsin streams. Journal of the American Water 
Resources Association 38:663-680. 
 
 119
Ward, J. V., and J. A. Stanford. 1982. Thermal responses in the evolutionary ecology of 
aquatic insects. Annual Review of Entomology 27:97-117. 
Weaver, L. A., and G. C. Garman. 1994. Urbanization of a watershed and historical 
changes in a stream fish assemblage. Transactions of the American Fisheries 
Society 123:162-172. 
Wiggins, G. B. 1996. Larvae of the North American caddisfly genera (Trichoptera). 
University of Toronto Press, Buffalo, NY. 
Yates, A. G., R. C. Bailey, and J. A. Schwindt. 2007. Effectiveness of best management 
practices in improving stream ecosystem quality. Hydrobiologia 583:331-344. 
Zampella, R. A. 1994. Characterization of surface water quality along a watershed 
disturbance gradient. Water Resources Bulletin 30:605-611. 
Zar, J. H. 1999. Biostatistical analysis. 4
th
 Edition.  Prentice-Hall, Upper Saddle River, 
New Jersey. 
 
  T
a
bl
e 1.
  L
a
n
d
 co
v
e
r
 an
d s
i
ze o
f
 s
t
ud
y
 wat
e
r
s
h
e
d
s
.
  
IS
 =
 %
 i
m
p
e
r
v
i
o
u
s
 s
u
r
f
ace co
v
e
r
,
 
Gr
as
s
/
T
r
an
s
i
s
t
ion 
= 
%
 pas
t
ur
e c
o
v
e
r
,
 
Decid
u
o
u
s
 
= %
 
dec
i
duo
us
 f
o
r
e
s
t
 
c
o
v
e
r
,
 
Ever
green
 =  %
 ev
er
gr
een f
o
r
e
s
t
 c
o
v
e
r
,
 an
d 
LU/
L
C
 =
 do
min
a
n
t
 
l
a
nd
 u
s
e
/
la
n
d
 c
o
ve
r
 in w
a
t
e
r
s
he
d
.
 
 
ID Si
t
e
 
St
r
e
a
m
 
W
a
t
e
r
she
d
 si
z
e
 
(k
m
2
)
 IS 
G
r
a
ss/
T
r
a
n
si
t
i
o
n
 
De
c
i
du
o
u
s 
E
v
e
r
g
r
e
e
n
 
L
U
/
L
C 
1 S
B
1 
S
c
h
l
ey
 C
r
eek
 
2
0
.
1
 
1.
88 20.
38 
35.
14 
38.
35 
D
e
ve
l
o
pi
n
g
 
2 S
B
2 
S
t
an
d
i
n
g
 B
o
y
 C
r
eek
 Tr
i
b
.
 
6
.
3
 
3.
41 20.
00 
35.
52 
37.
04 
D
e
ve
l
o
pi
n
g
 
3 S
B
4 
S
t
an
d
i
n
g
 B
o
y
 C
r
eek
 
2
6
.
6
 
3.
33 27.
70 
22.
81 
40.
98 
D
e
ve
l
o
pi
n
g
 
4 H
C
 
H
o
us
e
 Cr
e
e
k
 T
r
i
b
.
 
6.
6 
1.
33 19.
57 
26.
76 
47.
78 
S
i
l
v
i
c
ul
t
u
r
e
 
5 M
U
2 
M
u
l
b
er
r
y
 C
r
eek
 Tr
i
b
.
 
6
.
1
 
1.
39 14.
86 
37.
20 
41.
33 
S
i
l
v
i
c
ul
t
u
r
e
 
6 S
C
 
S
a
n
d
 C
r
e
e
k
 
9.
0 
1.
24 20.
89 
28.
86 
44.
68 
S
i
l
v
i
c
ul
t
u
r
e
 
7 
BC
 
B
e
ech
 C
r
eek
 
6
.
5
 
2.
29 13.
26 
34.
14 
46.
53 
F
o
r
e
s
t
 
8 B
L
N
 
B
l
a
n
t
o
n C
r
e
e
k
 
3.
6 
1.
41 18.
65 
28.
30 
47.
85 
Fo
r
e
st
 
9 M
K
 
F
l
at
 C
r
e
e
k
 T
r
ib
. 
6
.
6
 
2.
28 19.
82 
37.
98 
36.
31 
Fo
r
e
st
 
10 M
O
 
C
l
in
e'
s
 B
r
an
c
h
 
9
.
0
 
1.
53 13.
04 
33.
07 
48.
19 
Fo
r
e
st
 
11 M
U
3 
T
u
rn
t
i
m
e
 B
r
a
n
c
h
 
10.
4 
1.
39 36.
77 
24.
24 
29.
24 
Fo
r
e
st
 
12 F
S
2
 
Wi
l
d
ca
t C
r
eek
 Tr
i
b
.
 
1
4
.
5
 
2.
74 35.
79 
28.
21 
30.
71 
P
a
s
t
u
r
e
 
13 F
S
3
 
Wi
l
d
ca
t C
r
eek
 Tr
i
b
.
 
3
.
0
 
2.
58 33.
91 
29.
91 
31.
96 
P
a
s
t
u
r
e
 
14 H
C
2 
H
o
us
e
 Cr
e
e
k
 
14.
1 
1.
64 43.
95 
22.
22 
30.
46 
P
a
s
t
u
r
e
 
15 M
U
1 
O
s
s
a
ha
t
c
h
i
e
 C
r
e
e
k
 
T
r
i
b
.
 
12.
0 
1.
89 14.
86 
37.
20 
41.
33 
P
a
s
t
u
r
e
 
16 BU
1 
L
i
n
d
s
e
y
 Cr
e
e
k
 
25.
5 
40.
40 23.
31 
12.
66 
21.
37 
U
r
b
a
n
 
17 BU
2 
C
o
o
p
er
 C
r
eek
 
2
4
.
7
 
25.
06 24.
90 
15.
87 
30.
43 
U
r
b
a
n
 
18 RB
 
Ro
a
r
i
ng B
r
a
n
c
h
 
3.
7 
30.
30 27.
11 
11.
07 
28.
36 
U
r
b
a
n
 
19 BR
 
Br
ook
s
t
on
e
 Cr
e
e
k
 
4
.
7
 
19.
15 33.
84 
15.
66 
31.
07 
U
r
b
a
n
 
20 F
R
 
F
l
a
t
 R
o
ck
 C
r
eek
 
2
3
.
9
 
13.
70 43.
37 
8.
05 
32.
72 
U
r
b
a
n
 
 
120 
  T
a
bl
e 2.
  L
i
s
t
 
o
f
 hy
dr
o
l
o
g
i
ca
l
 
var
i
a
b
l
e
s
 us
ed 
i
n
 a
n
a
l
y
s
es
,
 t
h
e
i
r
 r
a
n
g
e o
f
 
v
a
l
u
es
,
 a
n
d s
i
g
n
if
i
c
a
n
t
 Pear
s
o
n
 co
r
r
e
l
a
t
i
o
n
s
 t
o
 per
c
e
n
t
age 
o
f
 wat
e
r
s
h
e
d as
 
im
per
v
i
o
us
 (
% I
S
)
,
 pas
t
ur
e
 (
%
 
P
)
,
 ev
er
gr
een
 
f
o
r
e
s
t
 (
%
 
EF
)
 an
d dec
i
duo
us
 
f
o
r
e
s
t
 
(
%
 
DF
)
.
  *  p <
 0.
05
,
 *
*
 p
 <
 0.
0
1
.
 
  
V
a
ri
a
b
le
 I
D
 
D
e
s
c
ri
p
t
i
o
n
 
R
a
n
g
e
  
% I
S
 
% P
 
% E
F
 
% D
F
 
M
agnitude 
  
  
 
 
Me
d
Q
 
Me
d
i
a
n
 d
i
sc
h
a
r
g
e
 (
L
 s
-1
)
 
0.
01 ?
 0.
90 
 0.
49* 
 
 
Ma
x
Q
 
Ma
x
i
m
u
m
 d
i
sc
h
a
r
g
e
 (
L
 s
-1
)
 
0.
54 ?
 21.
98 
  
 
 
Fr
equ
e
n
cy (
#
 t
i
m
es
 exceeded t
h
r
e
s
h
ol
d)
 
 
 
   
3x
M
e
d 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 3
x
 
m
e
d
i
a
n
 
f
l
o
w
 
5 ? 116 
  
 
 
5x
M
e
d 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 5
x
 
m
e
d
i
a
n
 
f
l
o
w
 
1 ? 70 
0.
50*  
 
 
7x
M
e
d 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 7
x
 
m
e
d
i
a
n
 
f
l
o
w
 
1 ? 64 
0.
56*  
 
 
9x
M
e
d
 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 9
x
 
m
e
d
i
a
n
 
f
l
o
w
 
0 ? 58 
0.
58*  
 
 
>75
th
 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 75t
h
 per
c
e
n
t
i
l
e
 
25 ? 115 
  
 
 
>95
th
 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 95t
h
 per
c
e
n
t
i
l
e
 
12 ? 66 
  
 
 
>99
th
 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 99t
h
 per
c
e
n
t
i
l
e
 
2 ? 35 
0.
62**  
 
 
D
u
ra
t
i
o
n
 (
#
 o
f
 h
o
u
rs sp
e
n
t
 a
b
o
v
e
 t
h
re
sh
o
l
d
)
 
 
 
   
>3xM
ed_d 
Ho
ur
s
 di
s
c
h
a
r
g
e was
 >3
x
 
m
e
d
i
a
n
 
f
l
o
w
 
36.
5 ?
 3026 
  
 
 
>5xM
ed_d 
Ho
ur
s
 di
s
c
h
a
r
g
e was
 >5
x
 
m
e
d
i
a
n
 
f
l
o
w
 
6 ? 2518 
  
 
 
>7xM
ed_d 
Ho
ur
s
 di
s
c
h
a
r
g
e was
 >7
x
 
m
e
d
i
a
n
 
f
l
o
w
 
1.
5 ?
 2412 
  
 
 
>9xM
ed_d 
Ho
ur
s
 di
s
c
h
a
r
g
e was
 >9
x
 
m
e
d
i
a
n
 
f
l
o
w
 
0 ? 17 
  
 
 
P
r
e
d
ic
tability
 and F
l
as
hine
s
s
 
 
  
 
 
 
 
C
.
V.
 
%
 C
o
ef
f
i
c
i
e
n
t
 o
f
 Var
i
at
i
o
n
 
42 ? 402 
  
 
 
I
n
c1
h
100 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 100%
 
i
n
 1 h
r
 
7 ? 109 
0.
55*  
 
 
I
n
c1
h
1000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 1000%
 
i
n
 1 h
r
 
0 ? 31 
  
 
 
I
n
c1
h
5000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 5000%
 
i
n
 1 h
r
 
0 ? 17 
 -
 
0.
47* 
 
 
I
n
c3
h
100 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 100%
 
i
n
 3 h
r
 
11 ? 122 
0.
53*  
 
 
121 
  
I
n
c3
h
1000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 1000%
 
i
n
 3 h
r
 
0 ? 44 
  
 
 
I
n
c3
h
5000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e 
i
n
cr
eas
es
 
by
 5000%
 
i
n
 3 h
r
 
0 ? 19 
  
 
 
Dec1h
100 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 100%
 
i
n
 1 h
r
 
0 ? 67 
  
 
 
Dec1h
1000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 1000%
 
i
n
 1 h
r
 
0 ? 28 
  
 
 
Dec1h
5000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 5000%
 
i
n
 1 h
r
 
0 ? 12 
  
 
 
Dec3h
100 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
e
s
 
by
 100%
 
i
n
 3 h
r
 
1 ? 92 
  
 
 
Dec3h
1000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 1000%
 
i
n
 3 h
r
 
0 ? 36 
  
 
 
Dec3h
5000 
# 
o
f
 e
v
e
n
t
s
 d
i
s
c
har
g
e decr
eas
es
 
by
 5000%
 
i
n
 3 h
r
 
0 ? 18 
  
 
 
B
a
s
e
f
l
ow
  
 
 
   
Me
d
B
F
 
Me
d
i
a
n
 b
a
se
f
l
o
w
  (
L
 s
-1
)
 
0.
04 ?
 730 
  
 
 
BI
 
B
a
se
f
l
ow
 inde
x (
?
 pr
e
d
ic
t
e
d
 ba
se
f
l
ow
/ 
?
o
b
s
e
r
v
ed
 fl
o
w
) 
0.
03 ?
 0.
82 
  
 
 
 
  
  
 
 
 
122 
  T
a
bl
e 3.
  
 P
h
y
s
i
c
o
c
h
e
m
i
ca
l
 
v
a
r
i
a
b
l
e
s
 us
ed 
i
n
 a
n
a
l
ys
es
,
 t
h
e
i
r
 r
a
n
g
e o
f
 
v
a
l
u
e
s
,
 an
d s
i
g
n
i
f
i
c
a
n
t
 P
e
ar
s
o
n
 co
r
r
el
at
i
o
ns
 t
o
 pe
r
c
en
t
a
g
e o
f
 wat
e
r
s
h
e
d as
 
im
per
v
i
o
us
 (
% I
S
)
,
 pas
t
ur
e
 (
%
 
P
)
,
 ev
er
gr
een
 
f
o
r
e
s
t
 (
%
 
EF
)
 an
d dec
i
duo
us
 
f
o
r
e
s
t
 
(
%
 
DF
)
.
  *  p <
 0.
05
,
 *
*
 p
 <
 0.
0
1
.
  
          
  
 
  
 
 
V
a
ri
a
b
le
 
D
e
s
c
ri
p
t
i
o
n
 
R
a
n
g
e
  
% I
S
 
% P
 
% E
F
 
% D
F
 
T
e
m
p
 
M
e
a
n
 wat
e
r
 
t
e
m
p
er
at
ur
e (
o
C
)
 
13.
3 ?
 15.
8 
0.
83***  
 
-
 
0.
66** 
D
O
 
Me
a
n
 d
i
sso
l
v
e
d
 O
2
 (
m
g 
L
-1
)
 
8.
6 ?
 14.
5 
-
 
0.
50*  
 0.
68** 
pH 
M
e
a
n
 pH 
5.
7 ?
 6.
9
  
  
 
 
T
S
S
 
Me
a
n
 t
o
t
a
l
 su
sp
e
n
d
e
d
 so
l
i
d
s c
o
n
c
e
n
t
r
a
t
i
o
n
 (
m
g
 L
-1
)
 
2.
1 ?
 8.
1 
 0.
54* 
 
 
T
D
S
 
Me
a
n
 t
o
t
a
l
 d
i
sso
l
v
e
d
 so
l
i
d
s c
o
n
c
e
n
t
r
a
t
i
o
n
 (
m
g
 L
-1
)
 
19.
95 ?
 58.
63 
0.
58  
 
 
NO
3
?
 
M
e
a
n
 
m
o
n
t
hl
y
 
ni
t
r
at
e co
n
cen
t
r
at
i
o
n
 (
m
g L
-1
)
 
  0
.
14 ?
 4.
45 
 0.
75*** 
-
 
0.
54* 
 
SO
4
?
 
M
e
a
n
 
m
o
nt
h
l
y
 
s
u
lp
ha
t
e
 c
o
nc
e
n
t
r
a
t
i
o
n (
m
g
 L
-1
)
 
0.
53 ?
 9.
19 
0.
82***  
 
-
 
0.
48* 
NH
4
+
 
M
e
a
n
 
m
o
n
t
hly
 
a
m
m
o
ni
u
m
 co
n
c
e
n
t
r
at
i
o
n
 (
m
g L
-1
)
 
0 ? 0
.
33 
  
 
-
 
0.
69** 
P 
M
e
a
n
 tot
a
l
 p
h
o
s
ph
at
e (
m
g L
-1
)
  
0.
05 ?
 0.
12 
  
 
 
D
O
C
 
M
e
a
n
 
m
o
nt
h
l
y
 d
i
s
s
o
l
ve
d
 o
r
g
a
n
i
c
 c
a
r
bo
n c
o
nc
e
n
t
r
a
t
i
o
n (
m
g
 L
-1
)
 
1.
59 ?
 8.
55 
  
 
 
F
C
 
M
e
a
n
 
m
o
n
t
hly
 
f
e
c
a
l
 co
l
i
f
o
r
m
 co
l
o
ni
es
 (
#
 100
m
l
-1
)
 
  
  127
 ? 15847 
0.
87*** 
 
-
 0.
54* 
-
 0.
 54* 
123 
  T
a
bl
e 4.
  Ge
o
m
o
r
phi
c v
a
r
i
a
b
l
e
s
 us
ed 
i
n
 a
n
a
l
y
s
es
,
 t
h
e
i
r
 r
a
n
g
e o
f
 
v
a
l
u
e
s
,
 an
d s
i
g
n
i
f
i
c
a
n
t
 P
ear
s
o
n
 cor
r
e
l
a
t
i
o
n
s
 t
o
 per
cen
t
a
ge 
o
f
 w
at
er
s
h
ed as
 
im
per
v
i
o
us
 (
% I
S
)
,
 pas
t
ur
e
 (
%
 
P
)
,
 ev
er
gr
een
 
f
o
r
e
s
t
 (
%
 
EF
)
 an
d dec
i
duo
us
 
f
o
r
e
s
t
 
(
%
 
DF
)
.
  *
 p <
 0.
0
5
,
 **
 p < 0
.
01.
 
