Parameters Affecting Bird Use of Stormwater Impoundments in the Southeastern United States: Implications for Bird-Aircraft Collisions by Brian James Fox A thesis submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Master of Science Auburn, Alabama May 9, 2011 Keywords: Bird Strikes, Airports, Stormwater, Wildlife, Habitat Selection, Wildlife Damage Management Copyright 2011 by Brian James Fox Approved by James B. Armstrong, Co-Chair, Professor of Wildlife Sciences Bradley F. Blackwell, Co-Chair, Research Wildlife Biologist, USDA/APHIS NWRC James B. Grand, Professor & Unit Leader, Alabama Cooperative Fish & Wildlife Research Unit ii Abstract Bird-aircraft collisions are a large and growing threat to aviation safety in the United States. Stormwater management impoundments in and around airports create conditions which attract hazardous wildlife species to air operations areas. Airport biologists and other stakeholders seek ways to design and manage these structures to reduce their relative attractiveness to hazardous wildlife species. Here I report on the results of a two-year observational study to quantify parameters influencing bird use of stormwater impoundments in a metropolitan area of the southeastern United States. My analysis demonstrates that while the influence of impoundment design features varies between foraging guilds, bird use of stormwater impoundments in the southeastern United States can broadly be reduced by minimizing impoundment area, eliminating standing water, increasing impoundment bank slope and locating impoundments so as to maximize their isolation from open water sources. iii Acknowledgments My work was funded through a Cooperative Service Agreement between the USDA APHIS WS Alabama office, in cooperation with the WS National Wildlife Research Center?s Ohio Field Station, and Auburn University. Funding for the agreement was provided via Interagency Agreement 07-7201-4430-GR between the FAA and WS. Considerable training and technical advice was provided by Doctors James B. Armstrong (Auburn University), Bradley F. Blackwell (WS National Wildlife Research Center) and James B. Grand (USGS). Additional guidance on analysis and data collection was provided by Doctors Todd Steury and Mark MacKenzie (deceased), respectively. Logistical support was provided by Mr. Frank Boyd, state director of WS Alabama. Most of the considerable field work on this project was completed with technician Wesley Brian Holland of APHIS WS Alabama. Advice and geospatial information were provided by Mr. Matt Dunn of the Auburn Water Works Board and Ms. Leisa Simpson of the City of Auburn?s Business Development Services. iv Table of Contents Abstract ......................................................................................................................................... ii Acknowledgments........................................................................................................................ iii List of Tables ............................................................................................................................... vi List of Figures ............................................................................................................................. vii List of Abbreviations ................................................................................................................. viii Chapter 1: Introduction ............................................................................................................... 1 Chapter 2: Methods ..................................................................................................................... 6 Study Area ...................................................................................................................... 6 Avian Guild Selection ..................................................................................................... 6 Sample Pond Selection ................................................................................................... 7 Covariate Selection & Models ........................................................................................ 7 Pond Characteristics ........................................................................................................ 9 Observations ................................................................................................................. 14 Model Fitting & Selection ............................................................................................ 15 Chapter 3: Results ..................................................................................................................... 17 Observations ................................................................................................................. 17 Pond Characteristics ...................................................................................................... 17 Analysis & Model Fitting ............................................................................................. 18 Chapter 4: Discussion ............................................................................................................... 20 v Managing Stormwater Run-off ..................................................................................... 24 Summary ....................................................................................................................... 27 Literature Cited ......................................................................................................................... 29 Appendix 1: Avian Guild Selection .......................................................................................... 44 Appendix 2: Vegetation Guilds ................................................................................................ 54 Appendix 3: Bird Species Observed Using Impoundments In This Study ............................... 56 Appendix 4: Other Vertebrates Observed In This Study Using Stormwater Impoundments ... 61 Appendix 5: Analysis Results for A Priori Models .................................................................... 64 vi List of Tables Table 1: A Priori model Set ....................................................................................................... 38 Table 2: Summary of encounters by avian guild ....................................................................... 41 Table 3: Summary statistics for study ponds ............................................................................. 42 vii List of Figures Figure 1: Study area and sample site distribution ...................................................................... 43 viii List of Abbreviations AC Advisory Circular AOA Air Operations Area EV Emergent Vegetation FAA Federal Aviation Administration ISR Impervious Surface Run-off OW Open Water WS Wildlife Services 1 INTRODUCTION Collisions between wildlife and aircraft (hereafter ?wildlife strikes?) are a serious and growing threat to civil (Dolbeer et al. 2009) and military (Zakrajsek & Bissonette 2005) aviation safety. Of these wildlife strikes, bird-aircraft collisions (hereafter ?bird strikes?) are by far the greatest concern. In 2008 alone, birds accounted for 96.9% of the 7,516 wildlife-aircraft collisions reported to the Federal Aviation Administration (FAA; Dolbeer et al. 2009). Between 1990 and 2008 there were 89,727 reported wildlife strikes in the United States resulting in approximately $308.3 million dollars in losses (Dolbeer et al. 2009), although the actual losses are far higher. Dolbeer et al. (2009) estimated that only 39% of all wildlife strikes are reported, while only 17% of reported strikes include any estimate of financial losses. This strike reporting rate continues to grow from reporting rates as low as 20% in the early 1990?s (Cleary & Dolbeer 2005). The danger posed by wildlife strikes is the resulting effect on flight (EOF). Wildlife strikes resulting in a negative effect on flight typically cause damage to engines, cockpit windshields, flight control surfaces or landing gear. The financial losses resulting from these strikes include aircraft repair and replacement costs as well as revenue lost due to flight delays, cargo loss/damage and increased bird-strike prevention efforts. Dolbeer (2006) reported that 74% of bird strikes occur less than 500 feet above ground level (AGL), when aircraft are within an airport?s perimeter or in close proximity (i.e. final approach, take-off/landing roll or initial ascent). The underlying assumption is that the birds involved are attracted to the area by habitat characteristics or resources in the immediate vicinity of the collision (Blackwell et al. 2009). Therefore, bird-strike prevention efforts are focused 2 primarily on airport property and adjoining private properties (however, see Blackwell et al. 2009). The FAA is responsible for advising airport managers and other stakeholders on managing hazardous wildlife attractants. The FAA (2007) instructs airport managers to address, and if possible eliminate, wildlife attractants within 1,524 meters of the airport?s air operations area (AOA) for airports serving piston aircraft and 3,048 m for airports serving turbine aircraft. The AOA encompasses all surface areas designed for aircraft movement including runways, taxiways and tarmacs. The FAA outlines hazardous wildlife attractants to be avoided in Advisory Circular (AC) 150/5200-33B (http://www.faa.gov/documentLibrary/media/advisory_circular/150-5200- 33B/150_5200_33b.pdf). However, the AC offers only broad recommendations for addressing these wildlife attractants. The AC is written with a national scope and offers no advice on adapting wildlife hazard management with respect to regional variation in vertebrate diversity or associated habitat preferences, beyond consulting a local wildlife biologist. As a result there has been a recent push to investigate and quantify regional factors influencing wildlife hazard attractants, particularly with respect to vegetation (Barras & Seamans 2002, Blackwell et al. 2009, Blackwell et al. unpublished manuscript) and water resources (Blackwell et al. 2008 & 2009). Water resources in an airport?s AOA are of great concern, because many of the avian genera considered most hazardous to aviation require open water in their habitats (De Graaf et. al 1985, Dolbeer et al. 2000, Sibley 2001). Stormwater impoundments are a particular problem, as they are necessary in and around airports to ensure environmental compliance by trapping and treating impervious surface runoff (ISR) (Baier 2003) and contribute to safe aircraft ground 3 movements by directing ISR away from the AOA. These impoundments create a wildlife attractant because run-off events produce standing water and associated vegetation communities (FAA 2007). Over time these impoundments may develop sediment deposits and vegetation complexes that support an array of invertebrate and vertebrate diversity (Le Viol et al. 2009), which combined may offer foraging, loafing, roosting and nesting space to many bird species. For example, Sharpe (2005) observed that bird, mammal and amphibian use of a dual-purpose stormwater impoundment/wetland mitigation site began even before site development was complete. Brand and Snodgrass (2009) found that stormwater impoundments were a major component of successful amphibian breeding habitat in a suburban landscape. These studies offer examples of the growing body of evidence demonstrating the ecological value of stormwater impoundments to wildlife across urban and suburban landscapes. Wildlife use of stormwater impoundments and other constructed wetlands has received a great deal of attention (e.g. Andersen at al. 2003, Brand & Snodgrass 2009, Sparling et al. 2004, Terman 1997), especially efforts to enhance stormwater facilities for wildlife attraction (e.g. Adams et al. 1985, Duffield 1986, McGuckin & Brown 1995, Sparling et al. 2007, White & Main 2005). Far less effort has focused on reducing wildlife use of stormwater impoundments to reduce or avoid wildlife-related hazards (Barras & Seamans 2002, Blackwell et al. 2008 & 2009). The desire of many community stakeholders to enhance stormwater impoundments for wildlife is a serious obstacle to safe airport operations. This creates an urgent need to investigate design and management strategies to reduce the relative attractiveness or utility of stormwater impoundments as habitat features. Blackwell et al. (2008) investigated parameters affecting bird use of stormwater impoundments in the Seattle-Tacoma, WA, USA, area, with an emphasis on identifying features 4 of pond designs that could be manipulated to make the sites less attractive to birds. Further, the authors selected ponds that could serve as surrogates to on-airport facilities, with respect to size and other design features. They found a model comprising surface area, a ratio of open water to emergent vegetation, irregularity and isolation to be a suitable predictor of bird use for 9 of 13 avian groups analyzed in their study. Post-hoc analysis for these groups showed isolation to be an important determining factor for use by blackbirds (Icteridae spp. & Sturnidae spp.), dabbling ducks and diving ducks (Anatidae spp.), such that probability of use equated to 0 at a 7-km separation between water resources. They also found models without surface area to be strong predictors of use by rock pigeons (Columba livia), killdeer (Charadrius vociferus), great blue herons (Ardea herodias) and geese (Anserinidae spp.). All four pose a significant bird strike risk (Dolbeer et al. 2009). The authors broadly recommended that managers reduce the likelihood of bird use by minimizing pond perimeter and maximizing pond isolation for new impoundments or minimizing open water in existing structures. However, the authors limited their inferences to the landscapes and avian communities of the Pacific Northwest and recommended that these bird-habitat associations be investigated across other regions of the United States. Further, the problem of impoundments as wildlife hazards includes not only a geographical perspective, but also a local component when considering properties adjacent to airports. Impoundments that are off an airport?s property but within the FAA siting criteria may still serve to attract wildlife to an airport?s vicinity, but are beyond the immediate control of airport managers and biologists. In some instances these impoundments are being managed for priorities that pose immediate hazards to aviation safety, such as enhancing avian wildlife use for residential enjoyment (Lee & Li 2009) and biodiversity mitigation or enhancement (Davis et al. 2008, Brand & Snodgrass 2009, Le Viol et al. 2009). 5 My objective was to quantify local- and landscape-level relationships associated with stormwater impoundments in the southeast USA that might serve as avian attractants. In so doing, I tested nine a priori models representing competing hypotheses to describe the probability of impoundment use by avian guilds (see below). These a priori models consisted of differing combinations of 11 variables: pond hydrology type (retention vs. detention), mean pond surface area, mean perimeter irregularity, the ratio of open water to emergent vegetation, the total surface area of adjacent open water resources, impoundment isolation (relative to other open water resources), mean impoundment bank slope, vegetation community diversity and adjacent landcover diversity. This suite of variables includes those parameters considered by Blackwell et al. (2008), as well as others. 6 METHODS Study Area I conducted my study in the Auburn-Opelika Metropolitan area in Lee County, Alabama (Figure 1) from March 2008 to March 2010. Average high and low temperatures are 23.9?C and 11.7?C, respectively with average annual rainfall of 134.6 cm. Historically, this region was dominated by southeastern coastal plain habitats and hardwood forests, including vast tracts of longleaf pine (Pinus palustris) forests. Today much of this area has been converted to agriculture, timber production, and urbanization (Commission for Environmental Cooperation 1997). The Auburn-Opelika area has experienced steady population and economic growth in the last 50 years (City of Auburn, 2010). Lee County now has a population of over 135,000 with more than 57,000 living in the Auburn-Opelika urbanized area. Avian Guild Selection I developed a set of 28 guilds encompassing all of the bird species known to occur in Alabama (Alabama Ornithological Society 2006). I excluded strictly pelagic species (i.e. Magnificent Frigatebird [Fregata magnificens]) and other strictly coastal species (i.e. Brown Pelican [Pelecanus occidentalis]), due to the extremely low probability of encountering these species in Alabama?s interior. The resulting species list represented all species in the study area that I hypothesized might utilize stormwater impoundments. Guilds were arranged primarily by foraging ecology (De Graaf et al. 1985, Sibley 2001) and with respect to each species? relative hazard to civil and military aviation (Dolbeer et al. 2000, Zakrajsek & Bissonettte 2005). Higher classification of species tentatively followed Hackett et al. (2008). A complete description of guild membership by family and genus is presented in Appendix I. 7 Sample Pond Selection I selected 40 stormwater impoundments (Figure 1) to serve as surrogates for stormwater impoundments occurring within or proximate to the 3048-m FAA siting criteria (FAA 2007). These surrogate ponds presented characteristics typical to all stormwater impoundments. They were all open basin designs with inlet and outflow pipes, rip-rap areas and spillways (Baier et al. 2003). All study ponds detained water in the weeks prior to the beginning of field observations. I was specifically interested in incorporating ponds with design and management features that were not guided by the FAA?s design and management recommendations in AC 150/5200-33B (i.e. unmanaged or naturalistic shorelines). Therefore, I did not constrain my selection of ponds by vegetation conditions in the impoundments? basins at the start of this study. The majority of my sample sites occurred in new residential construction and commercial sites including business parks, industrial parks, parking lot margins, and shopping malls. A large portion of my study area was established and developed before stormwater management was regulated (author personal observation), so my sample sites were not uniformly distributed across my study area. Covariate Selection and Models I developed a set of 9 a priori models to describe probability of use by each avian guild, including a null model (intercept only) and 8 reduced models (Table 1). Each model described the probability of bird use of stormwater impoundments in the southeastern United States as a combination of two or more of the following parameters: pond design type (Adams et al. 1985, Cleary & Dolbeer 2005), pond surface area (Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Cicero 1989, Duffield 1986), pond perimeter irregularity (Adams et al. 1985, Blackwell et al. 2008, Carbaugh et al. 2020, Cicero 1989), the ratio of open water to emergent vegetation (Blackwell et al. 2008, Duffield 1986), the total area of adjacent open water 8 (Brown & Dinsmore 1986, Duffield 1986, Steen et al. 2006), the minimum distance from an impoundment to the nearest open water resources (Blackwell et al. 2008, Brown & Dinsmore 1986, Duffield 1986, Dunton & Combs 2010), mean impoundment bank slope (Adams et al. 1985, Duffield 1986, Dunton & Combs 2010), vegetation community complexity within an impoundment basin (Bancroft et al. 2002, Cicero 1989, Cleary & Dolbeer 2005, Steen et al. 2006), the diversity of landcover types immediately surrounding an impoundment (Benoit & Askins 2002, Croci et al. 2008, Hostetler et al. 2005, Traut & Hostetler 2003) and season (Caula et al. 2008, Duffield 1986). My covariate set included the same suite of covariates investigated by Blackwell et al. (2008, citations therein). However, I expanded on their model set by investigating bank slope, pond design type, area of adjacent open water, vegetation community complexity, landcover diversity, and season (Duffield 1986). Furthermore, my model approach differs from Blackwell et al. (2008) in the inference drawn between models. In Blackwell et al (2008), each model differed by the sequential removal of each parameter. In my model set, each model represents a separate hypothesis as to which set of parameters may best describe bird use. While some models differ only by the inclusion or exclusion of one parameter, my intent is not to judge the contribution of individual parameters, but rather to identify a group of parameters that are suitable to describe the system in question for each avian guild analyzed. Each model, except the null model, was tested with and without the effect of season (see below), resulting in a set of 17 models run for each avian guild. During pond selection, I observed that bird use of impoundments in my study area appeared to decline briefly in the days after a large (>2.54 cm) rain event. I assumed an effect of precipitation on detection in my a priori models, either because species remained sheltered 9 during rain despite my efforts to flush them or because birds preferentially utilized natural ephemeral wetland resources during and immediately after rainfall. Pond Characteristics Variance Between Seasons I coded each weekly sampling period as spring, summer, fall or winter. This represents another departure from Blackwell et al. (2008). Each sampling occasion?s assignment to a season was based on the day on which each occasion began. Intervals were grouped as: March through May ? spring, June through August ? summer, September through November ? fall, December through February - winter. Some of my avian guilds consisted entirely of year-round residents, while others consisted primarily of fall migrants or winter residents. Testing each model with and without the effect of season allowed me to reduce some un-modeled heterogeneity in avian guilds for which time of year was correlated with probability of occurrence. This does not describe any additive or interactive effect between time of year and other covariates. Basin Type I identified each pond as either a detention pond (dry between run-off events) or retention pond (continuously wet between run-off events). Any pond that drained completely of water at any point during the survey period (defined as complete desiccation of the basin?s surface soil) was identified as a detention pond. Conversely, any basin which retained water continuously during the study period (defined as standing water or continuously saturated or muddy surface soil in the basin) was identified as a retention pond. My definition of retention ponds differed slightly from the intended design of these ponds, because some ponds in this study were designed to be detention ponds, but retained water continuously during the study period. This 10 may be attributed to the accumulation of sediment in older impoundments, which can alter their hydrology over time. Mean Area & Irregularity I defined mean area for each pond as the mean of 6 surface-area (km2) measurements made once every 5 weeks beginning the first full week of point counts. Pond area was defined as the total area of continuous water within each basin, and included surface area dominated by emergent vegetation as well as saturated or muddy shoreline areas. I made measurements using a TDS Nomad? GPS coupled with a Hemishpere Crescent? backpack-mounted antenna operating the GIS package SOLOForest?. I traced the antenna along the edge of the water area, and the software logged time interval waypoints producing a polygon representing the pond?s shape and area. In SoloForest? I adjusted the interval between waypoints (between 1 and 10 seconds) as needed to correct for the time needed to cross dense vegetation and steep terrain around each pond?s basin. This adjustment ensured that area measurements between ponds were conducted with the same level of accuracy across varying basin conditions. SoloForest? also calculated perimeter length for each pond as the total length of the distance traveled by the antenna during each area measurement. These perimeter measurements were also recorded as a mean value for each pond across all perimeter measurements made during the study period. From this value, I calculated perimeter irregularity as the ratio of the mean pond perimeter to the perimeter of a perfect circle of the same area (following Blackwell et al. 2008, citations therein). Open Water : Emergent Vegetation Ratio The ratio of open water to emergent vegetation has already been shown to correlate with waterfowl use (Duffield 1986, Hobaugh & Teer 1981). During each area measurement, I estimated the percentage of total pond surface which was dominated by emergent vegetation. 11 This estimate also included saturated or muddy soils and surface ice. From this percent cover, values were calculated for the area of open water (OW) and the area of emergent vegetation (EV). A ratio of OW:EV was calculated for each pond as the average ratio value across all six area measurements. Adjacent Open Water & Isolation It has been demonstrated that pond isolation (minimum distance to adjacent wetland resources) and the size of adjacent wetland patches correlate with avian use of wetland resources (Brown & Dinsmore 1986). I recorded open-water resources as the total area of open water resources within a 1-km buffer of each pond?s initial area measurement (Blackwell et al. 2008, Fairbairn & Dinsmore 2001). I manually digitized open-water resources and calculated their total area in ArcMap 9.2 (ESRI 2006) on digital orthorectified quarter quadrangle (DOQQ) aerial images (Alabama State Water Program 2008). Pond isolation for each pond was recorded as the minimum distance between each pond and any open-water resource, as calculated using the Near tool in ArcMap 9.2 (ESRI 2006). I scaled both of these variables by dividing each value by 10, because the scripts I used for analysis in Matlab? (Mathworks, Inc. 2010) became unstable with very large values. Bank Slope Bank slope may be an important determining factor in bird use of constructed wetlands either by facilitating foraging by shoreline foraging species (Cleary & Dolbeer 2005, De Graaf et al. 1985), or fostering the growth of emergent vegetation communities that provide shelter, nesting or additional foraging areas for birds (Duffield 1986). Bank slope for each pond was calculated as the mean percent slope for the cardinal points at the waterline of each pond 12 extending from the pond?s centroid. Percent slope was measured by a single observer using a Haglof? digital clinometer and recorded as a decimal value from 0 to 1. Vegetation Index Vegetation guilds were developed a priori to encompass the plant types occurring in the study area. Guilds were defined with primary respect to successional stage and secondary consideration to plant taxonomy (see Appendix 2). I surveyed an area of the basin encompassing all of the surface area in a buffer extending 5 meters in to the pond and 5 meters away from the pond?s shoreline. I recorded the coverage of each guild as a percent of the total buffer area. Vegetation diversity for each pond was calculated with a Shannon diversity index (Ricklefs 1990) at the midpoint of both observational years (August 2008 & 2009) as , where S is the number of guilds present at each pond and pi is the area of each guild as a proportion of the total buffer area. For my analysis I used the index values calculated in 2008 for all sampling periods that began in March 2008 to February 2009. I used the index calculated in 2009 for all point counts which began from March 2009 to March 2010. This was done to more accurately reflect the nature of vegetation community diversity at these ponds over time, because some ponds experienced significant shifts in vegetation community composition during the observation period attributed to management efforts (i.e. brush chopping) by landowners. Landscape Index A Shannon diversity index was also calculated for land cover/use within one km of each pond using data from the Alabama Gap Analysis Project (AL-Gap 2008). The Alabama land cover data set for the study area was modified by condensing the habitat types represented in the study area in to 6 broad land-use categories: 13 1. Open Water: All open-water resources including wetlands and constructed ponds 2. Open Development: <20% impervious surface, includes golf courses, rural homes, row crops and pastures 3. Low-intensity Development: 20 ? 