   
V
a
ri
a
b
le
 
D
e
s
c
ri
p
t
i
o
n
 
R
a
n
g
e
 
% I
S
 
% P
 
% E
F
 
% D
F
 
Fl
o
w
 
M
e
a
n
 c
u
rre
n
t
 v
e
l
o
c
i
t
y
 a
t
 i
n
v
e
rt
e
b
ra
t
e
 s
a
m
p
l
e
 (
m
 s
-1
)
 
0.
07 ?
 0.
38 
  
 
 
OM
 
M
e
a
n
 be
n
t
hi
c o
r
gani
c 
m
a
t
t
e
r
 
(
g
)
 
  0
.
3 ?
 1.
2 
  
0.
58* 
 
S
u
bs
t
r
a
t
e 
M
e
d
i
a
n
 s
u
bs
t
r
a
t
e s
i
z
e
 (
c
m
)
 
  0.
5
 ? 1
.
8
 
0.
52*  
 
 
T
g
 
T
r
ac
t
i
ve 
f
o
r
ce (
N
 m
-2
)
 
  2
.
7? 73
.
9
 
  
 
 
C
l
a
y
 
Pr
o
p
o
r
t
i
o
n
 be
n
t
hi
c c
l
a
y
 
0 ? 0
.
07 
  
 
 
S
i
l
t
 
Pr
o
p
o
r
t
i
o
n
 be
n
t
hi
c s
i
l
t
 
0 ? 0
.
18 
  
 
 
San
d
 
Pr
o
p
o
r
t
i
o
n
 be
n
t
hi
c s
a
n
d
 
0.
20 ?
 1.
00 
-
 0.
44* 
 
 
 
Gr
av
e
l
 
Pr
o
p
o
r
t
i
o
n
 be
n
t
hi
c gr
av
e
l
 
0  ?
 0.
52 
0.
44 -
 
0.
48* 
 
 
C
o
b
b
l
e
 
Pr
o
p
o
r
t
i
o
n
 be
n
t
hi
c co
b
b
l
e
 
0 ? 0
.
43  
  
 
 
W
ood
 
P
r
op
or
ti
on
 b
e
n
t
h
i
c
 w
o
od
 
0
 ?
 0
.
2
2
 
-
 0.
56* 
 
 
 
Det
r
i
t
us
 
Pr
o
p
o
r
t
i
o
n
 be
n
t
hi
c det
r
i
t
us
 
0 ? 0
.
17 
-
 0.
55* 
 
 
0.
67** 
B
e
dr
o
c
k 
Pr
o
p
o
r
t
i
o
n
 be
n
t
hi
c b
e
dr
o
c
k 
0 ? 0
.
22 
0.
45*  
-
 
0.
44* 
 
   
124 
 
 125
Table 5.  Mean seasonal and annual values for macroinvertebrate variables and their 
ranges over the study.  Seasonal values for each stream were used for Range values and 
to acquire averages for Summer, Winter, and Spring values and total stream averages 
were used for Annual values.   
 
 
?
 metrics used for Georgia Biotic Macroinvertebrate Index 
 
 
 
 
 
 
 
Variable Summer Winter Spring Annual Range 
Composite    
 
 
GA-IBI 49.4 30.2 30.0 
36.3 
3 ? 75  
Community    
 
 
H? 2.1 1.9 2.1 
2.0 
0.8 ? 2.9  
Richness 34.1 33.3 38.8 
35.5 
14 ? 64  
Biomass (g / m
2
) 16.5 26.9 43.5 
32.9 
6.9 ? 101.6  
Density (no. / m
2
) 12510 21307 19998 
15119 
2504 - 68429 
Functional    
 
 
 %Shredders
?
 38.4 0 0 
12.3 
0 ? 87.6 
%Scrapers
?
 3.5 0.9 1.2 
1.9 
0 ? 12.3  
%Clingers 4.7 0.5 1.8 
2.3 
0 ? 15.3 
%Filterers 25.5 10.8 19.8 
18.7 
0.7 ? 64.1 
%Gatherers 17.5 20.6 18.0 
18.7 
0.3 ? 41.6 
%Predators 10.3 11.0 10.7 
10.7 
0.3 ? 41.8 
Swimmers
?
 1.4 0.9 1.7 
1.3 
0 ? 3 
%Tolerant 3.1 1.8 10.0 
5.1 
0 ? 48.6 
Taxonomic    
 
 
EPT 6.9 7.6 7.3 
7.3 
1 ? 17  
%Ephemeroptera 9.0 1.6 2.8 
4.4 
0 ? 27.8 
%Plecoptera
?
 0.3 8.9 1.5 
3.5 
0 ? 32.8 
%Trichoptera 15.6 1.0 7.2 
7.9 
0 ? 55.8 
Diptera 18.2 18.2 21.1 
19.2 
7 ? 31  
%Chironomidae 65.1 70.0 72.9 
69.5 
22.9 ? 98.8 
Coleoptera
?
 2.2 70.0 72.9 
69.5 
22.9 ? 98.8 
%Oligochaete
?
 2.1 8.8 10.0 
7.1 
0 ? 67.6 
   
 
 
 
 
 
 126
 
 
 
 
Table 6.  Results from General Linear Model testing the effects of land use/land cover 
(LU/LC) and season on macroinvertebrate metrics across the study watersheds.  For 
LU/LC, F
4, 55
, for Season, F
2, 55
, and for LU/LC-season interactions (Interaction), F
6, 55
. 
GA-BMI = Georgia Benthic Macroinvertebrate Index  
 
 
     
Metric Source F p 
   
Richness LU/LC 24.96 < 0.001 
 Season 6.21 0.004 
 Interaction 0.25 0.978 
   
H? LU/LC 5.89 0.001 
 Season 3.47 0.04 
 Interaction 0.42 0.902 
   
Density LU/LC 2.01 0.11 
 Season 4.2 0.022 
 Interaction 2.64 0.02 
   
Biomass LU/LC 3.96 0.011 
 Season 5.33 0.006 
 Interaction 1.06 0.497 
   
GA-IBI LU/LC 18.12 < 0.001 
 Season 27.93 < 0.001 
  Interaction 0.61 0.763 
   
 
 
 
 
 
 
 
 
 
 
 
 
 
 127
 
 
Table 7.  Best models for Composite and Community macroinvertebrate responses.  
LU/LC column shows the best single land use / land cover predictor and Multivariate 
column shows the best model from multiple regressions with PCA axes as predictors, 
and.  For the Multivariate column, C-PC
1
 and C-PC
2
 are physicochemical PCA axes 1 
and 2, respectively, and S-PC
1
 and S-PC
2
 are substrate PCA axes 1 and 2, respectively 
while LU/LC classes are as defined in Table 1.  GA-BMI = Georgia Benthic 
Macroinvertebrate Index and NMDS
1
 = axis 1 from NMDS ordination .  Models are based 
on mean annual data.  All models are of the form y = ax + b. *p < 0.05, **p < 0.01, ***p 
< 0.001. 
 
 
 
      
Metric LU/LC R
2
 Multivariate
Standardized 
Estimate R
2
adj 
    
Richness Deciduous 0.73*** C-PC
1 
0.77 0.80***
   C-PC
2 
-0.48  
      
H' Deciduous 0.42** C-PC
1
 0.67 0.78***
   S-PC
1 
-0.34  
 C-PC
2
 -0.62  
     
Density ns ? S-PC
1
 -0.42 0.40***
   C-PC
2
 -0.56  
   
Biomass IS 0.39** S-PC
2
 -0.98 0.25* 
  C-PC
1
 -0.83  
   
GA-
BMI Deciduous 0.67*** C-PC
1
 0.74 0.78***
   C-PC
2
 -0.52  
   
NMDS
1 
IS 0.77*** C-PC
1
 0.91 0.83***
 
 
    
 
 
 
 
 
 
 
 
 
 
 
 128
Table 8.  Loadings (eigenvalues) and associated axis correlations for 1
st
 and 2
nd
 axes of 
Hydrologic, Geomorphic, and Chemical PCA ordinations.  Variable descriptions are as 
in Tables 1 ? 3. *p < 0.05, **p < 0.01, ***p < 0.001.  
          
 Loadings Correlations 
  PC
1
 PC
2
 PC
1
 PC
2
 
Hydrologic     
MedQ -0.17 -0.01  -0.70**  
MaxQ 0.05 -0.23   -0.47* 
3xMed 0.14 -0.32   -0.70** 
5xMed 0.21 -0.24 0.67**  -0.66** 
7xMed 0.22 -0.22 0.66**  -0.63** 
Inc1h1000 0.23 0.13 0.72**  
Inc3h1000 0.24 0.11 0.78***  
Inc3h5000 0.24 0.16 0.63**  
Dec1h100 0.24 0.15 0.72**  
Dec3h100 0.22 0.05 0.73**  
MedBF -0.22 0.07  -0.72**  
BI -0.19 0.22  -0.67** 0.56* 
Habitat     
OM -0.16 -0.36   -0.57* 
Substrate -0.22 0.42  -0.47** 0.72** 
Sand 0.47 0.11 0.93***  
Gravel -0.3 0.26  -0.58*  
Cobble -0.44 -0.12  -0.86***  
Wood 0.44 -0.24 0.86***  
Detritus 0.02 -0.5   -0.80*** 
Bedrock -0.41 -0.03  -0.80***  
Chemical     
Temp -0.37 0.11  -0.77***  
DO 0.38 -0.02 0.86***  
pH -0.24 -0.10  -0.54*  
TSS -0.06 0.48  0.75*** 
TDS -0.41 -0.17  -0.94***  
NO
3
0.02 0.57   -0.87*** 
SO
4
-0.38 -0.21  -0.86***  
NH4 -0.12 0.49  -0.49* 0.59* 
DOC -0.38 -0.19  -0.79***  
FC -0.30 0.26 -0.65**  
 
 
 
 
 129
Table 9.  Environmental variable and macroinvertebrate metric Pearson correlations with 
1
st
 and 2
nd
 axes of nonmetric multidimensional scaling (NMDS) ordination.  Only 
variables with significant correlations to NMDS Axis 1 or Axis 2 are shown.   
* p < 0.05, ** p < 0.01, *** p < 0.001. Hydrologic, Habitat, and Chemical variable 
defintions are as in Tables 2 ? 4.  ICI = Invertebrate Community Index  and GA-BMI = 
Georgia Benthic Macroinvertebrate Index  
 
 
 
 
Variable NMDS
1
 NMDS
2
 
Hydrologic   
5xMed -0.60**  
7xMed -0.73**  
9xMed -0.76***  
Inc1h100 -0.55*  
N>99 -0.56*  
BI 0.56*  
   
Habitat   
Flow 0.61**  
Substrate  -0.68**  
Detritus 0.55*  -0.54* 
  
Chemical   
Temp -0.89***  
DO 0.80***  
TSS  0.53* 
TDS -0.81***  
SO4 -0.87***  
NH4 -0.57* 0.62** 
DOC -0.65**  
FC -0.71**   
  
Macroinvertebrate    
Richness 0.82*** -0.83*** 
H? 0.60** -0.74*** 
Density  -0.64** 
Biomass -0.57*  
GA-BMI 0.79*** -0.74*** 
 
 
 
 
 
 
 130
Figure 1.  The study area included 20 small watersheds (shaded sections) of the 
Chattahoochee River Basin in 4 counties in the Lower Southern Piedmont ecoregion of 
western Georgia, USA.  Numbers refer to watersheds and correspond to Table 1. The city 
of Columbus is located in western Muscogee County.  County names are in capital letters 
 
.  
 
 
 
 
 
Columbus
20 020Ki lom eters 
 
?
9
3
10
8 
4
2
6 
14
1
12
13
7
11
5
19
16
15
18
17
20
MUSCOGEE 
HARRIS 
TROUP 
MERIWETHER 
Chattahoochee
River
 
 131
Figure 2.  Composite and community invertebrate metrics as a function of land use /land 
cover (LU/LC) category.  Horizontal lines are based on annual site means, symbols are 
seasonal site means.  Letters identify significant LU/LC category differences from annual 
General Linear Model.   
F
o
r
e
s
t
S
i
l
v
i
c
u
l
t
u
r
e
P
a
s
t
u
r
e
D
e
v
e
l
o
p
i
n
g
U
r
b
a
n
Bi
o
m
as
s (
g
/m
2
)
0
2
e
+
4
4
e
+
4
6
e
+
4
8
e
+
4
1
e
+
5
T
a
x
a
 R
i
c
hnes
s
10
20
30
40
50
60
Summer
Winter
Spring
GA
-
B
M
I
0
20
40
60
80
F
o
r
e
s
t
S
i
l
v
i
c
u
l
t
u
r
e
P
a
s
t
u
r
e
D
e
v
e
l
o
p
i
n
g
U
r
b
a
n
D
i
ve
rs
it
y (
H'
)
1.0
1.5
2.0
2.5
3.0
A
B B
B
C
A
B
B
B
C
A
BB
BC
B
AB
A
A
AB
B
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 132
 
Figure 3.  GA-BMI (y = 11.658x
-0.3104
) as function of Impervious surface (IS), illustrating 
potential watershed IS threshold beyond which biotic integrity is compromised. 
 
0.0 0.1 0.2 0.3 0.4
GA
-
B
M
I
0
10
20
30
40
50
60
R
2
 = 0.63
p < 0.001
Proportion Impervious Surface
  
 
 133
Figure 4.  Nonmetric multidimensional scaling ordination (NMDS) of sites in species 
space.  Symbols are study sites coded by land use classifications. Arrows on axes show 
direction of significant correlated LU/LC and macroinvertebrate community / composite 
metric scores and vectors show relative direction and strength of correlated values for A) 
macroinvertebrate functional feeding and taxonomic groups and B) environmental 
variables.  Axes 1 and 2 explained 79.3 and 9.8% of the total variation, respectively.  
Stress = 7, Final instability = 0.00001, number iterations = 66.  
 