49% impervious surface 4. Medium-intensity Development: 50 ? 79% impervious surface 5. High-intensity Development: 80 ? 100% impervious surface 6. Undeveloped/Rural area: All vegetation types unmodified by or for human use or activity These categories were produced by utilizing the existing definitions of low- to high- intensity development as defined in the AL-GAP classification scheme (AL-GAP 2008) and condensing all natural land cover types into the broader category of undeveloped/rural area. Areas of pasture and cultivation were condensed into open development, because they represent areas of anthropogenic landscape modification with little impervious surface and abundant vegetation. I calculated the percent coverage for each land cover type in a 1-km buffer zone around each pond and calculated a Shannon diversity index from the resulting data as , where S is the number of guilds present within 1 km of each pond and pi is the area of each guild as a proportion of the total buffer area. This method is similar in application to Croci et al. (2008) who used a Shannon diversity index of land cover type to describe avian community composition in relation to the diversity of fragmented land cover types across an urban to rural gradient. An increase in diversity of land cover types has been observed to benefit some avian species groups (Blair 1996, Dykstra et al. 2001, Stout et al. 2006, Traut & Hostetler 2001). 14 Precipitation Precipitation data for the study area were drawn from the National Oceanic and Atmospheric Administration?s (NOAA) National Climatic Data Center (NOAA 2010). Total precipitation for each week was recorded in meters. Observations I randomly assigned my sample of 40 ponds to 4 groups of 10 ponds each (sets A through D). Within each set no two ponds were located within 1 km of each other, to reduce the likelihood of double counting individual birds moving between ponds during counts. Each set was surveyed for 1 calendar week (5 days) on a rotating basis (beginning with set A), so that each set was surveyed 1 out of every 4 weeks. Within each week, each pond was surveyed in random order twice daily (ranging from 30 minutes to 2 hours after sunrise and from 2 hours to within 30 minutes of sunset). This variation in survey time not only allowed me to account for hourly variation in species use of the sites, but also facilitated access to sample sites located on government and commercial properties. The sampling regimen represents 20 attempts to detect each guild weekly. I believe this strategy afforded me a higher probability of detecting elusive or non-resident species. My counts continued for 102 weeks beginning the morning of March 17, 2008, and concluding the afternoon of February 26, 2010. My survey protocol consisted of a 3-minute walking survey of the pond?s perimeter. This walking survey allowed me to disturb vegetation around the pond?s perimeter and flush birds that might otherwise not have been observed. My survey approach was somewhat similar to a double-sampling methodology (Bart and Earnst 2002), in that each pond survey consisted of a rapid survey (e.g. all birds observed on initial approach), combined with a more intensive walking survey of these irregularly shaped survey plots. Because some ponds were highly 15 irregular in shape, the walking survey also ensured that the entire perimeter was visually surveyed. I note, however, that my survey approach yielded a ?snapshot? count adjusted by those individuals flushed or entering the site during the survey, not a corrected density estimate as outlined by Bart and Earnst (2002). During the 3-minute count I identified down to species each individual observed in the pond, within the pond?s basin or foraging in flight immediately above the pond (for example, Osprey [Pandion haliaetus] circling overhead). Each count included all individuals observed upon arrival at the pond, as well as all individuals that arrived in the aforementioned area during the count. Individuals who could not be identified down to species were still identified down to genera sufficiently to be assigned to a guild (i.e. unidentified sparrows [Emberizidae spp. & Passeridae spp.]). Any individual that could not be identified to guild was excluded from the point count. I also recorded the number of individuals of each guild observed at each count. For my analysis, however, I based pond use by guilds on a weekly interval as a binary value (detected or not detected). Each week?s 20 walking surveys represented 20 efforts to detect guild use at any point that week. Therefore, if a guild was detected at least once during these 20 walking surveys, then that guild was assigned as detected for that specific pond relative to that weekly interval. Model Fitting and Selection I used Matlab? for model fitting (Mathworks, Inc. 2010). I used occupancy analysis to estimate probability of use and detection (P) for each guild encountered on more than 20 occasions. I defined occupancy as use of a site by any member (species) of a guild in a given week. My detection model estimated the probability of detecting a guild, assuming it was present at the time of survey. To estimate occupancy I combined the encounter history of each 16 guild (see above) with my covariates. I obtained parameter estimates using a logit link in the form , where ? represents the parameter estimates of each parameter in the model. I calculated Corrected Akaike?s Information Criterion (AICc), model weights ( i) and evidence ratios for each avian guild. AICc is a measure of Kullback-Leibler (K-L) information with an additional term to correct for bias arising from a small sample pool (n) (Burnham and Anderson 2002). The AICc represents an estimation of the information lost between biological truth and the models being considered, given the data being analyzed (Anderson et al. 2000). Model evidence allowed me to infer the relative strength of each model?s ability to describe the response variable, P, for the system observed. A model was considered the best approximating model for a given guild if the evidence ratio between the best model and the second best approximating model was ?3.0. This evidence ratio was calculated from each model?s Akaike weight (?i), which represents the likelihood that a given model is the best approximating of several models being considered (Burnham & Anderson 2002). 17 RESULTS Observations I completed 104 weeks of point counts. No observations were made during the week of November 23, 2009 due to logistical constraints. An additional week of point counts was completed for this set of ponds at the end of the observational period in order to maintain an equal number of observations between all 4 sub-sets of ponds. I observed 145 bird species in 94 genera (see Appendix 3 for a complete list of species), representing 27 of my 28 avian guilds. The only guild not observed was the cuckoos (Coccyzus spp.). The most frequently observed guild was the Longtailed Ground Birds (i.e. Northern Mockingbird [Mimus polyglottos], Brown Thrasher [Toxostoma rufum], etc.), with 414 encounters (Table 2). In addition to the observed bird species I encountered 22 mammal, 10 amphibian and 18 reptile species or genera (see Appendix IV). Avian diversity averaged 4 species per week across all ponds, with weekly species counts ranging from 0 to 16. Bird use of stormwater impoundments reached its minimum in winter and peaked in summer. Pond Characteristics Twenty-nine ponds retained water continuously throughout the study, while the other 11 dried completely at least once. Mean weekly precipitation for the study area averaged 127 mm (SD ?157 mm), and ranged from a minimum of no measurable rainfall to a peak of 537mm. This extreme observation was recorded the week Hurricane Fay passed across the southeastern United States in August 2008. Descriptive statistics for my sample pond covariates can be found in Table 3. 18 Analysis & Model Fitting Eleven guilds were not encountered frequently enough to reliably perform model fitting (See Table 2). The number of observations was sufficient for model parameter estimates for the remaining 17 guilds, although some standard errors were inestimable. This may be due to the limitations of the maximum-likelihood estimator (MLE) I used in Matlab?. Increasing pond surface area was positively correlated with probability of use for all guilds analyzed and was a component of at least one of the top two best-approximating models for every guild except blackbirds (Appendix 5). Among all guilds whose impoundment use was best approximated by a model including slope, the correlation was negative. Season was a component of the at least one of the top two best-approximating models for all passerine guilds except Kingfishers. Pond use by aerials (Hirundinidae spp. & their allies, see appendix 1) was best approximated by a model composed of mean pond surface area, landscape diversity and season (Appendix 5). The same parameters without the effect of season, were the best approximation of impoundment use by kingfishers (Appendix 5). Impoundment use by anserinids and domestic/exotic waterfowl was best predicted by a model composed of area, irreg, OW:EV and isol (Appendix 5). This model was also found to be an adequate, although not the outright strongest, model to describe anserinid use in Blackwell et al. (2008) (Evidence ratio <3.0). The two models differ in that the model I test here assumed an effect of precipitation. Pond use by blackbirds was best described by a model composed of type, irreg, slope, veg and season (Appendix 5). For doves, this model was equal in strength to a model composed of area, irreg, OW:EV, slope, veg and season, so area may be considered superfluous to 19 describing use by doves (Appendix 5). Flycatcher use of impoundments was aqeduately described by two models. Both consist of area, irreg, OW:EV, slope and veg, while the weaker of the two included the effect of season. Therefore, season may be considered superfluous to describing use by Flycatchers (Appendix 5). For 6 passerine guilds (Brights, Corvids, Longtails, Small Forest, Sparrows and Warblers, see Appendices 1 & 5) and wading birds (Appendix 5) a model composed of area, irreg, OW:EV, isol and season was the best approximating model or at least adequate to describe impoundment use. Among these guilds, there was also evidence for a model composed of isol, irreg and season to describe use by sparrows (Appendix 5), while a model of area, irreg, OW:EV, slope, veg and season was also adequate to describe use by both longtailed (i.e. Mimidae spp., Appendix 5) and small forest (i.e. Troglodytidae spp., Appendix 5) passerines. The latter model was also 1 of 2 plausible models to describe impoundment use by flycatchers and doves (Appendix 5). The 2 best-approximating models for flycatchers differed only in the effect of season (Appendix 5). There was almost no difference in strength between a model of area, irreg, OW:EV, slope, veg and season versus a model composed of type, irreg, slope, veg, and season to describe dove use of impoundments (Appendix 5). Model evidence for dabbling ducks showed no best-approximating model among the four highest ranked (Appendix 5). The 4 best approximating models all included slope and veg as well as differing combinations of type, area, irreg, OW:EV and season). Raptor use was best approximated by a model of area, irreg, OW:EV, slope and veg (Appendix 5). Shorebird use was best approximated by a model composed of area, irreg, OW:EV, slope, veg and season (Appendix 5). 20 DISCUSSION Application of habitat-management practices simultaneously across multiple foraging guilds is challenging for airports (Linnell et al. 1996, Seamans et al. 2007). Moreover, controlling stormwater runoff on the airport poses a variety of direct and indirect safety issues (FAA 2007, Blackwell et al. 2008). Further, the challenges of mitigating wildlife the hazards posed to aviation on and near airport properties are enhanced by stormwater-management facilities on private property within or proximate to FAA citing criteria. Here, I report results pertaining to avian guild-specific probability of use of stormwater-management ponds relative to a set of a priori models and based on two years of weekly observations of bird use at 40 retention/detention ponds comparable to privately-owned stormwater-management facilities found on or near airports in the southeast USA. I, first, discuss my findings from the perspective of individual guilds, beginning with those guilds for which my model results showed the strongest evidence. I then relate my findings to how stormwater runoff can be better managed, including facility design considerations by urban/airport planners to reduce avian attractants on and near airports. Bank slope was negatively correllated with use by dabbling ducks, but not anserinids. This correlation has been demonstrated for Canada Geese (Branta canadensis) in previous studies (Dunton & Combs 2010, citations therein), so the latter observation may due to the contribution to model fit of other parameters in the best approximating model. The weak model evidence among the four best approximating models for dabbling ducks suggests they are responding to factors not measured in this study. This may be due in part to the influence on model fit of impoundment type and bank slope. Both variables showed strong negative correlation here, but were not analyzed by Blackwell et al. (2008). Furthermore, the 21 impoundments observed in this study were distributed across an urban to rural gradient (Figure 1) and waterfowl may have been responding to anthropogenic resources associated with urbanizing area, such as highly palatable landscaping, or the absence of predators. Waterfowl in the study area may have also been responding to reduced hunting pressure. Dieter et al. (2010) demonstrated that fall movement of Canada geese was influenced by hunting pressure in South Dakota, while Holevinski et al. (2007) demonstrated that suburban-dwelling Canada geese demonstrated high-site fidelity in areas closed to hunting, despite hazing efforts. To my knowledge there was no waterfowl hunting in the study area during the observation period, and I had few encounters at these impoundments with predators that might be expected to prey upon or harass waterfowl (i.e. coyotes [Canis latrans], author personal observation). It is therefore plausible that stormwater impoundments in this urban area offer waterfowl a refuge with anthropogenic resources (i.e. palatable landscaping) and reduced mortality pressure. Waterfowl are among the highest management priorities for airport biologists (Dolbeer et al. 2000) and identifying un-quantified factors influencing their use of impoundments is urgent. Diurnal raptors are also a high priority for airport managers (Dolbeer et al. 2000), although information on diurnal raptor use of stormwater impoundments is limited. Dykstra et al. (2001) suggested anthropogenic water resources to be an important component of suburban red-shouldered hawk (Buteo lineatus) habitat, while Stout et al. (2006) demonstrated that open water was a small and negatively correlated component of occupied red-tailed hawk (Buteo jamaicensis) habitat in a similar suburban setting. Dykstra et al. (2001) suggested that athropogenic ponds allowed suburban-dwelling red-shouldered hawks to sustain themselves on smaller territories by providing additional foraging sites. My estimates show a generally positive correlation between diurnal raptor use and increasing pond isolation. In this study those ponds 22 with the greatest isolation measurements were generally those in more heavily urbanized areas. I believe my observations on diurnal raptor use support Dykstra et al. (2001). Stakeholders should therefore be aware that impoundments isolated by suburban area may actually be used more frequently be diurnal raptors than those impoundments proximate to undeveloped areas. Given the broad distribution of diurnal raptors across the southeast (AOS 2006, Sibley 2000) and their risk of bird strikes (Dolbeer et al. 2000), future efforts should be made to identify (1) the relative importance of stormwater impoundments as a component of their available habitat and (2) landscape and habitat characteristics influencing diurnal raptor presence across urban to rural gradients, including airports, in the southeast (e.g. Dykstra et al. 2001, Stout et al. 2006). Wading birds (i.e. Herons & Egrets [Ardea, Butorides & Egretta spp.]; Appendix3) were frequent users of impoundments in my study (Table 2). Use by waders was positively correlated with increasing surface area, perimeter irregularity and the ratio of open water to emergent vegetation, while increasing isolation from other open water resources was negatively correlated in the best approximating model (Appendix 5). Reducing impoundment use by waders will be very difficult, as they appear capable of utilizing impoundments of very small surface area and minimal irregularity. Probability of impoundment use by waders, while holding other parameters constant, ranged from 0.54 to >0.90 when surface area was varied from 0.01 to 0.15 m2. Even with no perimeter irregularity (a perfect circle), probability of use by waders was >0.90. While maximizing isolation from adjacent open water and preventing emergent vegetation establishment may reduce impoundment use by wading birds, even small impoundments will still constitute major attractants to these species. Given their risk to civil aviation (Dolbeer et al. 2000), even small impoundments may require exclusion devices, hazing or lethal control to effectively reduce use by wading birds. 23 Blackbirds (Icteridae & Sturnidae spp.) and doves (Columbidae spp.) were also frequently encountered across my sample impoundments (Table 2) and present a risk to civil aviation (Dolbeer et al. 2000). My model output suggests that impoundment use by both groups may be reduced through complete drainage of ISR, maximizing bank slope and minimizing vegetation diversity. However, both groups are generally abundant across North America and common across urbanized areas (Otis et al. 2008, Yasukawa & Cercy 1995). As with wading birds, effectively reducing their use of impoundments may require more traditional wildlife damage management techniques such as harassment and lethal control (Conover 2002) in addition to the design recommendations I offer here. The vegetation and landscape indices I developed for this study were my attempt to develop metrics to describe bird use of impoundments, which could be produced by airport managers and other stakeholders with limited technical capabilities. However, these metrics probably do not describe bird use any more efficiently than other existing metrics. For instance, describing the diversity of vegetation types in an impoundment basin does not describe the contribution of a specific vegetation type (i.e. herbaceous cover) to impoundment use by a given guild. A measurement of native versus exotic plant cover might be a better alternative, as it is known to influence the composition of avian communities and the prey bases (Burghardt et al. 2008). Even mean vegetation height may be a more logistically feasible metric for use in or around airport environments (Millroy 2007, Washburn & Seamans 2007). In future, measures of housing density (Pidgeon et al. 2007) or canopy cover (MacGregor-Fors 2008) may be adequate to describe avian community assemblages at stormwater impoundments in developed landscapes, as the utility of these metrics has already been demonstrated (Cavia et al. 2009). In future it may be valuable to relate impoundment density across the landscape to these metrics to estimate (a) 24 the frequency of impoundment use as a portion of available suburban habitat and (b) the correlation of impoundment density with urban/suburban avian community abundance and composition. Managing Stormwater Runoff The difference in bird-habitat associations between foraging guilds represented in my study demonstrates the complex challenge faced by urban planners and airport managers in addressing bird-strike hazards from multiple avian guilds. The property owners and municipalities bordering airports, as well as airport managers, should take caution to insure that application of design recommendations to deter one guild does not encourage impoundment use by another. I suggest urban and airport planners prioritize their designs for stormwater- management methods and the potential attraction of birds relative to those species with the greatest percentage of total strikes that cause some form of damage (either direct aircraft damage or an effect on flight) for the airport?s geographic region (as per Devault et al. unpublished manuscript). For example, DeVault et al. (In review) note that 10 of the 15 most hazardous bird species or species groups are strongly associated with water (e.g., waterfowl and gulls [Larus spp.]). Complete stormwater drainage (detention pond) over a short period (e.g., 48 hours; FAA 2007) would likely reduce the probability of use by many aquatic foragers, simply by preventing establishment of fish, amphibian and invertebrate assemblages that serve as a prey base for some of the foraging guilds observed in this study (De Graaf et al. 1985). Ignoring water itself as an attractant, the establishment of aquatic food resources might be autocorrelated with pond size, as increasing pond area showed a strong positive correlation with probability of use across all guilds. Most ponds which drained completely in this study were relatively small (<1,000 m2) 25 compared to the largest ponds (?2 km), which were generally designed as retention ponds (continuously wet). Control of ISR via Low Impact Development techniques (http://cfpub.epa.gov/npdes/home.cfm?program_id=6), minimizing pond surface area, or complete draw down (Blackwell et al. 2008) will minimize bird use across multiple foraging guilds. ISR management using split-flow theory (Echols 2008) may also help to reduce open water available to hazardous wildlife species by more closely aligning post-construction ISR volumes with pre- development levels. However, the influence of such a system on bird use of ISR has not yet been tested or observed in the field. In addition, during the summer of 2006, I observed a sustained drop in bird use, across multiple foraging guilds, of a retention impoundment that was intentionally dyed by the land manager. It is unclear if the birds in this incident were responding to the effect of the dye, or other features of the pond. This dye may have reduced bird use by increasing turbidity, a premise that has been suggested but not tested (Glahn et al. 2000). It may also be possible that this dye altered the ultra violet reflectance of the pond in such a way as to make birds averse to its appearance. We know that UV reflectance has been demonstrated to influence avian foraging decisions (Koivula & Vittala 1999), although such an effect has not yet been tested on avian use of water resources. In future it will be important to bird-strike management to investigate dye use in impoundments to determine if dye can consistently influence bird use of water. To do so, it will be imperative to determine (a) the mechanism by which artificial dyes influence bird use of water resources (e.g. turbidity vs. spectral properties), (b) the visual configuration of the targeted avian species, particularly visual traits likely important to habitat selection and foraging, and (c) the logistical and economic viability of dye to reduce bird use of impoundments. 