. 
EPT
Diptera
%Olig
o 
%tolerant
%scraper
-2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0
0.5
1.0
0.0
-0.5
-1.0
Substrate
Detritus
Temp
DO
TDS
SO4
NH4
DOCFC
7xmed
9xmed
-2.5
-1.0
-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0
-0.5
0.0
0.5
1.0
Axis 1
LU/LC 
Mixed 
Silviculture 
Pasture 
Developing 
Urban 
Impervious 
D
ensi
t
y 
D
e
ci
duous 
Im
per
vio
u
s 
P
a
sture 
GA-BMI 
H? 
Richness 
Deciduous
 
 134
CHAPTER 5.  Land use effects and the influence of ecosystem size, resource 
availability, and disturbance on stream food webs. 
5.1  ABSTRACT 
 Food webs in aquatic systems can be constrained by several factors including, but 
not limited to, ecosystem size, resource availability and disturbance.  I evaluated the 
impact of these influences on stream food webs in watersheds that differed in their degree 
of land use / land cover (LU/LC) alteration in streams draining 12 watersheds (3 urban, 3 
suburban, 3 pasture, 3 forest) of the Lower Piedmont of western Georgia, USA.   
Specifically, I quantified the relationships between LU/LC and several measures of 
ecosystem size, resource availability, and disturbance and how these environmental 
factors interact to explain mean trophic position (TP), food chain length (FCL), and food 
web structure using stable C and N isotope signatures (?
13
C and ?
15
N, respectively).  ?
13
C 
showed significant enrichment with increasing watershed impervious surface and 
significant depletion with increasing forest cover, suggesting a basal resource shift in 
food web structure and the potential for ?
13
C as an indicator tool.  Also, many of the 
environmental predictors were related to LU/LC, particularly average stream area 
(ecosystem size), chlorophyll a abundance (resource availability), spate flow frequency 
(disturbance) and water temperature (disturbance).  Accordingly, FCL, which increased 
with taxa richness, showed strong negative associations with disturbance variables, but 
not ecosystem size or resource availability.  Further, average food web architecture 
 
 135
revealed that terrestrial subsidies were an important resource in all food webs and that 
forest stream food webs had a high degree of feeding segregation with more overlap 
among feeding compartments in streams with higher LU/LC disturbance.  Together, these 
results provide empirical evidence that LU/LC ? induced disturbance can dramatically 
influence the length and overall architecture of stream food webs. 
5.2 INTRODUCTION 
The study of energy transfers among basal resources, producers, and consumers 
has long been a topic of ecological interest (Elton 1927, MacArthur 1955).  From simple 
food chains to detailed energy transfer webs, the examination of food webs can provide 
inference on observed community structure as well as reveal patterns of resource use 
among organisms, including such interactions as herbivory, competition, and predation 
(Paine 1966, 1980, Carpenter et al. 1985, Power 1990, Navarrete et al. 2000).  In addition 
to these familiar interactions, food web structure in freshwater systems can be influenced 
by myriad factors such as colonization potential, terrestrial subsidies, and gene flow, all 
of which operate at various spatial and temporal scales (Post 2002a, Woodward and 
Hildrew 2002).   Recognizing the influence of such disparate drivers of freshwater food 
webs can help lead to a better understanding of the processes shaping these ecosystems. 
Food chain length (FCL), the number of transfers of energy from the base to the 
top of a food web, is a fundamental characteristic of the structure and function of food 
webs (Elton 1927, Vander Zanden and Fetzer 2007).  FCL influences foodweb dynamics 
and trophic interactions (Oksanen et al. 1981, Pace et al. 1999), modifies regulation of 
biogeochemical cycles (Wootton 1994, Schindler et al. 1997), and influences the 
bioaccumulation of contaminants (Kidd et al. 1995).  Inherent in the examination of FCL 
 
 136
is trophic position (TP), or the location of an organism in the food web (Post 2002a).  In 
contrast to discrete trophic levels, TP is a continuous measure of the energy path to a 
consumer that incorporates omnivory and other complexities of natural systems (Levine 
1980, Polis and Strong 1996, Vander Zanden and Rasmussen 1999).  FCL can be 
considered the maximum TP in a system; thus both measures are integral in 
understanding ecological processes and food web dynamics of a community (Post 
2002b). 
FCL, and food webs in general, are determined by several interacting factors 
operating at different scales.  First, the history of community organization, such as 
colonization history, membership, and biogeographical constraints can strongly influence 
food web structure, particularly in isolated or evolutionarily young systems (Post 2002a).    
When colonization constraints do not limit food web structure, resource availability 
and/or ecosystem size can exert influence.  According to the Productive Space 
Hypothesis, FCL should increase as a function of both ecosystem size and resource 
availability (Schoener 1989), however, most evidence points more to effects of 
ecosystem size than resources; there has been less evidence that resource availability 
strongly influences FCL in aquatic systems (Spencer and Warren 1996, Vander Zanden et 
al. 1999, Post et al. 2000).  Predator-prey interactions can also influence FCL by 
moderating species? abundances or by causing species absence and/or avoidance 
(Hildrew et al. 1984, McPeek 1990),  particularly at low-intensity levels of disturbance 
(Peckarsky 1983).  At higher-intensity disturbance, predation may become less important 
and competition between organisms increases, potentially influencing FCL (Peckarsky 
1983, Menge and Sutherland 1987).   In systems with frequent, extreme disturbance, FCL 
 
 137
should shorten with the dynamical constraints of abiotic instability largely regulating 
community structure (Pimm and Lawton 1977, Peckarsky 1983, Menge and Sutherland 
1987). 
Human activities can also strongly influence FCL and the structure of aquatic 
food webs.   Resource alteration through nutrient enrichment (sewage inputs, fertilizer 
runoff) is a significant factor affecting food web structure, often leading to increases in 
primary productivity of opportunistic producers and potentially causing changes in 
community structure (Carpenter et al. 1998, deBruyn et al. 2003, Tewfik et al. 2005).  
Species invasions and other anthropogenic alterations in species interactions such as 
predator removal have been shown to have dramatic effects on food webs by shortening 
FCL (Pauly et al. 1998, Rahel 2000).  Further, changes in disturbance regimes, 
particularly hydrologic alterations, can result in unstable environments that exclude all 
but the most tolerant of organisms and contribute to species invasion, all of which can 
simplify food webs and potentially shorten FCL (Wootton et al. 1996, Nystrom and 
McIntosh 2003, Walsh et al. 2005b).  However, extensive empirical evidence for the role 
of disturbance on FCL is lacking (Post 2002a) 
In this study, I examined the role of human-induced disturbance in the form of 
watershed land use / land cover (LU/LC) change on FCL and structure of stream food 
webs.  I used stable C and N isotope ratios (?
13
C and ?
15
N, respectively) to provide time-
integrated data regarding energy flow and feeding relationships of stream organisms in 
multiple watersheds with various levels of LU/LC disturbance.  
13
C of many primary 
producers vary but the stable C isotope ratios (?
13
C, where ? = deviation from a standard) 
of consumers are similar to that of their food (DeNiro and Epstein 1978).  However, the 
 
 138
N pools of animals are enriched with 
15
N
 
relative to their food and this enrichment is on 
average +3.4 ?, i.e. 3.4 ? difference in trophic levels (Deniro and Epstein 1981, 
Minagawa and Wada 1984).   Thus, the use of stable isotopes can be used to infer the 
continuous trophic position of organisms and, by avoiding the discreteness of trophic 
levels, may provide a more realistic picture of the feeding relationships by incorporating 
the influence of omnivory (Post 2002b).   
The objectives of my study were to relate stream FCL to watershed LU/LC and 
instream conditions in an attempt to understand the potential impacts of resource 
availability, system size, and environmental stability on food web patterns associated 
with anthropogenic disturbance.  Further, I constructed and compared generalized food 
webs among different types of LU/LC.  I predicted that FCL would decrease with 
increased LU/LC disturbance and that food webs in forested watersheds (reference 
conditions) would be more complex with upper trophic levels more reliant upon instream 
prey resources than food webs from LU/LC disturbed watersheds.  
5.3 METHODS 
5.3.1 Study area 
I studied stream reaches from tributaries of the Chattahoochee River in west 
Georgia, USA, in the Southern Outer Piedmont ecoregion (Figure 1).  Twelve small 
watersheds that varied in their degree of urbanization and consequently stream 
physicochemical conditions were selected in and around the city of Columbus, Muscogee 
County (Table 1).  Watersheds ranged in land cover from intense urbanization and active 
suburban development to pasture to heavily forested areas and streams draining these 
watersheds were 2
nd
 to 3
rd
 order.  This large range in landscape character allowed 
 
 139
comparison across geomorphically similar streams that differed primarily in watershed-
level LU/LC and associated variation in streamwater physicochemical conditions.   
I determined watershed boundaries and size from USGS 30-m resolution digital 
elevation models and ArcView 3.2a software (Environmental Research Systems Institute, 
Inc., Redwoods, Calfornia).   True color (3-band) aerial photographs of study watersheds 
were taken during leaf-off in March 2003 to determine LU/LC.  I also assigned each 
watershed to 1 of 4 broad LU/LC categories (Urban, Developing, Pasture, and Forest) to 
aid in developing LU/LC specific food webs (Table 1).  These categories were based on 
the dominant land cover in the watershed (% IS, pasture, and/or forest) from land cover 
analysis except for Developing which was largely forested with current or recent 
residential development (See (Lockaby et al. 2005) for detailed land cover analyses 
methods).   
5.3.2 Resource availability  
All food web and environmental variables were collected from a 100 to 200 m 
representative stream reach in each watershed.  To address potential differences among 
watersheds in resource availability, I determined mean macroinvertebrate biomass, algal 
biomass, and organic matter abundance.   Macroinvertebrate biomass was determined by 
quantifying benthic macroinvertebrates in 3 pools and 3 runs per stream reach during 
September (summer), February (winter), and April ? May (spring) in 2003 ? 2004 using a 
Surber sampler (250-?m mesh; 0.093 m
2
 sampling area).  I consolidated multiple Surber 
collections in each habitat (3 for pools, 4 for runs), resulting in a 0.27- and 0.36-m
2
 
sample from each pool and run, respectively (1.89 m
2
 total area sampled for each study 
reach per stream and season).  I elutriated excess sediment and preserved all samples with 
 
 140
95% EtOH in the field.  In the laboratory, I removed all organisms >2mm with the 
unaided eye, and subsampled the remainder under a dissecting microscope (? 300 
organisms, (Vinson and Hawkins 1996).  I counted, measured length (nearest mm), and 
identified macroinvertebrates to the lowest possible taxonomic level (usually genus or 
morphospecies) using keys in (Merritt and Cummins 1996), (Wiggins 1996) and 
estimated biomass using published lengh / mass equations (Benke et al. 1999).   
To estimate benthic algal biomass, I removed the top layer of benthic sediment 
near where invertebrates were sampled with a shallow inverted 5 cm diameter (15 mL 
volume) petri dish and trowel (Barbour et al. 1999).  Samples were kept on ice in the dark 
until processed for chlorophyll a.  I extracted chlorophyll a from the sediment of each 
sample in 90% acetone and measured this extracted pigment fluorometrically (Turner 
TD-700, Turner Designs, Sunnyvale, CA).  I averaged 3 petri dish samples for each of 
the 6 habitats (3 runs and 3 pools) and used these averaged habitat values to determine 
mean chlorophyll abundance.   
I quantified benthic organic matter (OM) and substrate size by sampling 
transitional areas between runs and pools where organisms were sampled to standardize 
efforts and avoid error associated with scour in runs and deposition in pools.  I sampled 
OM by determining the ash-free dry mass (AFDM) of 9 replicate 2.5 x 10 cm benthic 
cores. 
5.3.3  Environmental stability  
To assess potential differences among watersheds in environmental stability, I 
measured several physicochemical and hydrological variables that have been shown to be 
influential to aquatic taxa as well as computed a general habitat quality index in each 
 
 141
stream.  Stream temperature was measured continuously with HOBO? Temp data 
loggers from July 2003 ? July 2004, whereas total dissolved solids (TDS) and total 
suspended solids (TSS) were determined monthly, as detailed in (Schoonover and 
Lockaby 2006).  Continuous stream discharge (Q) was quantified from July 2003 to July 
2004 using a Mini-Troll? pressure transducer data logger (In-Situ Inc., Ft. Collins, 
Colorado) installed near the outflow point of each watershed to determine relative 
hydrologic signatures of the 12 streams (Schoonover et al. 2006).  I considered 2 
parameters of the hydrographs reflecting general magnitude and frequency of spate flows, 
maximum discharge (MaxQ) and # times discharge exceeded 7 x median flow (7xMed).  
These parameters were shown to be important predictors of biota in previous studies 
(Chapter 2,3).    
To assess available habitat quality/quantity, I used a comprehensive multimetric 
habitat assessment from the Georgia Environmental Protection Division designed for use 
in fish biomonitoring (GADNR 2007).  This Habitat Index included visual estimates of 
available cover (number and frequency of habitats), substrate characterization (type and 
condition), pool morphology (shape and frequency), channel alteration (frequency of 
riprap, dredging, etc), channel sinuosity (run-to-bend ratio), sediment deposition (particle, 
point bar, island size), flow status (degree to which channel is filled with water), bank 
condition (erosion potential and vegetation cover) and riparian condition (vegetation 
cover/quality) (GADNR 2007).  This assessment involves averaging 3 individuals? 
summed scores (1-10 or 1-20, depending on parameter) of the different habitat 
parameters to obtain an overall habitat quality value for the representative reach, with 
high average score indicating high habitat quality.  The same 3 observers were used at all 
 
 142
sites (BSH, KOM, RMM).  All environmental data were correlated with watershed 
LU/LC and food web variables (see below). 
5.3.4  System size  
 Along with watershed size, I also estimated mean available wetted area of the 
stream reach (i.e., mean stream size).  This variable was the product of the mean width 
and the mean depth of the stream as measured 3 times a year from 2002 ? 2004. 
5.3.5  Field collections  
 Each stream reach was sampled for periphyton, invertebrates and vertebrates in 
spring 2006.  Each stream had undergone exhaustive sampling in other studies (see 
Chapters 2 ? 5), so the resident organisms present in each stream were well known and, 
thus, sampling proceeded until at least all common organisms were collected.   All 
samples were replicated 3 to 5 times when possible and immediately put on ice in the 
field and later frozen until processed in the laboratory.   
Periphyton was collected by removing the top layer of benthic sediment with a 
shallow inverted 5 cm diameter (15 ml volume) petri dish and trowel (Barbour et al. 
1999).  Samples were collected in transitional areas between runs and pools to balance 
the influences of scour and deposition.  Organic matter, in the form of leaf detritus, was 
collected from leaf packs in depositional areas.  Terrestrial invertebrate drift was 
collected using 4 drift nets (0.09m
2
 sampling area, 250 ?m mesh) set for 4h during 
daylight hours in each stream.    Aquatic invertebrates and salamanders were collected 
with dip-nets (250 ?m mesh) and fish were collected with seines and a backpack 
electroshocker (Smith-Root LR-24) from all available habitats in the reach.   All aquatic 
macroinvertebrates were assigned to one of 7 functional feeding group (Merritt and 
 
 143
Cummins 1996) including collector/filterers (CF), collector/gatherers (CG),  predators 
(PR), scrapers (SC), shredders (SH), and scavengers/crayfish (SV).  Fish were assigned to 
one of four feeding guilds (Schleiger 2000), including herbivores (HBV), insectivores 
(ISV), omnivores (OMN), and piscivores (PSV).    
5.3.6 Sample preparation and isotope analysis 
In the laboratory, I removed, counted, measured length, and identified all 
macroinvertebrates (family level, subfamily for Chironomidae) from the dip net samples 
using standard keys (Merritt and Cummins 1996, Wiggins 1996, Epler 2001).  For each 
fish specimen, I measured length and weight and then removed a skinned, boneless 
portion of dorsal muscle for analysis, whereas for salamanders, a portion of the tail was 
removed.  After rinsing in distilled water, all foodweb samples were dried 48h at 60?C, 
ground to a fine powder with a mortar and pestle and then stored in clean scintillation 
vials.  Subsamples for most samples were then weighed (nearest 10
-5
 g) and placed in 4 x 
6 mm tin capsules for isotopic analysis.  For some smaller macroinvertebrates, several 
whole individuals were prepared without grinding to achieve necessary weight limits for 
analysis.   
 All samples were analyzed for isotope ratios (
13
C / 
12
C and 
15
N / 
14
N) by gas 
isotope-ratio mass spectroscopy at the Colorado Plateau Stable Isotope Laboratory at 
Northern Arizona University, Flagstaff, AZ, USA.  Isotope ratios are reported in parts per 
thousand (?) relative to standards (Pee Dee Belemnite for C, atmospheric N for N), 
defined in delta notation as: 
?
13
C or ?
15
N =  (R
sample
 / R
standard
 ? 1) x 10
3
 
where R = 
13
C / 
12
C or 
15
N / 
14
N (Craig 1957, Jepsen and Winemiller 2002).   
 
 144
5.3.7  Data analysis 
For each watershed, I calculated mean C (? ?
13
C) and N (? ?
15
N) isotopic 
signature as well as mean TP and FCL.  There is often considerable variation in ?
15
N and 
?
13
C in basal resources among different aquatic systems, so it is generally necessary to 
baseline-correct ?
15
N values to compare trophic position among systems (Cabana and 
Rasmussen 1996, Vander Zanden et al. 1997, Post 2002b).  Therefore to determine 
trophic position (TP) for each organism, I used:  
? + (?
15
N
sample 
? ?
15
N
base
) / ?
n
  
 and, similarly, to determine FCL, I used: 
? + (?
15
N
top consumer 
? ?
15
N
base
) / ?
n
  
where ? is the trophic position of the organism used to estimate ?
15
N
base
, ?
15
N
sample
 is the 
isotopic signature of an individual sample, ?
15
N
top consumer 
is the isotopic signature of the 
top consumer, ?
15
N
base
 is the isotopic signature of the baseline resource, and ?
n
 is the 
trophic fractionation of N, generally considered 3.4? per trophic level (Post 2002b, 
Vander Zanden and Fetzer 2007).  Detrital energy sources in aquatic systems, however, 
often show high temporal variation in ?
15
N and ?
13
C (France 1995c, a, Cabana and 
Rasmussen 1996, Post 2002b).  Also, after evaluation of C:N biplots, it became apparent 
that most organisms were not using leaf detritus or algae as a primary C source (i.e. most 
were not situated directly above organic matter or algae in C:N space), but instead, a 
slightly less enriched source.  Therefore, for the baseline resource I used an inferred 
averaged value for particulate organic matter (POM) based on the signatures of primary 
consumers (Post 2002b).  These consumers included all macroinvertebrates known to be 
 
 145
collectors in a given stream (Corbicula, Cheumatopsyche, Chironominae, etc.) and the 
isotopic signature of POM was estimated as  
?
13
C
POM
 = ?
13
C
b
 ? 1? and 
?
15
N
POM
 = ?
15
N
b
 ? 3.4? 
where POM is particulate organic matter, b is collector and 1? and 3.4? are the mean 
trophic fractionation values for C and N, respectively (Post 2002b, Vander Zanden and 
Fetzer 2007).  A POM value was generated for each CF and then averaged for an 
inferred, stream-specific POM value.   
Stable C and N values, mean TP, and FCL were then correlated with LU/LC, 
resource, size and stability variables and regressed against total taxa richness.  Using 
means for invertebrate functional feeding groups, fish feeding guilds, and salamanders 
for each stream, I constructed 4 functional food webs based on means for LU/LC (Forest, 
Pasture, Developing, and Urban).  Only functional groups found in at least 2 of the 3 
streams per category were included in the mean webs to minimize influence of local 
variation.  Finally, pairwise comparisons between Forest and all other food webs (Forest 
vs. Pasture, Forest vs. Developing, Forest vs. Urban) were made by calculating the 
directional change of each consumer component in food web biplots and circular 
statistics were used to determine any significant differences in these changes (Schmidt et 
al. 2007).  Directional change was measured as the length (r) and angle (?) of difference 
between each corresponding functional group in trophic space (?
13
C ? ?
15
N) in a 
comparison 2 mean food webs.  Rao?s spacing test for circular uniformity (U) was used 
to determine whether each foodweb comparison departed nonrandomly from uniformity 
and showed evidence for a preferred direction (Batschelet 1981). 
 
 146
5.4 RESULTS 
In total, 834 samples were collected for analysis, representing at least 54 different 
taxa, including 30 invertebrate families, 21 fish species, 1 salamander, 1 toad, and benthic 
algae taxa (Table 3).  Aquatic taxa richness ranged from 8 ? 30 across the watersheds and 
many watersheds were well represented with multiple taxa in each functional group 
(Table 4). 
5.4.1 Trophic position and food chain length 
Mean ?
13
C values ranged from -31.043 to -23.190 ? whereas mean ?
15
N values 
ranged from 5.141 to 8.842 ? (Table 5).  Carbon values were strongly correlated with 
LU/LC, showing overall enrichment with increasing impervious surface and concomitant 
depletion with increasing watershed forest cover (Table 6).  Somewhat surprisingly, ?
15
N 
values were not significantly correlated to any LU/LC parameter.  However, mean ?
15
N 
significantly increased with watershed size and TSS concentration (Table 6).   
TP ranged from 1.456 in the heavily urbanized BU1 to 2.356 in HC2, a pasture 
watershed.  FCL ranged from 2.623 in BU2, an Urban watershed, to 4.338 in the heavily 
forested MO (Table 5).  In all of the Urban and 2 of the 3 Developing watersheds, the top 
consumer was a sunfish (Family Centrarchidae, Table 5).  In all but 1 of the other 
streams, the top consumer was a benthic-feeding fish (Table 5). 
Mean TP and FCL decreased with increasing % impervious surface and FCL 
increased with % forest cover.  Neither mean TP nor FCL were correlated with system 
size variables, however, both were strongly correlated to several resource and 
environmental variables (Table 6).  Mean TP significantly decreased with increasing 
invertebrate biomass and chlorophyll a (Table 6), whereas both mean TP and FCL 
 
 147
decreased with median stream temperature and TDS concentrations (Table 6).  Spate 
frequency was also related significantly to mean TP and FCL, as both decreased with 
increasing frequency of spate flows, as measured by the variables 7xMed.  Further FCL 
was significantly longer in streams with higher quality habitat, as evidenced by the 
significant positive correlation with the habitat index score (Table 6).  Further, mean TP 
and FCL both significantly increased with increasing taxa richness (Figure 2). 
 5.4.2 Mean Food Webs 
 In general, streams draining urban watersheds showed slightly more variation 
within functional groups for ?
13
C, and less so for ?
15
N, than did the other LU/LC streams 
(Figure 3).  However, within-group variation was relatively consistent as there was no 
significant difference between the different LU/LC types in terms of C or N isotopic 
variation of functional groups (Figure 4).  The only functional group that was not found 
in all LU/LC categories was shredders (SH), which was absent from Urban streams.  
There also were several cases among Urban streams, and to a lesser extent in the Pasture 
and Developing streams, where a functional group was found in only 1 of the 3 replicate 
streams (Table 4).   
 Basal resources in the food webs were generally consistent among LU/LC 
categories, particularly OM (Figure 5).  Derived POM was generally less C-enriched than 
OM, except for Urban streams, where it was more enriched.  In all food webs, benthic 
algae were highly N-enriched, often at or above ?
15
N signatures of secondary consumers 
(Figure 5).    
Food webs seemed to show a distinct separation between vertebrates and 
invertebrates, and particularly so in the Developing web, with most fish being relatively 
 
 148
N-enriched and C-depleted as compared to invertebrates (Figure 5).  Exceptions to this 
generalization were with terrestrial insects, which were relatively C-enriched, and with 
crayfish, which showed C- and often N-enrichment compared with other aquatic 
invertebrates (Figure 5).  With few exceptions, most macroinvertebrate consumers were 
positioned over POM.  Exceptions were the SH category in the Forest web (?
13
C 
enriched) and in the Developing web (?
13
C depleted) as well as the SC category in the 
Forest web (Figure 5).    
The Urban food web, in general, was more C-enriched than the Forest and 
Developing food webs and significantly so for fish (F
3, 11
 = 5.99, p = 0.019; Figure 5).    
Based on their position in each web, OMN, ISV, and PSV fishes and salamanders 
appeared to be deriving their ultimate C from both terrestrial and aquatic resources.  
Piscivores and/or insectivores were consistently at the top of the food webs, except in the 
Pasture web, where HBV fish were particularly N-enriched.  Also, with the exception for 
the Urban web, HBV fish were depleted in ?
13
C compared with other fish feeding guilds 
(Figure 5).    
The Forest foodweb biplot was significantly different from the Pasture, 
Developing, and Urban foodweb biplots in terms of shared components (Figure 6).  There 
was a significant directional movement of shared taxa (? = 74.7?, r = 0.96, Rao?s U = 
266.1, p < 0.01) in the Forest to Pasture comparison of foodweb biplots (Figure 6).  There 
was also significant directional movement in the Forest to Developing foodweb biplot 
comparison (? = 326.2?, r = 0.71, Rao?s U = 176.2, p < 0.05).  Finally, the Forest to 
Urban foodweb comparison also showed strong, significant directional movement (? = 
79?, r = 0.99, Rao?s U = 283.1, p < 0.01).    
 
 149
5.5 DISCUSSION 
The analysis of food webs with stable C and N isotopes allows ecologists to 
describe broad-scale energy flow in and trophic arrangement of systems in a time-
integrated fashion (Peterson and Fry 1987).  My study shows that stable isotope analysis 
of food webs can be used to examine the community-level impact of LU/LC on streams.  
Specifically, these results provide empirical evidence that disturbance associated with 
LU/LC, particularly urbanization, can have strong effects on the structure and function of 
stream food webs.   
5.5.1  ?
13
C and ?
15
N differences among watersheds 
Interestingly, there was no overall ?
15
N-enrichment observed with increasing 
urbanization as has been observed in other studies of urban streams (McClelland et al. 
1997, Steffey and Kilham 2004, Ulseth and Hershey 2005).  However, there was 
considerable ?
13
C enrichment associated with the loss of watershed forest cover / increase 
of urban cover as well as with stream size, as evidenced by the correlation analyses and 
the average food webs.  Stable carbon isotope signatures often vary from system to 
system for various reasons (France 1995b, Finlay 2001), however systematic shifts over 
12 systems as observed in my study suggest alterations in basal resources.  Since stream 
size and loss of forest cover was correlated with C enrichment, it is possible that there 
was a shift in the importance of allochthonous litterfall in the smaller streams to 
autochthonous producers in the larger streams, particularly in the larger streams of the 
Urban and Pasture watersheds (Vannote et al. 1980, Rosenfeld and Roff 1992, Jones et al. 
1998).  Urban and Pasture watersheds in this area have been shown to have increased 
stream temperatures and levels of nutrients (Schoonover and Lockaby 2006), which could 
 
 150
lead to increases in primary productivity, yet, we saw little evidence in our biplots for 
major differences in ultimate source of basal carbon.  However, basal resources are 
notoriously variable (spatially and temporally) in terms of C and N isotopic signatures 
(France 1995b, Finlay 2001), so further sampling in other seasons would be necessary to 
fully address this shift in ?
13
C among the streams. 
I also observed ?
13
C enrichment with increased % impervious surface cover. A 
likely explanation for this enrichment is fecal contamination in urbanized watersheds as a 
result of increased faulty sewage lines, combined sewer overflow inputs, and other 
sources of fecal contamination.  In fact, near stream manhole covers were routinely 
displaced during heavy rain events in some of the urban watersheds in this study, 
allowing untreated sewage inputs directly to the stream (personal observation).  
Moreover, these urban watersheds have been shown to have elevated fecal coliform 
counts, particularly during storm events (Schoonover and Lockaby 2006).  Further, 
human sewage particulate organic matter generally has a heavier C isotope ratio than 
stream POM (Spies et al. 1989); thus consumers using such a fecal-contaminated source 
of OM would be expected to show enriched ?
13
C signatures.  Therefore in these 
watersheds, as in some other studies, ?
13
C of stream organisms can possibly serve as a 
time-integrated label for fecal contamination or urbanization in general and thus a 
potentially meaningful management tool (Spies et al. 1989, deBruyn and Rasmussen 
2002). 
5.5.2  Correlates of food chain length 
Several interacting factors are thought to explain food chain length, including 
resource availability, system size, colonization history, and environmental stability 
 
 151
(Pimm 1982, Briand and Cohen 1987, Schoener 1989, Post 2002a).  In addition, there is a 
growing literature integrating landscape influences on food web structure, emphasizing 
geology, hydrology, land use and habitat fragmentation on trophic interactions (Polis et 
al. 1997, Woodward and Hildrew 2002).  Specifically, effects of land use disturbance on 
stream systems often are considered functions of increased flood magnitude and 
frequency (Booth and Jackson 1997), altered streambed morphology and stability (Booth 
and Jackson 1997, Wang and Lyons 2003b) altered sediment and chemical inputs 
(Wernick et al. 1998, Allan 2004), and degraded riparian zones (Finkenbine et al. 2000, 
Groffman et al. 2003).  In my study, FCL decreased with increasing landscape 
disturbance, as reduced forest cover and elevated impervious surface cover.  It should be 
noted, however, that these 2 sources of landscape disturbance, although correlated, are 
not entirely reciprocal.  Except for the Urban streams, most watersheds had low % IS and 
were indistinguishable, so that FCL relationships with impervious surface largely reflect 
the influence of intense urbanization.  In contrast, % forest cover was more evenly spread 
across study watersheds, with Urban watersheds having the lowest % cover, followed by 
Pasture, Developing, and Forest watersheds.  Thus, the % Forest metric better reflects the 
entire spectrum of LU/LC disturbance in the watersheds than % IS or % pasture. 
Interestingly, on a reach scale, FCL was not correlated with watershed size or 
resource availability; however, there were strong correlations with reach measures of 
disturbance.  A similar trend was observed in relatively pristine Alaskan streams, where 
FCL was shorter where disturbance was greater (Parker and Huryn 2006).  In my streams, 
FCL decreased with increasing average stream temperature, TDS concentrations, and 
spate flow frequency and increased with increasing instream habitat quality.  There was 
 
 152
also a strong relationship between FCL and taxa richness.  These relationships suggest 
hydrologic alteration associated with LU/LC disturbance as a strong driver of food web 
structure in these streams.   
Of the multiple landscape disturbances affecting streams, hydrological alteration 
is one of the more obvious and pervasive (Booth and Jackson, 1997; Groffman et al., 
2003; Wang and Lyons, 2003; Walsh et al., 2005).  As a result of high levels of 
watershed imperviousness and connectedness, streams draining urban and urbanizing 
watersheds often display flashy hydrographs with multiple peak flows and reduced base 
flows (Ferguson and Suckling, 1990; Rose and Peters, 2001; Schoonover et al., 2006).  
Stormflow magnitude and frequency often increase in agricultural settings because of the 
use of drainage ditches, loss of wetlands, and soil compaction (Peterson and Kwak, 1999; 
Allan, 2004).  This effective transport of overland flow can have far-reaching 
implications on instream conditions, including elevated water contaminants and solutes, 
elevated water temperature, altered instream habitat, loss of species and alterations of 
assemblages; each of these stressors could have separately or in combination reduced 
FCL in study watersheds (Paul and Meyer 2001, Allan 2004). 
5.5.3  Food chain length and web structure 
My results in part support the model of dynamical constraints in that FCL was 
reduced in systems with higher disturbance (Pimm and Lawton 1977, Menge and 
Sutherland 1987).  However, the model suggests that rare species high on the food chain 
are more likely to be lost from a disturbance event than the more abundant lower 
consumers, resulting in shorter food chains in disturbed environments (Jenkins et al. 
1992, Thompson and Townsend 2005).  In my study, there was not the general loss of the 
 
 153
top consumer group, as insectivores and omnivores were consistently top consumer in all 
streams irrespective of disturbance intensity; however, there was a general loss of 
benthic-feeding fish in the more hydrologically disturbed streams.  In addition, increased 
disturbance appeared to have a strong effect on organisms in the middle trophic positions, 
particularly shredding macroinvertebrates and crayfish, and to a lesser extent filtering 
macroinvertebrates, salamanders, and herbivorous fish.  In fact, the longest and the 
shortest FCL in the study (MO and BU2, respectively) both contained Lepomis sunfish as 
the top consumer. This pattern of middle trophic position loss but maintenance of similar 
top consumers (at least functionally) across a disturbance gradient is partially 
contributable to the generalist behavior and broad diets of many top consumers in these 
streams.  Many of the fish species in these streams are cosmopolitan (Swift et al. 1986) 
and thus are capable of inhabiting a wide array of conditions (Boschung and Mayden 
2004); therefore, they should be somewhat resilient to compositional changes in food 
resources attributable to landscape differences in these streams.  In summary, disturbance 
and environmental stability appears to reduce FCL in these streams by limiting the 
species of top predator, but not the entire functional group, as well as consumer 
membership in middle trophic positions. 
Mean food web biplots revealed certain trends and trophic relationships that were 
not obvious with simple FCL examination.  As noted in other systems, terrestrial inputs 
into these streams, particularly terrestrial invertebrate prey for fish, appears to play an 
important energetic role (Nakano et al. 1999, Baxter et al. 2005).  Insectivorous and 
omnivorous fish in all streams, on average, were more ?
13
C enriched than most of the 
aquatic invertebrates, and were more positioned over terrestrial invertebrates in biplot 
 
 154
space, suggesting a heavy reliance on terrestrial prey.  However, there were many 
differences among the different types of LU/LC food webs.  In particular, mean biplots 
revealed a smaller spread of ?
13
C and very little overlap of functional groups in the Forest 
than the other streams, stronger segregation between benthos and fish in the Developing 
and Pasture streams than in Forest and Urban streams, and an almost chain-like 
appearance of the food web in the Urban streams.  Further, mean Forest foodwebs were 
significantly different than Pasture, Developing, and Urban foodwebs.  The mean vector 
of change from Forest to Pasture and from Forest to Urban foodwebs revealed strong 
?
13
C enrichment, while the mean vector of change from Forest to Developing suggested 
slight ?
15
N enrichment.    Taken together, these suggest 1) a close association between 
fish, benthos, and terrestrial resources in the Forest watersheds and this association tends 
to break down with increased landscape disturbance as benthos become less important of 
a resource for fish (Baxter et al. 2005); 2) a higher degree of feeding (niche) 
compartmentation in the Forest streams and higher feeding (niche) overlap in the more 
disturbed systems (Death 2004); and 3) that Urban stream food webs are dramatically 
simplified and retain minimum function with fewer functional groups (Walsh et al. 
2005b).  
5.5.4 Conclusions 
My results suggest that disturbance from watershed LU/LC can have a dramatic 
impact on food web architecture in southeastern streams subject to human influence.  In 
these streams, ?
13
C appears to be an indicator of general LU/LC disturbance.  This has 
important implications for management, as many physicochemical impacts of LU/LC 
disturbance are episodic in nature and can occur at relatively low levels of landscape 
 
 155
alteration (Allan et al. 1997, Barbour et al. 1999, Schoonover and Lockaby 2006).  Thus a 
time integrated marker of anthropogenic disturbance can be a valuable identification and 
management tool.   
FCL in these watersheds appears to be more determined by disturbance, 
particularly hydrologic alteration, than by ecosystem size or resource availability.  
Although the role of ecosystem size or resource availability should not be ruled out, 
especially given the temporal, interrelated nature of these influences (Post 2002a, 
Thompson and Townsend 2005), disturbance appears to be an overwhelming influence in 
these systems, particularly the Urban watersheds. 
Human population expansion and the inevitable landscape alteration caused by 
such growth have produced dramatic impacts on stream ecosystems.  By attempting to 
identify and address the specific drivers of biotic associations resulting from these 
perturbations, we can better address management and/or restoration needs designed to 
protect or minimize changes in stream biotic integrity. 
 