26 I am confident in the rigor of my observational methodology. I believe my walking survey of each pond?s perimeter was adequate to correct my initial count for species or individuals that did not flush upon my initial approach. Furthermore, those species which are of greatest concern for stakeholders in the bird-strike issue (e.g. geese, dabbling ducks, wading birds, etc.) are conspicuous and readily flush when approached. Furthermore, even if detection probability varied between the guilds I analyzed, I have no reason to believe it varied within guilds. I am planning a future analysis describing avian diversity at impoundments as a function of impoundment design and landscape characteristics. Some of the genera of greatest interest in this analysis (e.g. native passerines) were undoubtedly harder to flush and detect. Therefore, it will be necessary to account for variation in detection probability between avian guilds in this future analysis. The influence of precipitation on bird use of impoundments warrants further investigation. In future I may conduct a post hoc analysis in which I try to determine the extent to which the effect of precipitation I present here reduced un-modeled heterogeneity in my model set. I incorporated an effect of precipitation on detection in my models based only on my observation of reduced bird use of impoundments immediately after rainfall events in the weeks leading up to my observation period. My literature review has found no similar anecdotal data. While urban or suburban areas may offer viable habitat for some wildlife (e.g. Dykstra et al. 2001, Garaffa et al. 2009, Holevinski et al. 2007), there is little information on how their selection of anthropogenically modified habitats may be influenced by temporal variation in the availability of unmodified habitats (e.g. season; Caula 2008). It is my theory that birds in my study area preferentially occupied remnant ephemeral wetland resources when rainfall events replenished these resources. Conversely, I believe that birds in this study area used 27 impoundments when periods of little or no measureable rainfall reduced the availability of ephemeral wetland resources. A future effort should be made to investigate (1) the distribution of wetland-utilizing species across both constructed and natural wetlands in urbanized areas with respect to rainfall events and (2) the diversity of bird use of stormwater impoundments as a portion of all wetland resources in an urbanized landscape. Summary My study represents an improvement in the scale at which habitat management recommendations can be made with respect to differences in foraging guilds. In particular, quantifying changes in probability of use between seasons will allow airport managers to adjust their management priorities not just by avian guild but also by season for each guild. The frequency of bird strikes seems poised to continue growing, especially in the southeastern United States. The FAA (2010) continues to forecast steadily growing air traffic, while rapid urban and suburban growth is forecast for the southeastern United States (White et al. 2009). This urban expansion will carry with it the proliferation of stormwater impoundments and anthropogenic resources that sustain many hazardous bird species (Belant 1997, Burghardt et al. 2009, Chace & Walsh 2006, Conover 2002, Dykstra et al. 2001, Tilton 1995). As interest in stormwater management for wildlife attraction continues to grow (Brand & Snodgrass 2009, Davis et al. 2008, LeViol et al. 2009), airport biologists, researchers and other concerned stakeholders must work to ensure that the bird- strike issue remains in the forefront of this discussion (Blackwell et al. 2009). I suggest that future stormwater impoundments within the FAA siting criteria in the southeastern United States be designed with the steepest banks possible and present minimum 28 surface area. Furthermore, these impoundments must drain completely of water between run-off event and be situated so as to maximize their distance from other open water sources. 29 LITERATURE CITED Adams, L.W., L.E. Dove & T.M. Franklin. 1985. Mallard pair and brood use of urban stormwater-control impoundments. Wildlife Society Bulletin 13:46-51 Alabama Gap Analysis Project. Acessed March 1, 2008. Alabama Ornithological Society. 2006. Field checklist of Alabama birds. Accessed January 1, 2008. Alabama State Water Program. Geo-spatial Data. Accessed March 1, 2008. Andersen, D.C., J.J. Sartoris, J.S. Thullen & P.G. Reusch. 2003. The effects of bird use on nutrient removal in a constructed wastewater-treatment wetland. Wetlands 23:423-435. Anderson, D.R., K.P. Burnham & W.L. Thompson. 2000. Null hypothesis testing: Problems, prevalence, and an alternative. Journal of Wildlife Management 64:912-923. Baier, J.H. J. Holloway, R. Hulcher, T. Logiotatos, J.L. Johnson, E.L. Norton, P.L. Oakes, T. Paglione, K.M. Rogers, B.D. Smith, E.D. Surrency, & J. Thurmond. 2003. Erosion Control, Sediment Control and Stormwater Management on Construction Sites and Urban Areas. Alabama Soil & Water Conservation Committee. Montgommery, Alabama. Bancroft, G.T., D.E. Gawlik & K. Rutchey. 2002. Distribution of wading birds relative to vegetation and water depths in the northern Everglades of Florida, USA. Waterbirds 25:265-391. 30 Barras, S.C. & T.W. Seamans. 2002. Habitat management approaches for reducing wildlife use of airfields. Proceedings of the 20th Vertebrate Pest Conference. University of California- Davis. Davis, CA. 309-315. Bart, J. & S. Earnst. 2002. Double sampling to estimate density and population trends in birds. Auk 119:36-45. Belant, J.L. 1997. Gulls in the urban environment: landscape-level management to reduce conflict. Landscape and Urban Planning 38:245-258. Benoit, L.K. & R.A. Askins. 2002. Relationship between habitat area and the distribution of tidal marsh bits. Wilson Bulletin 114:314-323. Blackwell, B.F., L.M. Schafer, D.A. Helon & M.A. Linnell. 2008. Bird use of stormwater- management ponds: decreasing avian attractants on airports. Landscape and Urban Planning 86:162-170. Blackwell, B.F., T.L. DeVault, E. Fernandez-Juricic, R.A. Dolbeer. 2009. Wildlife collisions with aircraft: a missing component of land-use planning for airports. Landscape and Urban Planning 93:1-9. Blackwell, B.F., Seamans, T.W., Schmidt, P., DeVault, T.L., Belant, J., Whittingham, M.J., Martin, J.A., Fern?ndez-Juricic, E.. Airports, grasslands, and birds: a framework for management amidst conflicting priorities. In review. Ecological Applications. Blair, R.B. 1996. Land use and avian species diversity along an urban gradient. Ecological Applications 6:506-519. Brand, A.B. & J.W. Snodgrass. 2009. Value of artificial habitats for amphibian reproduction in altered landscapes. Conservation Biology 24:295-301. 31 Brown, M. & J.L. Dinsmore. 1986. Implications of marsh size and isolation for marsh bird management. Journal of Wildlife Management. 50:392-397. Burghardt, K.T., D.W. Tallamy, W.G. Shriver. 2009. Impact of native plants on bird and butterfly biodiversity in suburban landscapes. Conservation Biology 23:219-224. Burnham, K.P. & D.R. Anderson. 2002. Model Selection and Multimodel Inference- A Practical Information-Theoretic Approach, 2nd Edition. Springer. New York, NY. Carbaugh, J.S., D.L. Combs & E.M. Dunton. 2010. Nest-site selection and nesting ecology of giant Canada geese in central Tennessee. Human-Wildlife Interactions 4:207-212. Caula, S., P. Marty & J.L. Martin. 2008. Seasonal variation in species composition of an urban bird community in Mediterranean France. Landscape and Urban Planning 87:1-9. Cavia, R., G.R. Cueto, O.V. Su?rez. 2009. Changes in rodent communities according to the landscape structure in an urban ecosystem. Landscape and Urban Planning 90:11-19. Chace, J.F. & J.L. Walsh. 2006. Urban effects on native avifauna: a review. Landscape and Urban Planning 74:46-69. Cicero, C. 1989. Avian community structure in a large urban park: controls of local richness and diversity. Landscape and Urban Planning 17:221-240. City of Auburn. 2010. City of Auburn community profile. City of Auburn Economic Development Department. Auburn, AL. Cleary, E.C. & Dolbeer, R.A.. 2005. Wildlife Hazard Management at Airports, second edition. Federal Aviation Administration, Office of Airport Safety and Standards, Airport Safety and Compliance Branch, Washington, DC, USA. 32 Commission for Environmental Cooperation. 1997. Ecological regions of North America: towards a common perspective. Communications and Public Outreach Department of the CEC Secretariat. Montreal, Canada. Conover, M. 2002. Resolving Human-Wildlife Conflicts. CRC Press, Inc. Boca Raton, FL. Croci, S., A. Butet & P. Clergeau. 2008. Does urbanization filter birds on the basis of their biological traits? The Condor 110:223-240. Davis, D.E., C.H. Hanson & R.B. Hansen. 2008. Constructed wetland habitat for American avocet and black-necked stilt foraging and nesting. Journal of Wildlife Management 72:143-151. DeGraaf, R.M., N.G. Tilghman & S.H. Anderson. 1985. Foraging guilds of North American birds. Environmental Management 9: 493-536. DeVault, T. L., J. L. Belant, B. F. Blackwell, and T. W. Seamans. Interspecific variation in wildlife hazards to aircraft: implications for airport wildlife management. Journal of Wildlife Management. In review. Dieter, C.D., B.J. Anderson, J.S. Gleason, P.W. Mammenga & S. Vaa. 2010. Late summer movements by giant Canada geese in relation to a September hunting season. Human-Wildlife Interactions. 4:232-246. Dolbeer, R.A. 2006. Height distribution of birds as recorded by collisions with civil aircraft. Journal of Wildlife Management 70:1345-1350. Dolbeer, R.A., S.E. Wright & E.C. Cleary. 2000. Ranking the hazard level of wildlife species to aviation. Wildlife Society Bulletin 28:372-378. Dolbeer, R.A., S.E. Wright, J. Weller and M.J. Begier. 2009. Wildlife strikes to civil aircraft in the United States 1990-2008. U.S. Department of Transportation, Federal Aviation 33 Administration, National Wildlife Strike Database Serial Report Number 15. Office of Airport Safety and Standards, Airport Safety & Certification, Washington, DC. Duffield, J.M. 1986. Waterbird use of a urban stormwater wetland system in central California, USA. Colonial Waterbirds 9:227-235. Dunton, E.M. & D.L. Combs. 2010. Movements, habitat selection, associations and survival of giant Canada goose broods in central Tennessee. Human-Wildlife Interactions 4:192-201. Dykstra, C.R., J.L. Hays, B. Daniel & M. Simon. 2001. Home range and habitat use of suburban red-shouldered hawks in southwestern Ohio. The Wilson Bulletin 113:308-316. Echols, S. 2008. Split-flow theory: Stormwater to emulate natural landscapes. Landscape and Urban Planning 85:205-214. ESRI, Inc. 2006. ArcMap v9.2. Redlands, CA. Fairbairn, S.E. & J.J. Dinsmore. 2001. Local and landcape-level influences on wetland bird communities of the prairie pothole region of Iowa, USA. Wetlands 21:41-47. Federal Aviation Administration. 2007. Hazardous wildlife attractants on or near airports. Advisory Circular 150/5200-33B. Federal Aviation Administration. 2010. FAA Aerospace Forecast Fiscal Years 2010-2030. US Department of Transportation, Federal Aviation Administration, Aviation Policy & Plans. Washington, DC. Garaffa, P.I., J. Filloy & M.I. Bellocq. 2009. Bird community responses along urban-rural gradients: Does the size of the urban area matter? Landscape and Urban Planning 90:33- 41. Glahn, J. F., M. E. Tobin, AND B. F. Blackwell, editors. 2000. A science-based initiative to manage double-crested cormorant damage to southern aquaculture. USDA Animal and 34 Plant Health Inspection Service, Wildlife Services National Wildlife Research Center, Fort Collins, CO, APHIS 11-55-010. Hackett, S.J., R.T. Kimball, S. Reddy, R.C.K. Bowie, E.L. Braun, M.J. Braun, J.L. Chojnowski, W.A. Cox, K.L. Han, J. Harshman, C.J. Huddleston, B.D. Marks, K.J. Miglia, W.S. Moore, F.H. Sheldon, D.W. Steadman, C.C. Witt & T. Yuri. 2008. A phylogenetic study of birds reveals their evolutionary history. Science 320:1763-1768. Hobaugh, W.C. & J.G. Teer. 1981. Waterfowl use characteristics of flood-revention lakes in north-central Texas. Journal of Wildlife Management 45:16-26. Holevinski, R.A., P.D. Curtis & R.A. Malecki. 2007. Hazing of Canada geese is unlikely to reduce nuisance populations in urban and suburban communities. Human-Wildlife Conflicts. 1:257-264. Hostetler, M., S. Duncan & J. Paul. 2005. Post-construction effects of an urban development on migrating, resident and wintering birds. Southeastern Naturalist 4:421-434. Koivula, M. & J. Viitala. 1999. Rough-legged buzzards use vole scent marks to assess hunting areas. Journal of Avian Biology. 30:329-332. Lee, J.S. & M.H. Li. 2009. The impact of detention basin design on residential property value: Case studies using GIS in the hedonic price modeling. Landscape and Urban Planning 89:7-16. Le Viol, I., J. Mocq, R. Julliard & C. Kebiriou. 2009. The contribution of motorway stormwater retention ponds to the biodiversity of aquatic macroinvertebrates. Biological Conservation 42:3163-3171. Linnell, M.A., M.R. Conover & T.J. Ohashi. 1996. Analysis of bird strikes at a tropical airport. Journal of Wildlife Management 60:935-945. 35 MacGregor-Fors, I. 2008. Relation between habitat attributes and bird richness in a western Mexico suburb. Landscape and Urban Planning. 84:92-98. Mathworks Inc., 2010. Matlab? & Simulink? Student Version 7.10.0.499. Natick, MA. McGuckin, C.P. & R.D. Brown. 1995. A landscape ecological model for wildlife enhancement of stormwater management practices in urban greenways. Landscape and Urban Planning 33:227-246. Millroy, A.G. Impacts of Mowing on bird abundance, distribution and hazards to aircraft at Westover Air Reserve Base, Massachusetts. MS Thesis. University of Massachusetts Amherst, Amherst, 2007. National Oceanic and Atmospheric Administration. 2010. National Climactic Data Center. Accessed August 11, 2010. Otis, David L., John H. Schulz, David Miller, R. E. Mirarchi and T. S. Baskett. 2008. Mourning Dove (Zenaida macroura), The Birds of North America Online (A. Poole, Ed.). Ithaca: Cornell Lab of Ornithology; Retrieved from the Birds of North America Online: . Pidgeon, A.M., V.C. Radeloff, C.H. Flather, C.A. Lepczyk, M.K. Clanton, T.J. Hawbaker & R.B. Hammer. 2007. Associations of forest bird species richness with housing and landscape patterns across the USA. Ecological Applications. 17:1989-2010. Ricklefs, R.E. 1990. Ecology, 3rd edition. W.H. Freeman, New York, NY. Seamans, T.W., S.C. Barras, G.E. Bernhardt, B.F. Blackwell & J.D. Cepek. 2007. Comparison of 2 vegetation-height management practices for wildlife control at airports. Human- Wildlife Conflicts 1:97-105. 36 Sharpe, P.J. 2005. Two birds with one stone- the Fairview Church wetland mitigation case study. Wetland Science and Practice 22:20-24. Sibley, D.A. 2000. The Sibley Guide to Birds. Alfred A. Knopf, Inc. New York, NY. Sibley, D.A. 2001. The Sibley Guide to Bird Life and Behavior. Alfred A. Knopf, Inc. New York, NY. Sparling, D.W., J.D. Eisemann & W. Kuenzel. 