 
 
 
156 
5.6 REFERENCES 
Allan, J. D. 2004. Landscapes and riverscapes: The influence of land use on stream 
ecosystems. Annual Review of Ecology and Systematics 35:257-284. 
Allan, J. D., D. L. Erickson, and J. Fay. 1997. The influence of catchment land use on 
stream integrity across mulitiple spatial scales. Freshwater Biology 37:149-161. 
Barbour, M. T., J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. Rapid bioassessment 
protocols for use in streams and wadeable rivers: periphyton, benthic 
macroinvertebrates and fish, second edition. EPA 841-B-99-002, US 
Environmental Protection Agency, Office of Water, Washington D.C. 
Batschelet, E. 1981. Circular statistics in ecology. Academic Press, New York, New 
York, USA. 
Baxter, C. V., K. D. Fausch, and W. C. Saunders. 2005. Tangled webs: reciprocal flows 
of invertebrate prey link streams and riparian zones. Freshwater Biology 50:201-
220. 
Benke, A. C., A. D. Huryn, L. A. Smock, and J. B. Wallace. 1999. Length-mass 
relationships for freshwater macroinvertebrates in North America with particular 
reference to the southeastern United States. Journal of the North American 
Benthological Society 18:309-343. 
Booth, D. B., and C. R. Jackson. 1997. Urbanization of aquatic systems: degradation 
thresholds, stormwater detection, and the limits of mitigation. Journal of the 
American Water Resources Association 33:1077-1090. 
Boschung, H. T., Jr., and R. L. Mayden. 2004. Fishes of Alabama. Smithsonian Books, 
Washington D.C. 
 
157 
Briand, F., and J. E. Cohen. 1987. Environmental Correlates of Food-Chain Length. 
Science 238:956-960. 
Cabana, G., and J. B. Rasmussen. 1996. Comparison of aquatic food chains using 
nitrogen isotopes. Proceedings of the National Academy of Sciences 93:10844-
10847. 
Carpenter, S. R., N. F. Caraco, D. L. Correll, R. W. Howarth, A. N. Sharpley, and V. H. 
Smith. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. 
Ecological Applications 8:559-568. 
Carpenter, S. R., J. F. Kitchell, and J. R. Hodgson. 1985. Cascading tropic interactions 
and lake productivity. BioScience 35:634-639. 
Craig, H. 1957. Isotopic standards for carbon and oxygen and correction factors for mass-
spectrometric analysis of carbon dioxide. Geochimica et Cosmochimica Acta 
12:133-149. 
Death, R. G. 2004. Patterns of spatial resource use in lotic invertebrate assemblages. 
Hydrobiologia 513:171-182. 
deBruyn, A. M. H., D. J. Marcogliese, and J. B. Rasmussen. 2003. The role of sewage in 
a large river food web. Canadian Journal of Fisheries and Aquatic Sciences 
60:1332-1344. 
deBruyn, A. M. H., and J. B. Rasmussen. 2002. Quantifying assimilation of sewage-
derived organic matter by riverine benthos. Ecological Applications 12:511-520. 
DeNiro, M. J., and S. Epstein. 1978. Influence of diet on the distribution of carbon 
isotopes in animals. Geochemica et Cosmochimica Acta 42:495-506. 
 
158 
Deniro, M. J., and S. Epstein. 1981. Influence of Diet on the Distribution of Nitrogen 
Isotopes in Animals. Geochimica et Cosmochimica Acta 45:341-351. 
Elton, C. S. 1927. Animal Ecology. Sidgewick & Jackson, London. 
Epler, J. H. 2001. Identification manual for the Chironomidae (Diptera) of North and 
South Carolina.  North Carolina Department of Environment and Natural 
Resources, Division of Water Quality, Raleigh, NC. 
Finkenbine, J. K., J. W. Atwater, and D. S. Mavinic. 2000. Stream health after 
urbanization. Journal of the American Water Resources Association 36:1149-
1160. 
Finlay, J. C. 2001. Stable-carbon-isotope ratios of river biota: implications for energy 
flow in lotic food webs. Ecology 82:1052-1064. 
France, R. L. 1995a. Carbon isotopic variability in the composite pelagic foodweb of four 
oligotrophic lakes: feeding diversity or metabolic fractionations? Journal of 
Plankton Research 17:1993-1997. 
France, R. L. 1995b. Critical examination of stable isotope analysis as a means for tracing 
carbon pathways in stream ecosystems. Canadian Journal of Fisheries and 
Aquatic Science 52:651-656. 
France, R. L. 1995c. Source variability in d
15
N of autotrophs as a potential aid in 
measuring allochthony in freshwaters. Ecography 18:318-320. 
Groffman, P. M., D. J. Bain, L. E. Band, K. T. Belt, G. S. Brush, J. M. Grove, R. V. 
Pouyat, I. C. Yesilonis, and W. C. Zipperer. 2003. Down by the riverside: urban 
riparian ecology. Frontiers in Ecology and the Environment 1:315-321. 
 
159 
Hildrew, A. G., C. R. Townsend, and J. Francis. 1984. Community structure in some 
southern English streams - the influence of species interactions. Freshwater 
Biology 14:297-310. 
Jenkins, B., R. L. Kitching, and S. L. Pimm. 1992. Productivity, disturbance and food 
web structure at a local spatial scale in experimental container habitats. Oikos 
65:249-255. 
Jepsen, D. B., and K. O. Winemiller. 2002. Structure of tropical river food webs revealed 
by stable isotope ratios. Oikos 96:46-55. 
Jones, R. I., J. Grey, D. Sleep, and C. Quarmby. 1998. An assessment, using stable 
isotopes, of the importance of allochthonous organic carbon sources to the pelagic 
food web in Loch Ness. Proceedings of the Royal Society of London Series B-
Biological Sciences 265:105-111. 
Kidd, K. A., D. W. Schindler, D. C. G. Muir, W. L. Lockhart, and R. H. Hesslein. 1995. 
High-Concentrations of Toxaphene in Fishes from a Sub-Arctic Lake. Science 
269:240-242. 
Levine, S. 1980. Several measures of trophic structure applicable to complex food webs. 
Journal of Theoretical Biology 83:195-207. 
Lockaby, B. G., D. Zhang, J. McDaniel, H. Tian, and S. Pan. 2005. Interdisciplinary 
research at the urban-rural interface: the WestGa project. Urban Ecosystems 8:7-
21. 
MacArthur, R. 1955. Fluctuations of animal populations, and a measure of community 
stability. Ecology 36:533-536. 
 
160 
McClelland, J. W., I. Valiela, and R. H. Michener. 1997. Nitrogen-stable isotope 
signatures in estuarine food webs: a record of increasing urbanization in coastal 
watersheds. Limnology and Oceanography 42:930-937. 
McPeek, M. A. 1990. Determination of species composition in the Enallagma damselfly 
assemblages of permanent lakes. Ecology 71:83-98. 
Menge, B. A., and J. P. Sutherland. 1987. Community regulation: variation in 
disturbance, competition, and predation in relation to environmental stress and 
recruitment. The American Naturalist 130:730-757. 
Merritt, R. W., and K. W. Cummins. 1996. An Introduction to the Aquatic Insects of 
North America. Kendall/Hunt Publishing Company Dubuque, IA. 
Minagawa, M., and E. Wada. 1984. Stepwise enrichment of 
15
N along food chains: 
Further evidence and the relation between d
15
N and animal age. Geochemica et 
Cosmochimica Acta 48:1135-1140. 
Nakano, S., H. Miyasaka, and N. Kuhara. 1999. Terrestrial-aquatic linkages: riparian 
arthropod inputs alter trophi cascades in a stream food web. Ecology 80:2435-
2441. 
Navarrete, S. A., B. A. Menge, and B. A. Daley. 2000. Species interactions in intertidal 
food webs: prey or predation regulation of intermediate predators? Ecology  
Nystrom, P., and A. R. McIntosh. 2003. Are impacts of an exotic predator on a stream 
food web influenced by disturbance history. Oecologia 136:279-288. 
Oksanen, L., S. D. Fretwell, J. Arruda, and P. Niemela. 1981. Exploitation ecosystems in 
gradients of primary productivity. The American Naturalist 118:240-261. 
 
161 
Pace, M. L., J. J. Cole, S. R. Carpenter, and J. F. Kitchell. 1999. Trophic cascades 
revealed in diverse ecosystems. Trends in Ecology and Evolution 14:483-488. 
Paine, R. T. 1966. Food web complexity and species diversity. The American Naturalist 
100:65-75. 
Paine, R. T. 1980. Food webs: linkage, interaction strength and community infrastructure. 
Journal of Animal Ecology 49:667-685. 
Parker, S. M., and A. D. Huryn. 2006. Food web structure and function in two arctic 
streams with contrasting disturbance regimes. Freshwater Biology 51:1249-1263. 
Paul, M. J., and J. L. Meyer. 2001. Streams in the urban landscape. Annual Review of 
Ecology and Systematics 32:333-365. 
Pauly, D., V. Christensen, J. Dalsgaard, R. Froese, and F. T. Jr. 1998. Fishing down 
marine food webs. Science 279:860-863. 
Peckarsky, B. L. 1983. Biotic interactions or abiotic limitations?  A model of lotic 
community structure. Pages 303-324 in T. D. Fontaine III and S. M. Bartell, 
editors. Dynamics of lotic systems. Ann Arbor Science, Ann Arbor, MI. 
Peterson, B. J., and B. Fry. 1987. Stable isotopes in ecosystem studies. Annual Review of 
Ecology and Systematics 18:293-320. 
Peterson, J. T., and T. J. Kwak. 1999. Modeling the effects of land use and climate 
change on riverine smallmouth bass. Ecological Applications 9:1391-1404. 
Pimm, S. L. 1982. Food webs. Chapman and Hall, London. 
Pimm, S. L., and J. H. Lawton. 1977. Number of trophic levels in ecological 
communities. Nature 268:329-331. 
 
162 
Polis, G. A., W. B. Anderson, and R. D. Holt. 1997. Toward an integration of landscape 
and food web ecology: the dynamics of spatially subsidized food webs. Annual 
Review of Ecology and Systematics 28:289-316. 
Polis, G. A., and D. A. Strong. 1996. Food web complexity and community dynamics. 
American Naturalist 47:813-846. 
Post, D. M. 2002a. The long and short of food-chain length. Trends in Ecology and 
Evolution 17:269-277. 
Post, D. M. 2002b. Using stable isotopes to estimate trophic position: models, methods, 
and assumptions. Ecology 83:703-718. 
Post, D. M., M. L. Pace, and N. G. Hairston. 2000. Ecosystem size determines food-chain 
length in lakes. Nature 405:1047-1049. 
Power, M. E. 1990. Effects of fish in river food webs. Science 250:811-814. 
Rahel, F. J. 2000. Homogenization of fish faunas across the United States. Science 
288:854-856. 
Rosenfeld, J. S., and J. C. Roff. 1992. Examination of the carbon base in southern 
Ontario streams using stable isotopes. Journal of the North American 
Benthological Society 11:1-10. 
Schindler, D. E., S. R. Carpenter, J. J. Cole, J. F. Kitchell, and M. L. Pace. 1997. 
Influence of food web structure on carbon exchange between lakes and the 
atmosphere. Science 277:248-251. 
Schleiger, S. L. 2000. Use of an index of biotic integrity to detect effects of land uses on 
stream fish communities in west-central Georgia. Transactions of the American 
Fisheries Society 129:1118-1133. 
 
163 
Schoener, T. W. 1989. Food webs from the small to the large. Ecology 70:1559-1589. 
Schoonover, J. E., and B. G. Lockaby. 2006. Land cover impacts on stream nutrients and 
fecal coliform in the lower Piedmont of West Georgia. Journal of Hydrology 
331:371-382. 
Schoonover, J. E., B. G. Lockaby, and B. S. Helms. 2006. Impacts of land cover on 
stream hydrology in the west Georgia piedmont, USA. Journal of Environmental 
Quality 35:2123-2131. 
Sonneman, J. A., C. J. Walsh, P. F. Breen, and A. K. Sharpe. 2001. Effects of 
urbanization on streams of the Melbourne region, Victoria, Australia. II. Benthic 
diatom communities. Freshwater Biology 46:553-565. 
Spencer, M., and P. H. Warren. 1996. The effects of energy input, immigration and 
habitat size on food web structure: a microcosm experiment. Oecologia 108:764-
770. 
Spies, R. B., H. Kruger, R. Ireland, and D. W. Rice. 1989. Stable isotope ratios and 
contaminant concentrations in a sewage-distorted food web. Marine Ecology-
Progress Series 54:157-170. 
Sponseller, R. A., E. F. Benfield, and H. M. Valett. 2001. Relationships between land 
use, spatial scale and stream macroinvertebrate communities. Freshwater Biology 
46:1409-1424. 
Steffey, L. Y., and S. S. Kilham. 2004. Elevated d
15
N in stream biota in areas with septic 
tank systems in an urban watershed. Ecological Applications 14:637-641. 
 
164 
Strayer, D. L., R. E. Beighley, L. C. Thompson, S. Brooks, C. Nilsson, G. Pinay, and R. 
J. Naiman. 2003. Effects of land cover on stream ecosystems: roles of empirical 
models and scaling issues. Ecosystems 6:407 - 423. 
Sweeney, B. W. 1993. Effects of Streamside Vegetation on Macroinvertebrate 
Communities of White Clay Creek in Eastern North-America. Proceedings of the 
Academy of Natural Sciences of Philadelphia 144:291-340. 
Swift, C. C., C. R. Gilbert, S. A. Bortone, G. H. Burgess, and R. W. Yerger. 1986. 
Zoogeography of the freshwater fishes of the southeastern United States: 
Savannah River to Lake Pontchartrain. Pages 213-255 in C. H. Hocutt and E. O. 
Wiley, editors. Zoogeography of North American Fishes, Wiley-Interscience, 
New York. 
Tewfik, A., J. B. Rasmussen, and K. S. McCann. 2005. Anthropogenic enrichment alters 
a marine benthic food web. Ecology 86:2726-2736. 
Thompson, R. M., and C. R. Townsend. 2005. Energy availability, spatial heterogeneity 
and ecosystem size predict food-web structure in streams. Oikos 108:137-148. 
Ulseth, A. J., and A. E. Hershey. 2005. Natural abundances of stable isotopes trace 
anthropogenic N and C in an urban stream. Journal of the North American 
Benthological Society 24:270-289. 
Vander Zanden, M. J., G. Cabana, and J. B. Rasmussen. 1997. Comparing trophic 
position of freshwater fish calculated using stable nitrogen isotope ratios (delta N-
15) and literature dietary data. Canadian Journal of Fisheries and Aquatic 
Sciences 54:1142-1158. 
 
165 
Vander Zanden, M. J., and W. W. Fetzer. 2007. Global patterns of aquatic food chain 
length. Oikos 116:1378-1388. 
Vander Zanden, M. J., and J. B. Rasmussen. 1999. Primary consumer d
13
C and d
15
N and 
the trophic position of aquatic consumers. Ecology 80:1395-1404. 
Vander Zanden, M. J., B. J. Shuter, N. Lester, and J. B. Rasmussen. 1999. Patterns of 
food chain length in lakes: a stable isotope study. The American Naturalist 
154:406-416. 
Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell, and C. E. Cusing. 1980. 
The river continuum concept. Canadian Journal of Fisheries and Aquatic Science 
37:130-137. 
Vinson, M. R., and C. P. Hawkins. 1996. Effects of sampling area and subsampling 
procedure on comparisons of taxa richness among streams. Journal of the North 
American Benthological Society 15:392-399. 
Walsh, C. J., A. H. Roy, J. W. Feminella, P. D. Cottingham, P. M. Groffman, and R. P. 
Morgan. 2005. The urban stream syndrome: current knowledge and the search for 
a cure. Journal of the North American Benthological Society 24:706-723. 
Wang, L., and J. Lyons. 2003. Fish and benthic macroinvertebrate assemblages as 
indicators of stream dergadation in urbanizing watersheds. Pages 227-249 in T. P. 
Simon, editor. Biological response signatures: Indicator patterns using aquatic 
communities. CRC Press, New York. 
Wernick, B. G., K. E. Cook, and H. Schreier. 1998. Land use and streamwater nitrate-N 
dynamics in an urban-rural fringe watershed. Journal of the American Water 
Resources Association 34:639-650. 
 
166 
Wiggins, G. B. 1996. Larvae of the North American caddisfly genera (Trichoptera). 
University of Toronto Press, Buffalo, NY. 
Woodward, G., and A. G. Hildrew. 2002. Food web structure in riverine landscapes. 
Freshwater Biology 47:777-798. 
Wootton, J. T., M. S. Parker, and M. E. Power. 1996. Effects of disturbance on river food 
webs. Science 273:1558-1561. 
Wootton, J. Y. 1994. Predicting direct and indirect effects: An integrated approach using 
experiments and path analysis. Ecology 75:151-165. 
   T
a
bl
e 1.
  L
a
n
d
 co
v
e
r
 an
d p
h
y
s
i
c
a
l
 c
h
ar
act
er
s
 o
f
 s
t
ud
y
 wat
e
r
s
h
e
d
s
.
 
Wat
er
s
h
ed
 
i
s
 th
e
 w
a
te
r
s
h
e
d
 s
i
z
e
,
 
St
r
e
am
 
= 
m
e
a
n
 dept
h
 x
 w
i
dt
h
 
f
o
r
 
t
h
e s
t
r
eam
 dr
a
i
ni
ng t
h
e wat
e
r
s
h
e
d,
 %
IS
 = 
%
 i
m
p
er
vi
o
u
s
 s
u
r
f
ace co
v
e
r
,
 %
Pa
st
u
r
e
 
= %
 pas
t
ur
e
 c
o
v
e
r
,
 
an
d %
 
Fo
r
e
st
 =
  % f
o
re
s
t
 
co
v
e
r
,
 an
d 
LU/
L
C
 = do
m
i
n
a
n
t
 l
a
n
d
 us
e/
l
a
n
d
 co
v
e
r
 i
n
 wat
e
r
s
hed.
   
       
I
D
 Sit
e
 
St
r
e
am
 
Wat
er
s
h
ed 
(
k
m
2
) 
Str
e
am
 (m
2
) 
% I
S
 
% P
a
s
t
u
r
e
 
% F
o
re
s
t
 
L
U
/
L
C
 
1 SB
1 
Schl
e
y
 
C
r
eek 
20.
1 
0.
40 
2 
20 
74 
Dev
e
l
o
p
i
ng 
2 
SB
2 
St
an
d
i
n
g
 B
o
y
 C
r
eek T
r
i
b
.
 
6.
3 
0.
22 
3 
20 
73 
Dev
e
l
o
p
i
ng 
3 
SB
4 
St
an
d
i
n
g
 B
o
y
 C
r
eek 
26.
6 
0.
71 
3 
28 
64 
Dev
e
l
o
p
i
ng 
4 B
L
N 
B
l
a
n
t
o
n
 
C
r
eek 
3.
6 
0.
25 
1 
19 
76 
F
o
r
e
s
t
 
5 M
O
 
C
l
i
n
e
'
s
 
B
r
anc
h
 
9.
0 
0.
37 
2 
13 
81 
F
o
r
e
s
t
 
6 M
U
3 
T
u
r
n
t
i
m
e
 
B
r
a
n
c
h
 
10.
4 
0.
58 
2 
15 
78 
F
o
r
e
s
t
 
7 
F
S
2 
Wil
d
cat
 C
r
eek T
r
i
b
.
 
14.
5 
0.
64 
3 
36 
59 
Pas
t
ur
e 
8 
F
S
3 
Wil
d
cat
 C
r
eek T
r
i
b
.
 
3.
0 
0.
85 
3 
34 
62 
Pas
t
ur
e 
9 HC
2 
Ho
us
e 
C
r
eek 
14.
1 
0.
53 
2 
44 
52 
Pas
t
ur
e 
10 B
U
1 
L
i
n
d
s
e
y
 
C
r
eek 
25.
5 
2.
00 
40 
23 
34 
Ur
b
a
n
 
11 B
U
2 
C
o
o
p
er
 
C
r
eek 
24.
7 
0.
98 
25 
25 
46 
Ur
b
a
n
 
12 R
B
 
R
o
ar
i
n
g 
B
r
a
n
c
h
 
3.
7 
0.
68 
30 
27 
39 
Ur
b
a
n
 
167 
  T
a
bl
e 2.
  Res
o
u
r
ce,
 en
vi
r
o
nm
e
n
t
a
l
,
 an
d s
i
ze var
i
a
b
l
e
s
 us
ed 
i
n
 a
n
a
l
y
s
e
s
,
 t
h
e
i
r
 r
a
n
g
e o
f
 
v
a
l
u
es
,
 a
n
d t
h
e
i
r
 P
ear
s
o
n
 co
r
r
el
at
i
o
ns
 t
o 
pr
edo
m
i
n
a
n
t
 l
a
n
d
 co
v
e
r
 c
l
as
s
e
s
 
i
n
 t
h
e 12 wat
e
r
s
h
e
ds
.
  %
IS
 
=
 p
r
op
o
r
ti
on
 of
 
i
m
pe
r
v
i
ous
 s
u
r
f
a
c
e
 c
o
v
e
r
,
 %
Fo
r
e
st
 =
 p
r
op
or
ti
on
 of
 
f
o
r
e
s
t
 co
v
e
r
.
  
T
h
er
e wer
e
 n
o
 s
i
g
n
i
f
i
c
a
n
t
 cor
r
e
l
a
t
i
o
n
s
 b
e
t
w
ee
n
 any
 
v
a
r
i
a
b
l
e
s
 a
n
d pr
o
p
o
r
t
i
o
n
 o
f
 pas
t
ur
e 
co
v
e
r
 i
n
 t
h
e wat
e
r
s
h
e
d
s
.
 
 
*  p
 < 0.
05
,
 **
 p <
 0.
01
.
 
   
  
 
V
a
r
i
a
b
l
e
 ID
 
D
e
sc
r
i
p
t
i
o
n
 
R
a
n
g
e
  
%
 IS
  
%
 Fo
r
e
st
 
R
e
s
our
ce 
  
 
 
I
n
v
e
r
t
 Bi
o
m
a
s
s
 
M
e
a
n
 
m
a
cr
o
i
nv
er
t
e
b
r
at
e bi
o
m
a
s
s
 (
g
 /
 m
2
)
 
11.
2 ?
 51.
5 
 
 
C
h
l
o
r
o
phyll
 
a 
Me
a
n
 p
e
r
i
p
h
y
t
o
n
 b
i
o
m
a
ss (
g
 /
 m
2
)
 
0.
03 ?
 1.
97
  
0.
62* 
 
OM
  
B
e
n
t
hi
c o
r
gani
c 
m
a
t
t
e
r
 
(
g
)
 
  0
.
3 ?
 1.
2 
 
 
Habi
t
a
t
 I
n
dex
 
Habi
t
a
t
 as
s
e
s
s
m
e
n
t
 i
n
de
x
 s
c
o
r
e  
  
  54
.
2
 ?
 125.
5 
 
0.
48* 
E
n
vi
r
onment
a
l
 
  
 
 
Te
m
p
 
M
e
d
i
a
n
 wat
e
r
 
t
e
m
p
er
at
ur
e (
o
C
)
 
13.
3 ?
 15.
6 
0.
84** 
-
0
.
59* 
T
D
S
 
Me
a
n
 t
o
t
a
l
 d
i
sso
l
v
e
d
 so
l
i
d
s c
o
n
c
e
n
t
r
a
t
i
o
n
 (
m
g
 L
-1
)
 
19.
9 ?
 58.
6 
0.
65* 
 
T
S
S
 
Me
a
n
 t
o
t
a
l
 su
sp
e
n
d
e
d
 so
l
i
d
s c
o
n
c
e
n
t
r
a
t
i
o
n
 (
m
g
 L
-1
)
 
2.
1 ?
 8.
1 
 
 
Ma
x
Q
 
Ma
x
i
m
u
m
 d
i
sc
h
a
r
g
e
 (
L
 s
-1
)
 
0.
54 ?
 21.
98 
 
 
7x
M
e
d 
# 
o
f
 t
i
m
e
s
 d
i
s
c
h
a
r
g
e e
x
ceeded 7
x
 
m
e
d
i
a
n
 
f
l
o
w
 
1 ? 64 
0.
63* 
 
Siz
e
 
  
 
 
W
a
t
e
rsh
e
d
 
W
a
t
e
rsh
e
d
 si
z
e
 (k
m
2
)
 
3.
0 ?
 26.
6 
 
 
S
t
r
e
am
 
M
e
an
 wi
d
t
h
 
x
 
m
e
an
 d
e
p
t
h
 of
 s
t
r
e
am
 (
m
2
)
 
0.
34 ?
 1.
20 
0.
67* 
-
0
.
77** 
 
168 
 
169 
Table 3.  Taxa collected for stable isotope analysis over the 12 watersheds.  Group refers 
to broad functional grouping used in construction of food webs. 
 
       
Taxon Name Family Order Group 
benthic algae - - Basal/Terrestrial
organic matter - - Basal/Terrestrial
ant Formicidae Hymenoptera Basal/Terrestrial
Anchytarsus Elmidae Coleoptera Invertebrate 
Optioservus Elmidae Coleoptera Invertebrate
Cyclopoida - Cyclopoida Invertebrate
Cambarus latimanus Cambaridae Decapoda Invertebrate 
Procambarus versutus Cambaridae Decapoda Invertebrate 
Non -Tanypod chironomid Chironomidae Diptera Invertebrate 
Tanypod chironomid Chironomidae Diptera Invertebrate 
Tabanus Tabanidae Diptera Invertebrate
Hexatoma Tipulidae Diptera Invertebrate 
Tipula Tipulidae Diptera Invertebrate
Baetis Baetidae Ephemeroptera Invertebrate 
Ephemerella Ephemerellidae Ephemeroptera Invertebrate 
Hexagenia Ephemeridae Ephemeroptera Invertebrate 
Stenonema Heptageniidae Ephemeroptera Invertebrate 
Isonychia Isonychiidae Ephemeroptera Invertebrate 
Tricorythodes Tricorythidae Ephemeroptera Invertebrate 
Corydalus Corydalidae Megaloptera Invertebrate 
Boyeria Aeshnidae Odonata Invertebrate
Calopteryx Calopterygidae Odonata Invertebrate 
Cordulegaster Cordulegastridae Odonata Invertebrate 
Macromia Corduliidae Odonata Invertebrate
Gomphus Gomphidae Odonata Invertebrate 
Allocapnia Capniidae Plecoptera Invertebrate
Acroneuria Perlidae Plecoptera Invertebrate 
Physella Physidae Pulmonata Invertebrate
Cheumatopsyche Hydropsychidae Trichoptera Invertebrate 
Pycnopsyche Limnephilidae Trichoptera Invertebrate 
Lebertid mite Lebertidae Trombidiformes Invertebrate 
Corbicula fluminea Corbiculidae Veneroida Invertebrate 
 
 
 
 
 
 
 
 
 
170 
 
Table 3. continued. 
 
       
Taxon Name Family Order Group 
Bufo fowleri Bufonidae Anura Vertebrate 
Eurycea cirrigera Plethodontidae Caudata Vertebrate 
Hypentelium etowanum Catastomidae Cypriniformes Vertebrate 
Minytrema melanops Catastomidae Cypriniformes Vertebrate 
Campostoma pauciradii Cyprinidae Cypriniformes Vertebrate 
Luxilus zonistius Cyprinidae Cypriniformes Vertebrate 
Nocomis leptocephalus Cyprinidae Cypriniformes Vertebrate 
Notemigonus chrysoleucuas Cyprinidae Cypriniformes Vertebrate 
Notropis baileyi Cyprinidae Cypriniformes Vertebrate 
Notropis buccata Cyprinidae Cypriniformes Vertebrate 
Notropis longirostis Cyprinidae Cypriniformes Vertebrate 
Notropis texanus Cyprinidae Cypriniformes Vertebrate 
Semotilus atromaculatus Cyprinidae Cypriniformes Vertebrate 
Lepomis auritus Centrarchidae Perciformes Vertebrate 
Lepomis cyanellus Centrarchidae Perciformes Vertebrate 
Lepomis gulosus Centrarchidae Perciformes Vertebrate 
Lepomis macrochirus Centrarchidae Perciformes Vertebrate 
Micropterus salmoides Centrarchidae Perciformes Vertebrate 
Percina nigrofasciata Percidae Perciformes Vertebrate 
Ameiurus brunneus Ictaluridae Siluriformes Vertebrate 
Ameiurus natalis Ictaluridae Siluriformes Vertebrate 
Ameiurus nebulosus Ictaluridae Siluriformes Vertebrate 
Noturus sp. Ictaluridae Siluriformes Vertebrate 
  
 
 
 
  T
a
bl
e 4.
  Num
b
er
 o
f
 d
i
f
f
e
r
e
n
t
 t
a
x
a
 co
l
l
e
ct
ed f
r
o
m
 eac
h
 F
u
n
c
t
i
o
n
a
l
 Gr
o
up 
i
n
 eac
h
 wat
e
r
s
h
e
d.
  F
o
r
 i
nver
t
eb
r
a
t
e
s
,
 CF
 = co
l
l
ect
or
/
f
i
l
et
er
s
,
 
C
G
 =
 c
o
l
l
e
c
t
o
r
/g
a
t
h
e
re
rs
, PA
 =
 p
a
ra
si
t
e
s
,
 PR
 =
 p
r
e
d
a
t
o
r
s
,
 SC
 =
 s
c
ra
p
e
rs
,
 SH
 =
 s
h
re
d
d
e
rs
, a
n
d
 SV
 =
 s
c
a
v
e
n
g
e
rs
 (c
ra
y
f
i
s
h
)
.  F
o
r
 
v
e
rt
e
b
ra
t
e
s
,
 H
B
V
 =
 h
e
rb
i
v
o
r
e
s
, IS
V
 =
 i
n
s
e
c
t
i
v
o
r
e
s
, O
M
N
 =
 
o
m
n
i
v
o
re
s
,
 P
S
V
 =
 p
i
s
c
i
v
o
r
e
s
, S
A
L
 =
 s
a
l
a
m
a
n
d
e
r
, TO
A
D
 =
  t
o
a
d
. 
 S
i
t
e
 
co
des
 ar
e as
 i
n
 T
a
bl
e 1.
  Onl
y
 
i
nver
t
eb
r
a
t
e
s
 
t
h
at
 c
o
u
l
d be 
i
d
e
n
t
i
f
i
e
d
 to
 f
a
m
ily
-
l
e
v
e
l
 w
e
r
e
 
i
n
c
l
uded.
 