2004. Contaminant exposure and effects in Red- winged Blackbirds inhabiting stormwater retention ponds. Environmental Management 33:719-729. --. 2007. Nesting and foraging behavior of Red-winged Blackbirds in stormwater wetlands. Urban Ecosystems 10:1-15. Steen, D.A., J.P. Gibbs & S.T. Timmermans. 2006. Assessing the sensitivity of wetland bird communities to hydrologic change in the eastern Great Lakes region. Wetlands 26:605- 611. Stout, W.E., S.A. Temple & J.R. Cary. 2006. Landscape features of red-tailed hawk nesting habitat in an urban/suburban environment. The Journal of Raptor Research 40:181-192. Terman, M.R. 1997. Natural links: naturalistic golf courses as wildlife habitat. Landscape and Urban Planning 38:183-197. Tilton, D.L. 1995. Integrating wetlands into planned landscapes. Landscape and Urban Planning 32:205-209. Traut, A.H. & M.E. Hostetler. 2003. Urban lakes and waterbirds: effects of development on avian behavior. Waterbirds 26:290-302. Washburn, B.E. & T.W. Seamans. 2007. Wildlife responses to vegetation height management in cool-season grasslands. Rangeland Ecology Management 60:319-323. 37 White, C.L. & M.B. Main. 2005. Waterbird use of created wetlands in golf-course landscapes. Wildlife Society Bulletin 33:411-421. White, E.M., A.T. Morzillo & R.J. Alig. 2009. Past and projected rural land conversion in the US at state, regional, and national levels. Landscape and Urban Planning 89:37-48. Yasukawa, Ken and William A. Searcy. 1995. Red-winged Blackbird (Agelaius phoeniceus), The Birds of North America Online (A. Poole, Ed.). Ithaca: Cornell Lab of Ornithology; Retrieved from the Birds of North America Online: . Zakrajsek, Edward J. & J.A. Bissonette. 2005. Ranking the risk of wildlife species hazardous to military aircraft. Wildlife Society Bulletin 33 258-264. 38 Table 1. An a priori model set of 9 hypotheses to describe avian use of stormwater impoundments in the Southeastern United States. These models estimate the probability of impoundment use by a specified guild given the observed data. # Modela K 1 int 3 2 int + Area + Irreg + OW:EV + Isol + Spring + Summer + Fall 10 Adams et al. 1985, Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Caula et al. 2008, Cicero 1989, Duffield 1986, Dunton & Combs 2010 3 int + Isol + Irreg + Spring + Summer + Fall 8 Adams et al. 1985, Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Caula et al. 2008, Cicero 1989, Duffield 1986, Dunton & Combs 2010 4 int + Irreg + Veg + Spring + Summer + Fall 8 Adams et al. 1985, Bancroft et al. 2002, Blackwell et al. 2008, Carbaugh et al. 2010, Caula et al. 2008, Cicero 1989, Cleary & Dolbeer 2005, Duffield 1986, Steen et al. 2006 5 int + Area + Irreg + OW:EV + Slope + Veg + Spring + Summer + Fall 11 Adams et al. 1985, Bancroft et al. 2002, Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Caula et al. 2008, Cicero 1989, Cleary & Dolbeer 2005, Duffield 1986, Steen et al. 2006 6 int + Type + Slope + Veg + Spring + Summer + Fall 9 Adams et al. 1985, Bancroft et al. 2002, Caula et al. 2008, Cicero 1989, Cleary & Dolbeer 2005, Duffield 1986, Dunton & Combs 2010, Steen et al. 2006 7 int + Type + Irreg + Slope + Veg + Spring + Summer + Fall 10 Adams et al. 1985, Bancroft et al. 2002, Blackwell et al. 2008, Carbaugh et al. 2010, Caula et al. 2008, Cicero 1989, Cleary & Dolbeer 2005, Duffield 1986, Dunton & Combs 2010, Steen et al. 2006 8 int + Area + OW + Isol + Spring + Summer + Fall 9 Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Caula et al. 2008, Cicero 1989, Duffield 1986, Dunton & Combs 2010, Steen et al. 2006 9 int + Area + Landscape + Spring + Summer + Fall 8 Benoit & Askins 2002, Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Caula et al. 2008, Cicero 1989, Croci et al. 2008, Duffield 1986, Hostetler et al. 2005, Traut & Hostetler 2003 39 Table 1 (continued). An a priori model set of 9 hypotheses to describe avian use of stormwater impoundments in the Southeastern United States. These models estimate the probability of impoundment use by a specified guild given the observed data. # Modela K 10 int + Area + Irreg + OW:EV + Isol 7 Adams et al. 1985, Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Cicero 1989, Duffield 1986, Dunton & Combs 2010 11 int + Isol + Irreg 5 Adams et al. 1985, Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Cicero 1989, Duffield 1986, Dunton & Combs 2010 12 int + Irreg + Veg 5 Adams et al. 1985, Bancroft et al. 2002, Blackwell et al. 2008, Carbaugh et al. 2010, Cicero 1989, Cleary & Dolbeer 2005, Steen et al. 2006 13 int + Area + Irreg + OW:EV + Slope + Veg 8 Adams et al. 1985, Bancroft et al. 2002, Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Cicero 1989, Cleary & Dolbeer 2005, Duffield 1986, Steen et al. 2006 14 int + Type + Slope + Veg 6 Adams et al. 1985, Bancroft et al. 2002, Cicero 1989, Cleary & Dolbeer 2005, Duffield 1986, Dunton & Combs 2010, Steen et al. 2006 15 int + Type + Irreg + Slope + Veg 7 Adams et al. 1985, Bancroft et al. 2002, Blackwell et al. 2008, Carbaugh et al. 2010, Cicero 1989, Cleary & Dolbeer 2005, Duffield 1986, Dunton & Combs 2010, Steen et al. 2006 16 int + Area + OW + Isol 6 Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Cicero 1989, Duffield 1986, Dunton & Combs 2010, Steen et al. 2006 17 int + Area + Landscape Benoit & Askins 2002, Blackwell et al. 2008, Brown & Dinsmore 1986, Carbaugh et al. 2010, Cicero 1989, Croci et al. 2008, Duffield 1986, Hostetler et al. 2005, Traut & Hostetler 2003 5 aModel parameter definitions: int = model intercept (?0), Type = basin design (retention vs. detention), Area = mean impoundment surface area, Irreg = mean perimeter irregularity of impoundment surface area, OW:EV = mean ration of open water to emergent vegetation, OW = total area of open water resources within 1 km of an impoundment, Isol = minimum distance from an impoundment 40 to the nearest open water resource, Slope = mean impoundment bank slope, Veg = vegetation diversity index, Landscape = landscape diversity index, Spring Summer & Fall = season 41 Table 2. Summary of total encounters by avian guild at stormwater impoundments during this study. Any guild detected on less than 20 intervals during the study was excluded from analysis. Mean abundance represents the average number of individuals of each guild observed weekly at all 40 impoundments in this study. Total ponds occupied represents the number of ponds occupied at any time during this study be each guild. Guild Encounters Mean Weekly Abundance Maximum Mean Daily Counta Total Ponds Occupied Aerials 142 0.5 42 31 Anserinids 68 0.5 29.7 12 Blackbirds 361 2.7 400 36 Brights 367 0.8 45.5 40 Corvids 179 0.4 12 35 Cuckoos 0 0.0 0 0 Dabbling Ducks 111 0.7 43.4 14 Divers 9 0.0 1 6 Diving Ducks 13 0.1 24.1 7 Domestic & Exotic Waterfowl 70 1.8 71.3 5 Doves 209 0.5 28 38 Flycatchers 210 0.3 5 35 Gamebirds 5 0.0 2 4 Goatsuckers 1 0.0 1 2 Kingfishers 100 0.1 2 20 Longtail Ground Birds 414 0.8 57 40 Open Ground Birds 5 0.0 1 8 Pelicaniformes 1 0.0 1.2 2 Raptors 41 0.1 10.5 23 Shorebirds 122 0.3 11 23 Shrikes 5 0.0 1 1 Small Forest Birds 151 0.3 12 29 Sparrows 278 0.8 24 39 Vireos 2 0.0 1 2 Waders 210 0.2 3.7 29 Warblers 148 0.4 14.3 33 Waxwings 14 0.4 111 13 Woodpeckers 19 0.0 4 13 aMean daily abundance was calculated for each guild across all ponds. This value represents the largest observed value across the 104 week survey period. 42 Table 3. Summary statistics for the 40 stormwater impoundments at which I conducted field observations for two years. Variable Mean Value + Standard Deviation Min/Max Values Surface Area 0.41 km2 ? 0.64 <0.01 km2 / 2.76 km2 Shoreline Irregularity 1.41 ? 0.27 0.70 / 2.10 OW:EV 0.48 ? 0.75 0.00 / 3.40 Area of adjacent OW 0.79 km2 ? 0.45 0.00 km2 / 2.02 km2 Pond Isolation 0.35 km ? 0.35 0.02 km / 1.47 km Bank Slope 0.41 ? 0.17 0.17 / 1.00 Vegetation Diversity Index 1.09 ? 0.37 0.10 / 1.68 Landscape Diversity Index 1.13 ? 0.27 0.50 / 1.57 43 Figure 1. Study area and sample site distribution; this study took place in Lee County, Alabama, USA. My sample sites were distributed across the Auburn-Opelika Metropolitan area. 44 APPENDIX 1: Avian Guild Selection 45 Genera appearing in bold were observed utilizing sample ponds during the study. Anatid Guilds 1. Anserinids- 1400-6800g anatids highly evolved for an aquatic existence; this particular group is characterized by large bodies with long necks and herbivorous forgaing ecology. This guild includes species that forage both by dabbling and grazing. a. Anseriformes i. Anatidae (Anserinae) 1. Anser Greater White-fronted Goose 2. Branta Typical Geese 3. Chen Snow Goose 4. Cygnus Swans 2. Dabbling Ducks- 600-1200g anatids highly evolved for an aquatic existence; this particular group is characterized by having legs that are placed farther forward on the body to allow greater mobility on land. This allows for these species to take off directly from the water. These species include herbivorous, granivorous, and omnivorous species that employ a wide variety of foraging strategies including dabbling, grazing, and straining. a. Anseriformes i. Anatidae (Anatinae) 1. Aix Wood Duck 2. Anser Typical Ducks 3. Diving Ducks- 380-1500g anatids highly evolved for an aquatic existence; this particular group is characterized by having legs that are placed far back on the body to aid in diving. This places a physiological restraint on terrestrial mobility and most of these species require a running start to exit the water. These species include omnivorous, crustaceovorous, insectivorous, molluscovorous, piscivorous, and herbivorous species that are all bottom foragers and gleaners. a. Anseriformes i. Anatidae (Anatinae) 1. Aythya Diving Ducks 2. Bucephala Buffleheads and Goldeneyes 3. Clangula Long-tailed Duck 4. Lophodytes Hooded Merganser 5. Melanitta Scoters 6. Mergus Merganser 7. Nomonyx Masked Duck 8. Oxyura Ruddy Duck 46 4. Domestic and Exotic Ducks and Geese- 2700-9000g domesticated Anatids; typically heavy-bodied granivores and herbivores dependant on, or habituated to, anthropogenic food sources. a. Anseriformes i. Anatidae (Anserinae) 1. Anser anser Grayleg (Barnyard) Goose 2. Anser cygnoides Swan (Chinese) Goose ii. Anatidae (Anatinae) 1. Anas platyrhynchos Domestic Mallard or Pekin Duck 2. Cairina moschata Domestic Muscovy Aquatic Guilds 1. Divers- 300-4100g birds highly adapted for swimming and diving; they are characterized by having legs that are set farther back on the body to aid in propulsion. These species are piscivorous freshwater divers. a. Gaviiformes i. Gaviidae 1. Gavia Loons b. Podicipediformes i. Podicipedidae 1. Podiceps Typical Grebes 2. Podilymbus Pied-billed Grebe 2. Kingfisher- ~150g bird species separated from the other guilds for its unique foraging behavior (piscivorous aerial diver) and conspicuous plumage a. Coracciformes i. Alcedinidae 1. Ceryle Belted Kingfisher 3. Pelicaniforms- ~1250g birds characterized by totipalmate feet and a bare throat patch a. Pelicaniformes i. Anhingidae 1. Anhinga Anhinga ii. Phalacrocoracidae 1. Phalacrocorax Cormorants 4. Shorebirds-20-1200g birds extremely varied in aspects of morphology and behavior, but primarily non-herbivorous shoreline or tidal zone feeders a. Charadriiformes i. Charadriidae 1. Charadrius Small Plovers and Killdeer 2. Pluvialis Large Plovers ii. Laridae 47 1. Larus Gulls iii. Scolopacidae 1. Actitis Spotted Sandpiper 2. Bartramia Upland Sandpiper 3. Calidris Sandpipers 4. Gallinago Wilson?s Snipe 5. Limnodromus Dowitcher 6. Scolopax American Woodcock 7. Tringa Yellowlegs 8. Tryngites Buff-breasted Sandpiper iv. Sternidae 1. Chlidonias Black Tern 2. Sterna Typical Terns 5. Waders- 85-5000g birds that inhabit areas near water and exhibit a wide array of foraging and social behavior including a convergent group of long-legged, large-billed wading birds; Ardeiforms, ciconiiforms, and threskiornithiforms include carnivorous, crustaceovorous, insectivorous, molluscovorous, and piscivorous species that employ water ambushing, ground gleaning, water straining, and mud gleaning & probing foraging strategies. Gruiforms include crustaceovorous, insectivorous, molluscovorous, and omnivorous species. These birds utilize gleaning, probing, dabbling and diving in freshwater to saltwater wetlands and tidal zones. a. Pelicaniformes i. Ardeidae 1. Aredea Greater Egrets and Herons 2. Botaurus American Bitterns 3. Bubulcus Cattle Egret 4. Butorides Green Heron 5. Egretta Lesser Egrets and Herons 6. Ixobrychus Least Bittern 7. Nyctanassa Yellow-crowned Night Heron 8. Nycticorax Black-crowned Night Heron b. Ciconiiformes i. Ciconiidae 1. Eudocimus White Ibis 2. Mycteria Wood Stork c. Gruiformes i. Gruidae 1. Grus Sandhill Crane ii. Rallidae 48 1. Fulica American Coot 2. Gallinula Common Moorhen 3. Laterallus Black Rail 4. Porphyrio Purple Gallinule 5. Porzana Sora 6. Rallus Typical Rails Terrestrial Guilds 1. Game Birds- 130-7400g ground-dwelling birds that also forage on the ground a. Galliformes i. Odontophoridae 1. Colinus Northern Bobwhite ii. Phasianidae 1. Meleagris Wild Turkey 2. Cuckoos- 50-65g perching birds insectivorous in nature and solitary in habit a. Cuculiformes i. Cuculidae 1. Coccyzus Cuckoos 3. Doves- 30-270g robust-bodied perching granivorous birds with small heads and feet a. Columbiformes i. Columbidae 1. Columba Rock Dove 2. Columbina Common Ground Dove 3. Streptopelia Eurasian Collared Dove 4. Zenaida Mourning and White-winged Doves 4. Goatsuckers- Perching 50-120g crepuscular cryptic birds that are insectivorous in nature a. Caprimulgiformes i. Caprimulgidae 1. Caprimulgus Poor-wills 2. Chordeiles Nighthawk 5. Raptors-A