  
 
  
  
  
  
  
  
  
  
  
  
  
  
 
  
S
I
TE LU/LC 
CF
 
CG
P
R
S
C
S
H
S
V
HB
V
I
S
V
OM
N
 
P
S
V
S
AL
TOAD 
TOTAL
SB
1 D
e
ve
l
o
ping 
1 
4 
3 
0 1 1 
1 
5 
1 
0 
1 
0 
18 
SB
2 D
e
ve
l
o
ping 
0 
3 
1 
1 1 0 
0 
4 
1 
1 
0 
0 
12 
SB
4 D
e
ve
l
o
ping 
0 
3 
3 
1 0 1 
0 
6 
2 
2 
1 
0 
19 
BL
N
 F
o
r
e
s
t
 
2 
3 
6 
1 2 2 
1 
9 
2 
1 
1 
0 
30 
MO
 F
o
r
e
s
t
 
1 
4 
6 
1 1 1 
1 
7 
3 
1 
1 
0 
25 
M
U
3 F
o
r
e
s
t
 
2 
3 
4 
1 0 1 
0 
5 
3 
2 
0 
0 
21 
FS
2
 Pa
st
u
r
e
 
0 
2 
3 
1 1 2 
1 
9 
2 
0 
1 
0 
22 
FS
3
 
Pa
st
u
r
e
 
1 
3 
3 
1 0 2 
1 
4 
3 
0 
1 
0 
19 
HC2 P
a
s
t
u
r
e
 
0 
3 
3 
1 0 2 
0 
7 
2 
0 
0 
0 
18 
B
U
1 U
r
ban 
0 
3 
4 
1 0 0 
1 
2 
1 
1 
0 
1 
14 
B
U
2 U
r
ban 
1 
2 
2 
0 0 0 
0 
5 
0 
1 
1 
0 
12 
RB
 Ur
b
a
n
 
2 
2 
2 
1 0 0 
0 
1 
0 
0 
0 
0 
8 
    
171 
 
172 
Table 5.  Stable isotope results for 12 watersheds.  ?
13
C and ?
15
N values are site averages, 
?TP is mean trophic position, FCL is maximum trophic position (food chain length), 
Top Consumer is the organism with the maximum trophic position, and FG is the 
functional group of the top consumer (see Table 4 for abbreviation explanations) 
 
            
Site ? ?
13
C (?) ? ?
15
N (?) ? TP FCL Top consumer FG 
SB1 -31.043 5.232 1.645 3.024 L. macrochirus ISV 
SB2 -29.677 5.195 1.769 2.985 L. auritus ISV 
SB4 -29.348 8.655 1.9 2.97 P. nigrofasciata ISV 
BLN -28.17 6.192 1.833 3.675 P. nigrofasciata ISV 
MO -29.436 5.141 2.265 4.338 L. macrochirus ISV 
MU3 -29.262 6.912 1.815 3.253 P. nigrofasciata ISV 
FS2 -27.639 7.361 2.247 3.393 M. melanops ISV 
FS3 -28.61 6.895 1.726 2.934 H. etowanum ISV 
HC2 -27.116 6.616 2.356 3.525 A. natalis OMN 
BU1 -23.19 7.126 1.456 2.741 L. auritus ISV 
BU2 -26.538 8.842 1.676 2.623 L. auritus ISV 
RB -28.195 4.233 1.373 3.015 L. macrochirus ISV 
      
 
 
 
 
 
173 
Table 6.  Significant correlation coefficients between stable isotope results and land 
cover, resource, environmental and system size.  Variable definitions are as in Tables 1 ? 
2  *  p < 0.05, ** p < 0.01. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
          
Variable ? ?
13
C ? ?
15
N ? TP FCL 
% IS 0.75**  -0.67* -0.58* 
% Pasture     
% Forest  -0.78**   0.58* 
Invert Biomass   -0.60*  
Chlorophyll a   -0.68* 
Temp   -0.81** -0.65* 
TDS  -0.73** -0.66* 
TSS   0.70*   
7xMed  -0.71** -0.63* 
Habitat    0.75** 
Watershed   0.65*   
Stream 0.76**  
     
 
174 
Figure 1.  The study area included 12 small watersheds (shaded sections) of the 
Chattahoochee River Basin in 4 counties in the Lower Southern Piedmont ecoregion of 
western Georgia, USA.  Numbers refer to watersheds and correspond to Table 1. The city 
of Columbus is located in western Muscogee County.  County names are in capital 
letters. 
 
 
Columbus
2 0 020Kilom eters 
 
?
3
6
4 
4 
2
9
1
8
7
5
10
12
11
MUSCOGEE 
HARRIS 
TROUP MERIWETHER 
Chattahoochee
River
 
175 
Figure 2.  Relationship between mean trophic position and taxa richness and maximum 
food chain length and taxa richness in 12 streams. 
 
 
Taxa Richness
5 1015202530
Food
 C
hai
n
 Le
ngt
h
2.5
3.0
3.5
4.0
4.5
R
2
 = 0.57
p = 0.004
5 1015202530
M
ean Trop
hi
c
 
P
o
s
i
ti
o
n
1.0
1.5
2.0
2.5
R
2
 = 0.36
p = 0.040
   
 
176 
 
Figure 3.  Standard errors of mean ?
13
C and ?
15
N functional group values for 4 land use 
types. 
 
 
 
? 
13
C (SE)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
? 
15
N (S
E
)
0.0
0.3
0.6
0.9
1.2
1.5
Forest
Pasture
Developing
Urban
 
 
 
 
177 
Figure 4.  Mean standard deviations for averaged ?
13
C and ?
15
N values in each LU/LC 
category.  There is no significant difference in the variation between categories. 
 
 
Mixed Urban Pasture Developing
M
ean
 Stan
dard
 D
e
v
i
ati
o
n
0
1
2
3
4
5
? 
15
N
? 
13
C
 
 
  F
i
gur
e 5.
  Bi
p
l
o
t
s
 
o
f
 a
v
er
age 
?
13
C an
d
 
?
15
N
 (
+
 1
 S
E
)
 v
a
l
u
es
 
f
o
r
 f
u
nct
i
o
n
a
l
 gr
o
ups
 f
o
r
 4 l
a
n
d
 us
e cat
e
go
r
i
es
 (
3
 s
t
r
eam
s
 per
 cat
eg
or
y
)
.
  A
 
g
i
v
e
n
 
f
u
n
c
t
i
o
n
a
l
 gr
o
up h
a
d 
t
o
 b
e
 pr
es
en
t
 i
n
 at
 l
eas
t
 2 
o
f
 t
h
e 3 s
t
r
eam
s
 per
 cat
eg
o
r
y
 t
o
 b
e
 i
n
c
l
ud
ed.
  Da
r
k
 gr
ey
 va
l
u
es
 ar
e v
e
r
t
eb
r
a
t
e
s
,
 
l
i
g
h
t
 gr
ey
 ar
e 
i
n
v
e
r
t
eb
r
a
t
e
s
,
 an
d bl
ack 
v
a
l
u
es
 ar
e bas
a
l
 /
 t
e
r
r
e
s
t
r
i
a
l
 r
e
s
o
ur
ces
.
  
E
r
r
o
r
 b
a
r
s
 ar
e pr
es
en
t
 b
u
t
 
o
b
s
c
ur
ed o
n
 s
o
m
e
 
po
i
n
t
s
.
 
Fo
r
e
s
t
? 
15
N
-2
02468
101214
S
a
l
a
ma
nd
er
P
i
scivo
r
e
He
rb
i
v
o
r
e
Om
n
i
vo
r
e
I
n
se
ct
i
v
o
r
e
Co
lle
ct
o
r
 /
 
Fi
lt
e
r
e
r
C
o
l
l
e
c
t
o
r
 /
 
G
a
t
h
er
er
P
r
ed
at
o
r
Sc
r
a
pe
r
S
h
re
dd
er
Cra
y
f
i
s
h
PO
M
T
e
r
r
e
s
tr
i
a
l
 In
s
e
c
t
Al
g
a
e
OM
D
e
v
e
l
opi
ng
? 
13
C
-
3
6-
3
4
-
3
2-
3
0
-
2
8-
2
6
-
2
4-
2
2
-
2
0-
1
8
? 
15
N
-2
02468
101214
Pa
s
t
u
r
e
U
r
ban
? 
13
C
-
3
6-
3
4
-
3
2-
3
0
-
2
8-
2
6
-
2
4-
2
2
-
2
0-
1
8
178 
 
179 
Figure 6.  Arrow diagrams of direction and magnitude of change for consumer groups that 
correspond between compared food webs.  Each diagram represents a comparison of 1 
mean food web to another.  The length of the arrows represent the magnitude of change 
for each group shared in both food webs, concentric circles correspond to the magnitude 
of change, dashed lines are the mean vector of change, and curved dashed lines are 95% 
confidence intervals of the vector of change. 
 
 
 
 
 
 
5 5
5
5
4 4
4
4
3 3
3
3
2 2
2
2
1
1
1
1
0?
90?
180?
270? 
A.  Forest to Pasture 
8 8
8
6 6
6
6
4 4
4
4
2
2
2
90?
180?
270? 
0?
B.  Forest to Developing 
6 6
6
6
4 4
4
4
2 2
2
2
90?
180?
270? 
C.  Forest to Urban 
0?
 
180 
APPENDIX.  Watershed and biotic summaries for each watershed sampled during the 
studies.  AFDM:  Ash-free dry mass, Diversity:  Shannon?s H?, GA-BMI: Georgia 
Benthic Macroinvertebrate Index, GA-IBI:  Georgia Index of Biotic Integrity 
 
 
Watershed ID BC    
Watershed Area 647 ha    
Tributary Name Beech Creek    
Stream Order 2  
UTM Zone 16 Coordinates 16 N 703868 3657575  
Seasons Sampled 6    
Total Macroinvertebrate Richness 145  
Total Fish Richness 15    
          
Parameter Mean Max Min SE
Algae     
AFDM (g / m
2
) 0.45 0.04 1.08 0.25 
Chlorophyll a (mg / m
2
) 0.32 0.01 0.64 0.11 
     
Macroinvertebrates 
Biomass (g / m
2
) 9.72 0.42 20.59 2.48 
Diversity (H') 2.64 2.44 3.12 0.09 
Richness 42.50 22.00 59.00 5.20 
GA-BMI 53.33 32.00 80.00 7.22 
     
Fish 
Diversity (H') 1.30 1.03 1.57 0.10 
Richness 9.00 6.00 11.00 0.84 
GA-IBI 40.80 32.00 44.00 2.33 
 
 
 
 
181 
 
Watershed ID BLN    
Watershed Area 364 ha    
Tributary Name Blanton Creek   
Stream Order 1  
UTM Zone 16 Coordinates 16 N 677570 3625622  
Seasons Sampled 3   
Total Macroinvertebrate Richness 92   
Total Fish Richness 18   
          
Parameter Mean Min Max SE 
Algae    
AFDM (g / m
2
) 0.02 0.02 0.02 0.00 
Chlorophyll a (mg / m
2
) 0.70 0.47 0.92 0.23 
     
Macroinvertebrates 
Biomass (g / m
2
) 31.43 9.82 62.04 12.85 
Diversity (H') 2.74 2.59 2.97 0.10 
Richness 56.67 51.00 64.00 3.84 
GA-BMI 56.00 41.00 75.00 10.02 
     
Fish 
Diversity (H') 2.09 1.94 2.21 0.08 
Richness 12.33 11.00 14.00 0.88 
GA-IBI 36.00 34.00 38.00 1.15 
     
 
 
182 
 
Watershed ID BSB    
Watershed Area 697 ha    
Tributary Name Blue Springs Branch   
Stream Order 1   
UTM Zone 16 Coordinates 16 N  690427 3621407  
Seasons Sampled 3   
Total Macroinvertebrate Richness 90    
Total Fish Richness 16  
          
Parameter Mean Min Max SE 
Algae     
AFDM (g / m
2
) 0.82 0.05 1.56 0.44 
Chlorophyll a (mg / m
2
) 0.71 0.51 0.95 0.13 
     
Macroinvertebrates 
Biomass (g / m
2
) 15.95 3.20 23.57 5.24 
Diversity (H') 2.51 1.74 3.04 0.32 
Richness 48.67 23.00 63.00 12.86 
GA-BMI 53.67 46.00 58.00 3.84 
     
Fish 
Diversity (H') 1.95 1.87 2.04 0.05 
Richness 10.67 9.00 12.00 0.88 
GA-IBI 46.67 46.00 48.00 0.67 
     
 
 
183 
 
Watershed ID BU1    
Watershed Area 2546 ha    
Tributary Name Lindsey Creek   
Stream Order 2  
UTM Zone 16 Coordinates 16N 693323 3592891  
Seasons Sampled 6   
Total Macroinvertebrate Richness 70   
Total Fish Richness 12   
          
Parameter Mean Min Max SE 
Algae     
AFDM (g / m
2
) 0.58 0.00 1.49 0.36 
Chlorophyll a (mg / m
2
) 0.61 0.11 1.82 0.41 
     
Macroinvertebrates 
Biomass (g / m
2
) 40.58 13.70 113.14 13.96 
Diversity (H') 1.82 1.43 2.31 0.12 
Richness 20.33 15.00 29.00 1.98 
GA-BMI 28.67 10.00 43.00 4.70 
     
Fish 
Diversity (H') 1.17 0.23 1.69 0.26 
Richness 5.40 2.00 8.00 0.98 
GA-IBI 32.00 26.00 36.00 1.90 
     
 
 
184 
 
Watershed ID BU2    
Watershed Area 2469 ha    
Tributary Name Cooper Creek   
Stream Order 2  
UTM Zone 16 Coordinates 16 N 695222 3595956  
Seasons Sampled 6   
Total Macroinvertebrate Richness 83   
Total Fish Richness 20   
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 0.51 0.02 1.16 0.29 
Chlorophyll a (mg / m
2
) 0.41 0.05 0.93 0.18 
     
Macroinvertebrates 
Biomass (g / m
2
) 34.43 14.15 55.97 5.83 
Diversity (H') 1.72 1.24 2.37 0.15 
Richness 23.50 19.00 30.00 1.67 
GA-BMI 24.83 7.00 35.00 4.18 
     
Fish 
Diversity (H') 1.64 1.28 1.77 0.09 
Richness 10.00 9.00 11.00 0.45 
GA-IBI 36.40 32.00 40.00 1.83 
     
 
 
 
185 
 
Watershed ID FPBW    
Watershed Area 489 ha
Tributary Name Five Points Branch    
Stream Order 2    
UTM Zone 16 Coordinates 16 N 689255 3621326  
Seasons Sampled 3    
Total Macroinvertebrate Richness 104 
Total Fish Richness 10    
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 0.85 0.03 1.58 0.45 
Chlorophyll a (mg / m
2
) 0.81 0.47 1.00 0.17 
     
Macroinvertebrates 
Biomass (g / m
2
) 26.23 9.66 42.23 7.68 
Diversity (H') 2.34 1.76 3.04 0.31 
Richness 57.67 29.00 74.00 14.38 
GA-BMI 51.67 48.00 59.00 3.67 
     
Fish 
Diversity (H') 1.83 1.72 1.96 0.07 
Richness 8.00 7.00 9.00 0.58 
GA-IBI 44.67 44.00 46.00 0.67 
     
 
 
186 
 
Watershed ID FR    
Watershed Area 2396 ha 
Tributary Name Flat Rock    
Stream Order 2 
UTM Zone 16 Coordinates 16 N 699161 3599824  
Seasons Sampled 1    
Total Macroinvertebrate Richness 26 
Total Fish Richness 8    
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) - - - - 
Chlorophyll a (mg / m
2
) - - - - 
     
Macroinvertebrates 
Biomass (g / m
2
) 61.24 61.24 61.24 - 
Diversity (H') 2.20 2.20 2.20 - 
Richness 26.00 26.00 26.00 - 
GA-BMI 24.00 24.00 24.00 - 
    
Fish  
Diversity (H') 1.46 1.46 1.46 - 
Richness 8.00 8.00 8.00 - 
GA-IBI 38.00 38.00 38.00 - 
     
 
 
 
187 
 
Watershed ID FS1    
Watershed Area 2420 ha 
Tributary Name Wildcat Creek   
Stream Order 2    
UTM Zone 16 Coordinates 16 N 684091 3641414  
Seasons Sampled 3    
Total Macroinvertebrate Richness 66 
Total Fish Richness 15    
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 1.00 0.04 2.87 0.94 
Chlorophyll a (mg / m
2
) 0.38 0.03 0.88 0.25 
     
Macroinvertebrates 
Biomass (g / m
2
) 15.98 8.19 20.96 3.22 
Diversity (H') 2.20 1.57 2.78 0.29 
Richness 37.00 24.00 46.00 6.66 
GA-BMI 44.67 38.00 50.00 3.53 
     
Fish 
Diversity (H') 1.93 1.73 2.13 0.12 
Richness 11.00 7.00 15.00 2.31 
GA-IBI 45.33 42.00 48.00 1.76 
     
 
 
 
188 
 
Watershed ID FS2    
Watershed Area 1449 ha    
Tributary Name Wildcat Creek   
Stream Order 2  
UTM Zone 16 Coordinates 16 N 685946 3639085  
Seasons Sampled 3   
Total Macroinvertebrate Richness 55   
Total Fish Richness 15   
          
Parameter Mean Min Max SE 
Algae    
AFDM (g / m
2
) 0.00 0.00 0.00 0.00 
Chlorophyll a (mg / m
2
) 0.01 0.01 0.01 0.00 
     
Macroinvertebrates 
Biomass (g / m
2
) 11.17 4.85 18.09 3.13 
Diversity (H') 1.93 1.40 2.61 0.29 
Richness 33.67 31.00 35.00 1.33 
GA-BMI 40.00 25.00 66.00 13.05 
     
Fish 
Diversity (H') 1.96 1.90 2.06 0.05 
Richness 10.67 10.00 11.00 0.33 
GA-IBI 37.33 34.00 42.00 2.40 
     
 
 
189 
 
Watershed ID FS3    
Watershed Area 296 ha    
Tributary Name Wildcat Creek   
Stream Order 1  
UTM Zone 16 Coordinates 16 N 685956 3640196  
Seasons Sampled 3   
Total Macroinvertebrate Richness 66   
Total Fish Richness 16   
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 0.62 0.04 1.20 0.58 
Chlorophyll a (mg / m
2
) 1.03 1.02 1.03 0.00 
     
Macroinvertebrates 
Biomass (g / m
2
) 41.14 24.45 65.65 10.22 
Diversity (H') 2.37 2.31 2.42 0.03 
Richness 41.00 36.00 51.00 5.00 
GA-BMI 41.33 36.00 52.00 5.33 
     
Fish 
Diversity (H') 2.00 1.87 2.08 0.06 
Richness 11.00 10.00 12.00 0.58 
GA-IBI 30.67 30.00 32.00 0.67 
     
 
 
190 
 
Watershed ID FS5    
Watershed Area 1183 ha    
Tributary Name Flat Shoals    
Stream Order 2   
UTM Zone 16 Coordinates 16 N 707467 3652031  
Seasons Sampled 3   
Total Macroinvertebrate Richness 84   
Total Fish Richness 9   
          
Parameter Mean Min Max SE 
Algae    
AFDM (g / m
2
) 0.96 0.03 2.73 0.89 
Chlorophyll a (mg / m
2
) 0.23 0.09 0.37 0.08 
     
Macroinvertebrates 
Biomass (g / m
2
) 8.95 7.30 10.61 0.96 
Diversity (H') 2.78 2.68 2.88 0.06 
Richness 59.00 56.00 62.00 2.45 
GA-BMI 31.50 22.00 41.00 7.76 
     
Fish 
Diversity (H') 1.17 0.80 1.41 0.19 
Richness 4.33 3.00 6.00 0.88 
GA-IBI 38.00 36.00 40.00 1.15 
     
 
 
 
191 
 
Watershed ID FS6    
Watershed Area 922 ha    
Tributary Name Flat Shoals    
Stream Order 2   
UTM Zone 16 Coordinates 16 N 707446 3651910  
Seasons Sampled 3    
Total Macroinvertebrate Richness 21  
Total Fish Richness 9    
          
Parameter Mean Min Max SE
Algae     
AFDM (g / m
2
) 0.51 0.22 0.81 0.17 
Chlorophyll a (mg / m
2
) 0.23 0.06 0.48 0.13 
     
Macroinvertebrates 
Biomass (g / m
2
) 0.68 0.00 1.36 0.39 
Diversity (H') 0.84 0.00 1.68 0.48 
Richness 10.50 1.00 20.00 7.76 
GA-BMI 29.00 20.00 38.00 7.35 
     
Fish 
Diversity (H') 1.30 1.11 1.48 0.11 
Richness 5.33 4.00 6.00 0.67 
GA-IBI 33.33 30.00 36.00 1.76 
     
 
 
192 
 
Watershed ID HC    
Watershed Area 655 ha    
Tributary Name House Creek    
Stream Order 2  
UTM Zone 16 Coordinates 16 N 678141 3630775  
Seasons Sampled 5   
Total Macroinvertebrate Richness 115   
Total Fish Richness 17    
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 0.40 0.01 1.10 0.35 
Chlorophyll a (mg / m
2
) 0.08 0.04 0.10 0.02 
     
Macroinvertebrates 
Biomass (g / m
2
) 16.56 1.66 41.87 6.91 
Diversity (H') 2.11 1.13 2.70 0.24 
Richness 35.20 30.00 41.00 1.85 
GA-BMI 45.00 33.00 61.00 6.19 
     
Fish 
Diversity (H') 1.63 1.17 2.03 0.20 
Richness 7.75 6.00 10.00 0.85 
GA-IBI 37.50 28.00 48.00 4.27 
     
 
 
193 
 
Watershed ID HC2    
Watershed Area 655 ha    
Tributary Name House Creek    
Stream Order 3  
UTM Zone 16 Coordinates 16 N 683811 3634666  
Seasons Sampled 3   
Total Macroinvertebrate Richness 52    
Total Fish Richness 17  
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 0.01 0.00 0.02 0.01 
Chlorophyll a (mg / m
2
) 0.22 0.03 0.40 0.18 
     
Macroinvertebrates 
Biomass (g / m
2
) 6.96 1.38 15.83 3.66 
Diversity (H') 1.69 0.81 2.56 0.41 
Richness 31.67 25.00 40.00 4.41 
GA-BMI 33.33 23.00 39.00 5.17 
     
Fish 
Diversity (H') 1.94 1.79 2.01 0.08 
Richness 12.33 12.00 13.00 0.33 
GA-IBI 40.67 40.00 42.00 0.67 
     
 
 
 
194 
 
Watershed ID MK    
Watershed Area 663 ha    
Tributary Name McKoon Creek   
Stream Order 2  
UTM Zone 16 Coordinates 16 N 703701 3668559  
Seasons Sampled 6   
Total Macroinvertebrate Richness 126   
Total Fish Richness 20   
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 0.52 0.01 1.06 0.30 
Chlorophyll a (mg / m
2
) 0.13 0.02 0.33 0.07 
     
Macroinvertebrates 
Biomass (g / m
2
) 5.87 1.91 10.97 1.39 
Diversity (H') 2.46 2.11 2.60 0.08 
Richness 37.83 28.00 44.00 2.46 
GA-BMI 48.67 26.00 60.00 5.05 
     
Fish 
Diversity (H') 1.94 1.65 2.24 0.11 
Richness 12.00 7.00 16.00 1.52 
GA-IBI 42.00 30.00 52.00 4.60 
     
 
 
 
195 
 
Watershed ID MO    
Watershed Area 897  
Tributary Name Cline's Branch   
Stream Order 2  
UTM Zone 16 Coordinates 16 N 680997 362509  
Seasons Sampled 6   
Total Macroinvertebrate Richness 167   
Total Fish Richness 25   
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 0.82 0.01 1.94 0.48 
Chlorophyll a (mg / m
2
) 0.31 0.16 0.57 0.09 
     
Macroinvertebrates 
Biomass (g / m
2
) 19.71 3.25 55.01 7.82 
Diversity (H') 2.48 2.06 2.95 0.12 
Richness 47.83 40.00 57.00 2.94 
GA-BMI 53.83 40.00 67.00 4.71 
     
Fish 
Diversity (H') 1.78 1.43 2.12 0.11 
Richness 10.50 6.00 15.00 1.28 
GA-IBI 38.33 28.00 46.00 2.85 
     
 
 
 
196 
 
Watershed ID MU1    
Watershed Area 1178 ha    
Tributary Name Ossahatchie Creek   
Stream Order 3  
UTM Zone 16 Coordinates 16 N 712939 3615237  
Seasons Sampled 32    
Total Macroinvertebrate Richness 102  
Total Fish Richness 11    
          
Parameter Mean Min Max SE
Algae     
AFDM (g / m
2
) 0.62 0.00 1.60 0.49 
Chlorophyll a (mg / m
2
) 0.11 0.00 0.25 0.07 
     
Macroinvertebrates 
Biomass (g / m
2
) 32.55 6.79 75.78 12.25 
Diversity (H') 2.02 1.45 2.37 0.13 
Richness 28.00 22.00 35.00 1.86 
GA-BMI 33.00 23.00 49.00 3.82 
     
Fish 
Diversity (H') 1.03 0.17 2.01 0.33 
Richness 5.40 3.00 9.00 1.29 
GA-IBI 30.33 26.00 36.00 1.58 
     
 
 
 
197 
 
Watershed ID MU2    
Watershed Area 606 ha    
Tributary Name Mulberry Creek   
Stream Order 2  
UTM Zone 16 Coordinates 16 N 708913 3621461  
Seasons Sampled 5    
Total Macroinvertebrate Richness 90  
Total Fish Richness 14    
          
Parameter Mean Min Max SE
Algae     
AFDM (g / m
2
) 0.49 0.06 1.33 0.42 
Chlorophyll a (mg / m
2
) 0.10 0.05 0.16 0.03 
     
Macroinvertebrates 
Biomass (g / m
2
) 5.08 0.88 15.97 2.16 
Diversity (H') 2.06 1.06 2.47 0.20 
Richness 23.50 14.00 35.00 2.93 
GA-BMI 31.83 18.00 42.00 3.22 
     
Fish 
Diversity (H') 1.65 1.38 1.96 0.10 
Richness 6.80 5.00 10.00 0.97 
GA-IBI 38.40 30.00 44.00 3.06 
     
 
 
 
198 
 
Watershed ID MU3    
Watershed Area 1044 ha    
Tributary Name Turntime Branch   
Stream Order 2  
UTM Zone 16 Coordinates 16 N 701261 3618978  
Seasons Sampled 5    
Total Macroinvertebrate Richness 124  
Total Fish Richness 19    
          
Parameter Mean Min Max SE
Algae     
AFDM (g / m
2
) 0.39 0.02 1.14 0.37 
Chlorophyll a (mg / m
2
) 0.19 0.01 0.42 0.12 
     
Macroinvertebrates 
Biomass (g / m
2
) 32.40 7.27 111.39 16.41 
Diversity (H') 2.42 2.08 2.62 0.10 
Richness 40.20 32.00 45.00 2.31 
GA-BMI 50.20 40.00 58.00 3.46 
     
Fish 
Diversity (H') 1.81 1.47 2.08 0.13 
Richness 9.75 8.00 12.00 0.85 
GA-IBI 36.50 24.00 46.00 5.19 
     
 
 
 
199 
 
Watershed ID RB    
Watershed Area 367 ha    
Tributary Name Roaring Branch   
Stream Order 1  
UTM Zone 16 Coordinates 16 N 691357 3602110  
Seasons Sampled 6    
Total Macroinvertebrate Richness 71  
Total Fish Richness 13    
          
Parameter Mean Min Max SE
Algae     
AFDM (g / m
2
) 0.56 0.01 1.29 0.25 
Chlorophyll a (mg / m
2
) 0.39 0.07 0.65 0.10 
     
Macroinvertebrates 
Biomass (g / m
2
) 40.77 15.68 65.67 7.37 
Diversity (H') 1.86 1.25 2.47 0.15 
Richness 20.00 15.00 30.00 2.22 
GA-BMI 20.33 3.00 34.00 5.41 
     
Fish 
Diversity (H') 0.99 0.59 1.53 0.15 
Richness 6.00 5.00 8.00 0.45 
GA-IBI 27.67 22.00 36.00 2.65 
     
 
 
 
200 
 
Watershed ID RC    
Watershed Area 471 ha    
Tributary Name Rushing Creek   
Stream Order 1  
UTM Zone 16 Coordinates 16 N 688583 3600821  
Seasons Sampled 1    
Total Macroinvertebrate Richness 23  
Total Fish Richness 8    
          
Parameter Mean Min Max SE
Algae     
AFDM (g / m
2
) - - - - 
Chlorophyll a (mg / m
2
) - - - - 
     
Macroinvertebrates 
Biomass (g / m
2
) 101.55 101.55 101.55 - 
Diversity (H') 1.90 1.90 1.90 - 
Richness 23.00 23.00 23.00 - 
GA-BMI 9.00 9.00 9.00 - 
    
Fish  
Diversity (H') 1.47 1.47 1.47 - 
Richness 8.00 8.00 8.00 - 
GA-IBI 36.00 36.00 36.00 - 
    
 
 
 
201 
 
Watershed ID SB1    
Watershed Area 2009 ha    
Tributary Name Schley Creek   
Stream Order 3  
UTM Zone 16 Coordinates 16 N 685293 3608160  
Seasons Sampled 6    
Total Macroinvertebrate 
Richness 131    
Total Fish Richness 15  
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 0.53 0.01 1.97 0.48 
Chlorophyll a (mg / m
2
) 0.66 0.11 1.14 0.27 
     
Macroinvertebrates 
Biomass (g / m
2
) 13.37 1.33 38.81 5.26 
Diversity (H') 2.40 2.05 2.70 0.09 
Richness 38.33 27.00 48.00 3.54 
GA-BMI 41.17 27.00 51.00 3.98 
     
Fish 
Diversity (H') 1.48 1.04 1.76 0.11 
Richness 6.83 3.00 12.00 1.22 
GA-IBI 34.00 22.00 46.00 3.76 
     
 
 
 
202 
 
Watershed ID SB2    
Watershed Area 634 ha    
Tributary Name Standing Boy Creek   
Stream Order 2  
UTM Zone 16 Coordinates 16 N 692405 3613831  
Seasons Sampled 6    
Total Macroinvertebrate Richness 114  
Total Fish Richness 12    
          
Parameter Mean Min Max SE
Algae     
AFDM (g / m
2
) 0.42 0.02 1.40 0.33 
Chlorophyll a (mg / m
2
) 0.36 0.05 0.48 0.11 
     
Macroinvertebrates 
Biomass (g / m
2
) 17.01 1.40 36.87 4.96 
Diversity (H') 2.37 2.15 2.63 0.09 
Richness 37.40 27.00 43.00 3.08 
GA-BMI 46.00 34.00 61.00 4.71 
     
Fish 
Diversity (H') 1.01 0.49 1.65 0.24 
Richness 6.00 3.00 9.00 1.47 
GA-IBI 33.00 28.00 40.00 2.65 
     
 
 
 
203 
 
Watershed ID SB4    
Watershed Area 2659 ha    
Tributary Name Standing Boy Creek   
Stream Order 3  
UTM Zone 16 Coordinates 16 N 697366 3612240  
Seasons Sampled 6    
Total Macroinvertebrate Richness 132  
Total Fish Richness 23    
          
Parameter Mean Min Max SE
Algae     
AFDM (g / m
2
) 1.05 0.00 1.62 0.53 
Chlorophyll a (mg / m
2
) 0.22 0.19 0.24 0.02 
     
Macroinvertebrates 
Biomass (g / m
2
) 31.68 4.22 69.02 9.04 
Diversity (H') 1.88 1.26 2.33 0.16 
Richness 36.33 30.00 44.00 2.63 
GA-BMI 41.67 26.00 58.00 4.59 
     
Fish 
Diversity (H') 1.68 1.00 2.09 0.19 
Richness 10.20 9.00 12.00 0.58 
GA-IBI 40.00 32.00 50.00 3.52 
     
 
 
 
204 
 
Watershed ID SC    
Watershed Area 896 ha    
Tributary Name Sand Creek    
Stream Order 2  
UTM Zone 16 Coordinates 16 N 680233 3635938  
Seasons Sampled 6    
Total Macroinvertebrate Richness 116  
Total Fish Richness 14    
          
Parameter Mean Min Max SE
Algae     
AFDM (g / m
2
) 0.58 0.03 1.13 0.55 
Chlorophyll a (mg / m
2
) 0.27 0.06 0.48 0.21 
     
Macroinvertebrates 
Biomass (g / m
2
) 11.90 0.90 26.79 3.15 
Diversity (H') 1.73 1.21 2.39 0.16 
Richness 33.83 26.00 40.00 2.29 
GA-BMI 35.17 21.00 44.00 3.65 
     
Fish 
Diversity (H') 1.52 0.69 1.85 0.17 
Richness 7.00 2.00 10.00 1.15 
GA-IBI 37.33 28.00 44.00 2.67 
     
 
 
 
205 
 
Watershed ID WC    
Watershed Area 2193 ha    
Tributary Name Weracoba Creek   
Stream Order 2  
UTM Zone 16 Coordinates 16 N 691980 3590902  
Seasons Sampled 3   
Total Macroinvertebrate Richness 47   
Total Fish Richness 10   
          
Parameter Mean Min Max SE 
Algae   
AFDM (g / m
2
) 2.41 2.41 2.41 0.00 
Chlorophyll a (mg / m
2
) 0.05 0.05 0.05 0.00 
     
Macroinvertebrates 
Biomass (g / m
2
) 20.13 1.33 53.06 13.49 
Diversity (H') 1.74 1.46 2.25 0.21 
Richness 22.67 17.00 33.00 5.17 
GA-BMI 17.33 15.00 20.00 1.45 
     
Fish 
Diversity (H') 1.76 1.58 2.04 0.14 
Richness 7.33 6.00 9.00 0.88 
GA-IBI 35.67 35.00 36.00 0.33