paraphyletic group of 140g-4600g carnivorous/piscivorous/insectivorous birds characterized by sharp, hooked claws and beaks and refined binocular vision; their feeding strategies include both diurnal and nocturnal hunting and scavenging. a. Accipitriformes i. Accipitridae 1. Accipiter Accipiters 2. Aquila Golden Eagle 3. Buteo Buteos 4. Circus Northern Harrier 5. Elanoides Swallow-tailed Kite 6. Elanus White-tailed Kite 49 7. Haliaeetus Bald Eagle 8. Ictinia Mississippi Kite 9. Pandion Osprey b. Cathartiformes i. Cathartidae 1. Cathartes Turkey Vulture 2. Coragyps Black Vulture c. Falconiformes i. Falconidae 1. Falco Peregrine Falcon & American Kestrel d. Strigiformes i. Strigidae 1. Aegolius Northern Saw-whet Owl 2. Asio Eared Owls 3. Athene Burrowing Owl 4. Bubo Great Horned Owl 5. Otus Eastern Screech Owl 6. Strix Barred Owl ii. Tytonidae 1. Tyto Barn Owl 6. Woodpeckers- 27g-290g climbing birds that have hard, chisel-like bills used to obtain insects from underneath the bark of trees; these birds are also characterized by long stiff tails to help maintain balance and zygodactyls feet. a. Piciformes i. Picidae 1. Colaptes Northern Flicker 2. Drycopus Pileated Woodpecker 3. Melanerpes Food-storing Woodpeckers 4. Picoides Typical Woodpeckers 5. Sphyrapicus Yellow-bellied Woodpecker Passerine Guilds 1. Aeriels- 14-55g songbirds characterized by small feet and long wings relative to body length; all species are air salliers except for the frugivorous Tree Swallow. a. Apodiformes i. Apodidae 1. Chaetura Chimney Swift ii. Trochilidae 1. Archilochus Ruby-throated Hummingbird b. Passeriformes 50 i. Hirundinidae 1. Hirundo Barn Swallow 2. Petrochelidon Cave and Cliff Swallows 3. Progne Purple Martin 4. Riparia Bank Swallow 5. Stelgidopteryx Northern Rough-winged Swallow 6. Tachycineta Tree Swallow 2. Blackbirds- 20-215g songbirds that are gregarious, conspicuous and noisy; many prefer habitats close to water; all are dark in color and are ground foragers. a. Passeriformes i. Icteridae 1. Agelaius Red-winged and Tricolored Blackbirds 2. Euphagus Brewer?s and Rusty Blackbirds 3. Molothrus Brown-headed Cowbird 4. Squiscalus Grackles 5. Xanthocephalus Yellow-headed Blackbird ii. Sturnidae 1. Sturnus European Starling 3. Brights- 15-60g songbirds characterized by bright highly conspicuous plumage; this group includes granivores, frugivores and insectivores. a. Passeriformes i. Cardinalidae 1. Cardinalis Northern Cardinal 2. Passerina Indigo Bunting & Blue Grosbeak 3. Pheucticus Rose-breasted Grosbeak ii. Fringillidae 1. Carpodacus House and Purple Finch 2. Spinus American Goldfinch iii. Icteridae 1. Icterus Orioles iv. Thraupidae 1. Piranga Tanagers v. Turdidae 1. Sialia Eastern Bluebird 4. Corvids- 85-1200g conspicuous songbirds that are often aggressive toward smaller birds; they are omnivorous upper canopy and ground foragers. 51 a. Passeriformes i. Corvidae 1. Corvus Crows and Ravens 2. Cyanocitta Blue Jay 5. Flycatchers- 10-30g perching birds often identified by their habit of tail-dipping when perched; All are insectivorous air salliers except for the eastern phoebe and the great crested flycatcher, both of which are lower-canopy frugivores. a. Passeriformes i. Tyrannidae 1. Contopus Pewees 2. Empidonax Typical flycatchers 3. Myiarchus Great Crested Flycatcher 4. Sayornis Eastern Phoebe 5. Tyrannus Eastern Kingbird 6. Longtail Groundbirds- 30-80g songbirds that typically ground forage for insects; all members have long tails relative to their body mass. a. Passeriformes i. Emberizidae 1. Pipilo Eastern Towhee ii. Mimidae 1. Dumetella Gray Catbird 2. Mimus Northern Mockingbird 3. Toxostoma Brown Thrasher iii. Turdidae 1. Catharus Typical Thrushes 2. Hylocichla Wood Thrush 3. Turdus American Robin 7. Open Ground Birds- 20-90g songbirds that are generally drab in color and occupy relatively open ground in fields and meadows; all species are ground foragers. a. Passeriformes i. Aluadidae 1. Eremophila Horned Lark ii. Cardinalidae 1. Spiza Dickcissel iii. Emberizidae 1. Calcarius Lapland Longspur iv. Icteridae 52 1. Dolichonyx Bobolink 2. Sturnella Eastern Meadowlink v. Motacillidae 1. Anthus American Pipit 8. Shrikes- ~50g carnivorous songbirds with strongly hooked beaks. a. Passeriformes i. Laniidae 1. Lanius Loggerhead Shrike 9. Small Forest Birds- 6-20g songbirds that vary in foraging behaviors and appearance; all are tree dwellers and prefer canopy or dense brush to open ground. a. Passeriformes i. Certhiidae 1. Certhia Brown Creeper ii. Paridae 1. Baeolophus Tufted Titmouse 2. Poecile Carolina Chickadee iii. Regulidae 1. Regulus Kinglets iv. Sittidae 1. Sitta Nuthatches v. Slyviidae 1. Polioptila Blue-gray Gnatcatcher vi. Troglodytidae 1. Cistothorus Marsh and Sedge Wrens 2. Thryomanes Bewick?s Wren 3. Thryothorus Carolina Wren 4. Troglodyes House and Winter Wrens 10. Sparrows- 12-40g songbirds that are usually drab in color with conical bills; although varied in habitat preference, all are omnivorous, granivorous, or insectivorous ground feeders. a. Passeriformes i. Emberizidae 1. Aimophila Bachman?s Sparrow 2. Ammodramus Meadow Sparrows 3. Chondestes Lark Sparrow 4. Junco Dark-eyed Junco 5. Melospiza Marsh Sparrows 53 6. Passer House Sparrow 7. Passerculus Savannah Sparrow 8. Passerlla Fox Sparrow 9. Pooecetes Vesper Sparrow 10. Spizella Chipping, Clay-colored and Field Sparrows 11. Zonotrichia White marked Sparrows 11. Vireos-12-18g songbirds characterized by stocky bodies; large, hooked bills; and short legs; Members of this group are all canopy foragers. a. Passeriformes i. Vireonidae 1. Vireo Vireos 12. Wood Warblers- 7-20g songbirds that are highly active with short, pointed bills; all are canopy foragers. a. Passeriformes i. Parulidae 1. Dendroica Bright Warblers 2. Geothlypis Common Yellowthroat 3. Helmitheros Worm-eating Warbler 4. Icteria Yellow-breasted Chat 5. Limnothlypis Swainson?s Warbler 6. Mniotilta Black-and-white Warbler 7. Oporornis Connecticut, Kentucky, & Mourning Warblers 8. Parula Northern Parula 9. Protonotaria Prothonotary Warbler 10. Seiurus Ovenbirds and Waterthrushes 11. Setophaga American Redstart 12. Vermivora Drab Warblers 13. Wilsonia ?Water Thicket? Warblers 13. Waxwings- ~30g songbirds characterized by bright, sleek plumage and crest, with a short, yellow-tipped tail; they are both insectivorous air salliers and frugivorous upper canopy foragers. a. Passeriformes i. Bombycillidae 1. Bombycilla Cedar Waxwing 54 APPENDIX 2: Vegetation Guilds 55 1) Successional Guilds: These guilds generally follow the normal process of ecological succession. These guilds were selected for easy sight recognition to be utilized by airport managers without extensive backgrounds in ecology or botany. These guilds include both terrestrial and semi-aquatic guilds and include seed, nut, and fruit producing species. (a) Bare Rock: earliest stage of succession, or has factors that prevent soil pockets from forming and producing plants. (b) Bare Soil: Includes sand, mud, clay, or loam not supporting vegetation (c) Detritus: dominated by leaf litter and decaying woody debris (d) Archaic: dominated by primitive plants including mosses (bryophyte), ferns (Pteridophyta spp.), liverworts (Marchantiophyta spp.) and cycads (Cycadophyta spp.) (e) Grasses & Forbes: includes sedges (Cyperaceae spp.), rushes (Juncaceae spp.), and grasses (Poaceae spp.) except monoculture turfgrasses (Festuca spp.) that dominate dry soils. (f) Turfgrasses: short monoculture grasses (Festuca spp.) selected and managed for landscaping (g) Shrub/Seedling Trees: characterized by short dense woody vegetation and trees in the earliest stages of development (h) Saplings & Small Trees: composed of trees that do not produce fruits and nuts at full capacity or belong to smaller species; includes poletimber trees and saplings of DBH <10 inches. (i) Mature Trees: characterized by large mature trees, or trees with a DBH >10?; includes trees suitable for sawtimber. 2) True Aquatic Guilds: Unlike semi-aquatic components of the successional guilds, these guilds are truly dependent on water as a means for reproduction and growth and should not be confused with emergent vegetation covered in the above guilds. (a) Algae and Free-floating Flora: water cover that includes floating algae, duckweed (Lemnaceae spp.), and other vascular and non-vascular free- floating flora that are either rootless or whose roots are not attached to the substrate. (b) Aquatic Rooted Plants: floating submergent vegetation the bottom of the pond and roots itself in the sediment. This group includes water lilies. (c) Reeds & Their Allies: Vascular rooted plants that dominate wet soils and shallow water. Includes horsetails, (Equisetum spp.) cattails (Typha spp.), as well as some grasses (Poaceae spp.), sedges (Cyperaceae spp.), and rushes (Juncaceae spp.) that favor wet soils. (d) Aquatic Trees: Includes mangroves and cypress trees. (e) Open Water: Includes standing water not dominated by emergent or floating submergent vegetation. 3) Anthropogenic Guilds: These guilds encompass impervious surfaces of anthropogenic origins. (a) Impervious Surface: This includes all artificial impervious surfaces including lumber, concrete and asphalt. 56 APPENDIX 3: Bird Species Observed Using Impoundments In This Study 57 The following species were observed at my study sites during the study period. This list includes only species observed utilizing the study ponds and does not include bird species observed in the study area but not at the sample sites. Species are arranged alphabetically by common name (Alabama Ornithological Society 2006). Acadian Flycatcher Empidonax virescens Alder Flycatcher Empidonax alnorum American Black Duck Anas rubripes American Coot Fulica americana American Crow Corvus brachyrhynchos American Goldfinch Spinus tristis American Kestrel Falco sparverius American Redstart Setophaga ruticilla American Robin Turdus migratorius Anhinga Anhinga anhinga Bank Swallow Riparia riparia Barn Swallow Hirundo rustica Belted Kingfisher Ceryle alcyon Black and White Warbler Mniotilta varia Black Vulture Coragyps atratus Blackburnian Warbler Dendroica fusca Black-crowned Night Heron Nycticorax nycticorax Black-throated Green Warbler Dendroica virens Blue Grosbeak Passerina caerulea Blue Jay Cyanocitta cristata Blue-gray Gnatcatcher Polioptila caerulea Blue-headed Vireo Vireo solitarius Blue-winged Teal Anas discors Blue-winged Warbler Vermivora pinus Bobolink Dolichonyx oryzivorus Brewer's Blackbird Euphagus cyanocephalus Broad-winged Hawk Buteo platypterus Brown Thrasher Toxostoma rufum Brown-headed Cowbird Molothrus ater Brown-headed Nuthatch Sitta pusilla Bufflehead Bucephala albeola Canada Goose Branta canadensis Carolina Chickadee Poecile carolinensis Carolina Wren Thryothorus ludovicianus Cave Swallow Petrochelidon fulva Cedar Waxwing Bombycilla cedrorum Chimney Swift Chaetura pelagica Chipping Sparrow Spizella passerina Chuck Will's Widow Caprimulgus carolinensis Cliff Swallow Petrochelidon pyrrhonota 58 Common Grackle scalus quiscula Common Ground-Dove Columbina passerina Common Yellowthroat Geothlypis trichas Cooper's Hawk Accipiter cooperii Dickcissel Spiza americana Domestic Duck Anas spp. Domestic Goose Anser spp. Downy Woodpecker Picoides Eastern Bluebird Sialia sialis Eastern Kingbird Tyrannus tyrannus Eastern Meadowlark Sturnella magna Eastern Phoebe Sayornis phoebe Eastern Towhee Pipilo erythrophthalmus Eastern Wood Pewee Contopus virens Eurasian Collared Dove Streptopelia decaocto European Starling Sturnus vulgaris Field Sparrow Spizella pusilla Fox Sparrow Passerella iliaca Common Goldeneye Bucephala clangula Grasshopper Sparrow Ammodramus savannarum Gray Catbird Dumetella carolinensis Great Blue Heron Ardea herodias Great Crested Flycatcher Myiarchus crinitus Great Egret Ardea alba Greater Scaup Aythya marila Green Heron Butorides virescens Hairy Woodpecker Picoides villosus Henlsow Sparrow Ammodramus henslowii Hooded Merganser Lophodytes cucullatus House Finch Carpodacus mexicanus House Sparrow Passer domesticus House Wren Troglodytes aedon Indigo Bunting Passerina cyanea Killdeer Charadrius vociferus Laughing Gull Larus atricilla Least Flycatcher Empidonax minimus Least Sandpiper Calidris minutilla Lesser Scaup Aythya affinis Lincoln's Sparrow Melospiza lincolnii Little Blue Heron Egretta caerulea Loggerhead Shrike Lanius ludovicianus Louisiana Waterthrush Seiurus motacilla Mallard Anas platyrhynchos Mourning Dove Zenaida macroura Northern Bobwhite Colinus virginianus 59 Northern Cardinal Cardinalis cardinalis Northern Flicker Colaptes auratus Northern Mockingbird Mimus polyglottos Northern Parula Parula americana Northern Rough-winged Swallow Stelgidopteryx serripennis Northern Shoveler Anas clypeata Northern Waterthrush Seiurus noveboracensis Orange-crowned Warbler Vermivora celata Orchard Oriole Icterus spurius Ovenbird Seiurus aurocapillus Palm Warbler Dendroica palmarum Pied-billed Grebe Podilymbus podiceps Pine Siskin Spinus pinus Pine Warbler Dendroica pinus Prairie Warbler Dendroica discolor Prothonotary Warbler Protonotaria citrea Purple Finch Carpodacus purpureus Purple Martin Progne subis Red-bellied Woodpecker Melanerpes carolinus Red-breasted Merganser Mergus serrator Red-eyed Vireo Vireo olivaceus Redhead Aythya americana Red-shouldered Hawk Buteo lineatus Red-tailed Hawk Buteo jamaicensis Red-winged Blackbird Agelaius phoeniceus Ring-billed Gull Larus delawarensis Ring-necked Duck Aythya collaris Rock Pigeon Columba livia Rose-breasted Grosbeak Pheucticus ludovicianus Ruby-crowned Kinglet Regulus calendula Ruby-throated Hummingbird Archilochus colubris Rusty Blackbird Euphagus carolinus Savannah Sparrow Passerculus sandwichensis Semipalmated Plover Charadrius semipalmatus Snow Goose Chen caerulescens Snowy Egret Egretta thula Solitary Sandpiper Tringa solitaria Song Sparrow Melospiza melodia Spotted Sandpiper Actitis macularia Surf Scoter Melanitta perspicillata Swainson's Warbler Limnothlypis swainsonii Swamp Sparrow elospiza georgiana Tennessee Warbler Vermivora peregrina Tree Swallow Tachycineta bicolor Tricolored Heron Egretta tricolor 60 Tufted Titmouse Baeolophus bicolor Turkey Vulture Cathartes aura Vesper Sparrow Pooecetes gramineus White-crowned Sparrow Zonotrichia leucophrys White-eyed Vireo Vireo griseus White-throated Sparrow Zonotrichia albicollis Wild Turkey Meleagris gallopavo Willow Flycatcher Empidonax traillii Winter Wren Troglodytes troglodytes Wood Duck Aix sponsa Wood Thrush Hylocichla mustelina Yellow Warbler Dendroica petechia Yellow-bellied Sapsucker Sphyrapicus varius Yellow-throated Warbler Dendroica dominica Yellow-rumped Warbler Dendroica coronata 61 APPENDIX 4: Other Vertebrates Observed In This Study Using Stormwater Impoundments 62 The following species and genera were encountered incidentally during the study. Mammals Bats Vespertilionid spp. Beaver Castor canadensis Coyote Canis latrans Domestic Cat Felis sylvestris Domestic Dog Canis lupus Eastern Chimpmunk Tamias striatus Eastern Cottontail Slyvilagus floridanus Feral Swine Sus scrofa Gray Fox Urocyon cinereoargenteus Gray Squirrel Sciurus carolinensis Hispid Cotton Rat Sigmodon hispidus Muskrat Ondatra zibethicus Nine-banded Armadillo Dasypus novemcinctus Old World Mice Rattus spp. Raccoon Procyon lotor Red Fox Vulpes vulpes Shrews Sorex spp. Swamp Rabbit Sylvilagus aquaticus Virginia Opossum Didelphis virginiana White-tailed Deer Odocoileus virginianus Wood Mice Peromyscus spp. Woodchuck Marmota monax Amphibians American Bullfrog Lithobates catesbeiana Chorus Frogs Pseudacris spp. Eastern Narrow-mouthed Toad Gastrophryne carolinensis Green Frog Lithobates clamitans Northern Spring Peeper Pseudacris crucifer Slimy Salamander Plethodon glutinosis Southern Cricket Frog Acris gryllus Southern Leopard Frog Lithobates sphenocephalus Toads Anaxyrus spp. Treefrogs Hyla spp. Reptiles Black Racer Coluber constrictor Carolina Anole Anolis carolinensis Common Snapping Turtle Chelydra serpentina Cottonmouth Agkistrodon piscivorus Eastern Box Turtle Terrapene carolina Eastern Chicken Turtle Deirochelys reticularia Eastern Fence Lizard Sceloporus undulatus Eastern Mud Turtle Kinosternon subrubrum 63 Loggerhead Musk Turtle Sternotherus minor Painted Turtle Chrysemys picta Pond Slider Trachemys scripta Red-bellied Watersnake Nerodia erythrogaster erythrogaster Ring-necked Snake Diadophis punctatus River Cooter Pseudemys concinna Skinks Eumeces spp. Softshell Turtle Apalone spp. Stinkpot Sternotherus odoratus Water Snakes Nerodia spp. 64 Appendix 5: Analysis Results for A Priori Models Describing Bird Use of Stormwater Impoundments in the Southeastern United States 65 Aerials Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -31.07 (--) 1.42 (0.33) -32.14 (--) 0.79 (0.47) 0.18 (0.68) 0.67 (0.40) -16.65 (--) 1.42 (0.34) -0.41 (0.47) -21.99 (--) 0.69 (0.49) -0.16 (0.69) 0.76 (0.41) -54.19 (--) -2.39 (0.59) -31.30 (--) -2.47 (0.59) -0.61 (0.71) -33.11 (--) -0.09 (0.66) -18.62 (--) 0.36 (0.66) 0.07 (0.63) 0.95 (0.25) -0.66 (0.46) 0.91 (0.25) -0.31 (0.39) -1.13 (0.80) 0.52 (0.34) 0.09 (0.56) 0.43 (0.30) -2.34 (1.06) 0.46 (0.36) -0.16 (0.59) 0.51 (0.31) 0.28 (0.84) -1.79 (0.54) 0.89 (1.17) -1.83 (0.54) -0.42 (0.56) -1.92 (0.83) -0.16 (0.54) -0.79 (0.73) 0.22 (0.51) -0.88 (0.14) Averageb -29.15 (0.00) -0.05 (0.01) 1.25 (0.35) -0.05 (0.06) 0.01 (0.14) 0.14 (0.08) 66 Aerials Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -1.74 (0.67) 32.88 (--) 32.29 (--) -1.33 (0.60) 32.11 (--) 31.53 (--) -1.51 (0.67) 17.31 (--) 16.77 (--) 0.39 (1.12) 1.30 (0.55) 20.44 (--) 19.79 (--) -1.65 (1.06) -0.41 (0.74) 56.39 (--) 55.72 (--) -1.79 (1.07) -0.41 (0.74) 34.44 (--) 33.75 (--) 1.60 (0.50) 31.93 (--) 31.31 (--) -1.60 (0.57) 19.06 (--) 18.50 (--) -1.24 (0.55) -1.18 (0.58) -1.04 (0.53) 0.30 (0.92) 0.98 (0.47) -1.21 (0.88) -0.27 (0.58) -1.29 (0.88) -0.25 (0.59) 1.14 (0.43) -1.27 (0.51) Averageb -0.38 (0.17) -0.01 (0.09) 0.07 (0.05) -1.13 (0.43) 30.36 (0.00) 29.77 (0.00) 67 Aerials Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip 29.87 (--) -2.48 (1.09) 8 586.0 0.0 0.6484 1.0 29.07 (--) 0.23 (0.18) 10 589.1 3.1 0.1381 4.7 14.33 (--) -2.39 (1.09) 9 589.2 3.2 0.1283 5.1 17.40 (--) -2.57 (1.09) 11 590.6 4.7 0.0628 10.3 53.50 (--) -2.44 (1.09) 9 593.5 7.5 0.0150 43.1 31.53 (--) -2.45 (1.09) 10 594.9 8.9 0.0074 87.6 29.21 (--) -2.48 (1.09) 8 608.5 22.5 <0.0001 77381.0 16.42 (--) -2.42 (1.09) 8 609.9 23.9 <0.0001 157216.4 -2.43 (1.06) 5 681.9 95.9 <0.0001 >106 -2.39 (1.06) 6 684.6 98.6 <0.0001 >106 -2.45 (1.06) 7 685.0 99.0 <0.0001 >106 -2.49 (1.06) 8 686.6 100.7 <0.0001 >106 -2.50 (1.07) 6 687.5 101.5 <0.0001 >106 -2.50 (1.07) 7 689.0 103.0 <0.0001 >106 -2.52 (1.07) 5 699.3 113.3 <0.0001 >106 -2.49 (1.07) 5 699.7 113.7 <0.0001 >106 -2.49 (1.07) 3 703.1 117.1 <0.0001 >106 Averageb 27.35 (0.00) -2.10 (0.96) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 68 Anserinids Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -4.66 (2.34) 2.27 (0.49) 0.66 (1.46) 0.97 (0.33) -4.50 (2.39) 2.28 (0.49) 0.65 (1.46) 0.97 (0.33) -4.37 (2.95) 2.60 (0.74) 1.51 (1.45) 0.65 (0.39) -3.38 (1.03) 2.98 (0.46) -0.10 (0.85) -6.90 (1.78) 3.17 (0.47) -4.19 (2.98) 2.61 (0.74) 1.50 (1.45) 0.64 (0.38) -3.24 (1.13) 2.99 (0.46) -0.10 (0.85) -6.77 (1.84) 3.18 (0.47) 7.05 (--) -34.68 (--) 6.49 (--) -33.69 (--) 0.33 (--) 7.14 (--) -33.94 (--) 6.58 (2.20) -33.09 (0.00) 0.33 (0.86) -2.68 (1.02) 1.09 (0.66) -2.62 (1.07) 1.09 (0.66) -2.06 (0.19) -3.81 (1.14) 1.04 (0.69) -3.75 (1.18) 1.04 (0.69) Averageb -4.63 (2.34) 0.00 (0.00) 2.29 (0.49) 0.00 (0.01) 0.66 (1.44) 0.95 (0.32) 69 Anserinids Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -6.24 (2.34) -6.25 (2.33) -0.34 (0.85) -0.05 (0.83) -2.45 (2.79) -1.68 (1.05) -5.04 (2.14) 2.00 (1.37) -2.47 (2.79) -1.69 (1.05) -0.31 (0.83) -0.05 (0.80) -5.06 (2.14) -0.30 (0.80) -0.03 (0.78) 2.00 (1.37) -0.26 (0.77) -0.02 (0.75) -12.54 (--) -3.74 (--) -12.34 (--) -3.72 (--) -12.55 (--) -3.74 (--) -0.19 (--) -0.01 (--) -12.35 (2.51) -3.72 (1.10) -0.19 (0.64) -0.01 (0.62) -3.74 (1.21) -3.74 (1.21) -0.12 (0.54) -0.01 (0.52) 0.23 (0.51) 0.23 (0.51) -0.12 (0.52) -0.01 (0.51) Averageb -6.11 (2.29) -0.04 (0.05) -0.03 (0.02) 0.00 (0.00) -0.01 (0.04) 0.00 (0.04) 70 Anserinids Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip -0.59 (1.86) 7 237.8 0.0 0.9237 1.0 -0.25 (0.87) 0.52 (0.26) 10 243.9 6.1 0.0430 21.5 -0.48 (1.85) 8 245.9 8.2 0.0156 59.1 -0.26 (1.77) 6 246.1 8.3 0.0146 63.5 -0.22 (1.77) 5 250.4 12.6 0.0017 548.4 -0.26 (0.84) -0.44 (1.85) 11 252.1 14.3 0.0007 1304.2 -0.31 (0.80) -0.23 (1.76) 9 252.1 14.4 0.0007 1308.0 -0.28 (0.77) -0.20 (1.76) 8 256.4 18.7 0.0001 11252.6 -0.50 (--) 6 285.4 47.6 <0.0001 >106 -0.49 (--) 7 287.3 49.5 <0.0001 >106 -0.15 (--) -0.47 (--) 9 291.5 53.8 <0.0001 >106 -0.15 (0.64) -0.47 (1.49) 10 293.5 55.7 <0.0001 >106 -0.24 (1.74) 5 344.7 106.9 <0.0001 >106 -0.13 (0.54) -0.23 (1.74) 8 350.9 113.1 <0.0001 >106 -0.25 (1.75) 3 359.4 121.6 <0.0001 >106 -0.24 (1.74) 5 360.8 123.1 <0.0001 >106 -0.12 (0.52) -0.23 (1.74) 8 367.0 129.2 <0.0001 >106 Averageb -0.01 (0.04) -0.53 (1.79) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 71 Blackbirds Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -2.37 (1.38) -2.17 (0.62) 1.50 (0.64) -4.81 (1.15) 2.21 (0.93) 1.33 (0.61) -1.09 (0.46) -0.12 (0.99) -2.25 (0.61) -3.27 (0.95) 5.79 (2.68) 1.82 (0.67) -2.43 (1.07) -5.99 (1.07) 1.74 (0.60) 0.49 (0.52) 1.16 (0.32) -1.01 (0.40) -3.38 (0.87) 2.47 (0.61) 0.76 (0.69) 1.15 (0.30) -2.65 (1.03) 3.31 (1.93) 1.00 (0.58) -1.49 (0.79) -1.36 (1.19) -1.39 (0.44) 1.25 (0.58) 0.52 (0.83) -1.46 (0.43) -1.49 (0.79) 5.87 (1.87) 1.35 (0.60) -2.46 (0.79) -4.07 (0.90) 1.48 (0.56) 1.28 (0.45) 1.04 (0.32) -0.81 (0.36) 1.50 (0.62) 1.03 (0.30) -1.91 (0.73) 2.06 (0.54) 0.44 (0.13) Averageb -2.45 (1.30) -1.89 (0.53) 0.30 (0.13) 0.00 (0.00) 1.31 (0.57) -0.15 (0.06) 72 Blackbirds Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -2.48 (1.03) 0.99 (0.64) 3.12 (0.66) 2.32 (0.51) -0.88 (0.98) 2.16 (0.48) 2.56 (0.49) 2.13 (0.45) -2.78 (1.01) 1.03 (0.64) 3.07 (0.64) 2.26 (0.50) -1.36 (0.50) 2.50 (0.50) 2.11 (0.47) 2.68 (0.46) 2.52 (0.50) 2.00 (0.43) -1.96 (0.53) 2.37 (0.47) 1.87 (0.41) -1.78 (0.45) 2.32 (0.47) 1.87 (0.42) -1.53 (0.56) 2.29 (0.46) 1.79 (0.40) -0.87 (0.95) 1.66 (0.45) -1.62 (0.85) 1.01 (0.55) -1.91 (0.83) 1.04 (0.55) -1.06 (0.46) 2.27 (0.42) -1.58 (0.47) -1.26 (0.51) -1.46 (0.39) Averageb 0.00 (0.00) -2.29 (1.02) 1.15 (0.62) 0.00 (0.00) 3.04 (0.64) 2.29 (0.50) 73 Blackbirds Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip 0.27 (0.40) -0.44 (0.73) 10 1076.1 0.0 0.7544 1.0 0.33 (0.42) -0.50 (0.73) 11 1079.5 3.5 0.1335 5.6 0.25 (0.39) -0.44 (0.73) 9 1079.9 3.8 0.1110 6.8 0.40 (0.42) 0.39 (0.13) 10 1090.4 14.3 0.0006 1296.1 0.24 (0.39) -0.46 (0.73) 8 1090.8 14.7 0.0005 1560.8 0.22 (0.38) -0.51 (0.74) 9 1110.5 34.4 <0.0001 >106 0.21 (0.37) -0.50 (0.74) 8 1115.2 39.2 <0.0001 >106 0.21 (0.37) -0.49 (0.74) 8 1115.7 39.6 <0.0001 >106 -0.62 (0.72) 8 1125.7 49.6 <0.0001 >106 -0.59 (0.73) 7 1128.9 52.8 <0.0001 >106 -0.58 (0.73) 6 1131.9 55.9 <0.0001 >106 -0.60 (0.72) 7 1132.6 56.5 <0.0001 >106 -0.58 (0.73) 5 1137.0 61.0 <0.0001 >106 -0.62 (0.74) 6 1154.6 78.6 <0.0001 >106 -0.61 (0.74) 5 1158.8 82.7 <0.0001 >106 -0.63 (0.74) 5 1159.0 83.0 <0.0001 >106 -0.60 (0.74) 3 1182.4 106.3 <0.0001 >106 Averageb 0.27 (0.40) -0.45 (0.73) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 74 Brights Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -3.17 (1.10) 3.71 (1.94) 3.06 (0.84) -2.13 (0.82) -2.32 (1.00) 4.12 (1.72) 2.81 (0.80) -2.26 (0.74) -1.94 (1.24) 5.90 (2.13) 2.77 (0.82) -2.95 (0.85) -1.25 (1.15) 5.82 (1.92) 2.64 (0.81) -2.90 (0.78) -3.42 (1.11) 3.38 (0.87) -2.55 (0.96) 3.05 (0.78) -3.17 (1.07) 2.71 (0.72) -2.44 (0.95) 2.55 (0.70) -3.83 (1.70) 0.62 (0.69) 2.79 (0.76) -3.05 (1.61) 0.56 (0.67) 2.62 (0.74) 2.25 (0.73) 0.25 (0.29) -1.26 (0.52) 2.50 (0.67) 0.28 (0.29) -1.10 (0.49) 1.19 (0.19) 0.73 (0.79) 0.33 (0.30) 0.82 (1.10) 0.01 (0.61) 1.02 (0.75) 0.35 (0.30) 1.21 (1.05) 0.02 (0.59) Averageb -2.55 (1.09) 0.00 (0.00) 4.35 (1.89) 0.00 (0.00) 2.90 (0.82) -2.35 (0.79) 75 Brights Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -0.99 (0.55) 1.29 (0.60) 1.00 (0.54) -0.83 (0.55) -1.60 (1.17) -0.55 (0.65) 1.23 (0.62) 0.94 (0.56) -1.44 (1.13) -0.59 (0.64) -1.35 (0.54) 1.35 (0.65) 1.02 (0.54) -1.18 (0.50) 0.14 (0.52) 1.22 (0.60) 0.87 (0.49) 0.10 (0.51) -0.68 (1.00) 0.78 (0.84) 1.29 (0.63) 0.86 (0.50) -0.58 (0.98) 0.67 (0.82) -1.65 (0.67) 1.28 (0.64) 0.89 (0.50) -1.43 (0.62) -0.07 (0.64) 1.00 (0.55) 0.82 (0.49) -1.53 (1.02) 0.55 (0.73) 1.08 (0.59) 0.88 (0.52) 0.03 (0.62) -1.35 (0.99) 0.51 (0.70) Averageb -0.70 (0.42) -0.37 (0.28) -0.14 (0.16) 0.00 (0.00) 0.76 (0.36) 0.59 (0.33) 76 Brights Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip 0.23 (0.49) 0.00 (0.11) 10 1258.4 0.0 0.4657 1.0 -1.17 (0.64) 7 1259.4 1.0 0.2865 1.6 0.16 (0.50) -1.11 (0.64) 11 1261.0 2.6 0.1287 3.6 -1.18 (0.64) 8 1261.2 2.8 0.1163 4.0 0.04 (0.45) -1.02 (0.66) 8 1269.1 10.7 0.0022 210.7 -1.13 (0.65) 5 1272.4 14.0 0.0004 1112.4 0.00 (0.44) -1.04 (0.66) 8 1275.1 16.7 0.0001 4322.8 -1.12 (0.66) 5 1277.4 19.0 <0.0001 13433.0 -0.02 (0.44) -1.04 (0.66) 10 1277.5 19.1 <0.0001 13990.0 -1.13 (0.66) 7 1280.0 21.6 <0.0001 50115.7 0.05 (0.43) -0.94 (0.66) 9 1285.6 27.3 <0.0001 831153.1 -1.03 (0.66) 6 1287.8 29.4 <0.0001 >106 -1.10 (0.66) 3 1290.5 32.1 <0.0001 >106 -0.02 (0.42) -1.05 (0.66) 8 1292.3 33.9 <0.0001 >106 -0.03 (0.43) -1.05 (0.66) 9 1292.6 34.2 <0.0001 >106 -1.12 (0.65) 5 1293.0 34.6 <0.0001 >106 -1.14 (0.65) 6 1293.9 35.6 <0.0001 >106 Averageb 0.13 (0.30) -0.62 (0.39) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 77 Corvids Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -0.11 (1.10) 7.69 (2.11) -0.62 (0.75) -2.57 (0.94) 0.30 (0.97) 7.83 (2.23) -0.70 (0.71) -2.49 (0.94) -2.00 (0.91) 4.46 (1.49) 0.40 (1.13) 7.50 (2.31) -0.74 (0.68) -2.40 (0.95) -0.26 (0.76) 3.76 (1.47) -0.58 (0.54) -1.69 (0.83) 4.68 (1.53) -0.15 (0.67) 4.10 (1.48) -0.40 (0.51) 0.40 (1.41) 8.66 (2.89) -0.30 (0.82) -3.16 (1.31) 2.43 (1.15) -2.27 (0.62) 3.71 (1.87) -2.44 (0.68) -0.70 (0.75) 2.56 (1.10) -2.14 (0.60) 4.22 (2.01) -2.41 (0.72) -0.87 (0.78) -2.04 (1.14) 0.36 (0.68) -0.77 (1.11) 0.89 (0.77) -1.63 (1.04) 0.24 (0.66) -0.21 (0.97) 0.67 (0.70) 0.36 (0.23) Averageb -0.17 (1.05) 0.00 (0.00) 7.26 (2.06) -0.03 (0.03) -0.56 (0.65) -2.22 (0.82) 78 Corvids Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -0.25 (0.58) 1.33 (0.56) -0.04 (0.51) -0.14 (0.57) 0.72 (0.67) 1.25 (0.53) 0.01 (0.49) -0.74 (1.26) 0.22 (0.53) -0.99 (0.66) 1.38 (0.57) 0.01 (0.50) 0.73 (0.65) -0.79 (0.63) -1.86 (1.52) -0.11 (0.65) 16.67 (0.00) 0.00 (0.55) -2.98 (1.03) -0.60 (0.71) 1.51 (0.62) -0.06 (0.48) -3.25 (1.09) -0.67 (0.73) 1.47 (0.62) -0.09 (0.50) -2.78 (1.00) -0.50 (0.68) -3.15 (1.10) -0.60 (0.73) 1.48 (0.53) 1.36 (0.62) -0.11 (0.47) -1.25 (0.52) 1.69 (0.80) -0.08 (0.46) 1.57 (0.55) -1.08 (0.46) Averageb -0.23 (0.50) -0.04 (0.07) 0.01 (0.03) 0.06 (0.05) 1.04 (0.44) -0.03 (0.40) 79 Corvids Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip 0.08 (0.52) -0.75 (0.15) 10 874.0 0.0 0.6768 1.0 0.28 (0.78) 7 877.2 3.2 0.1367 5.0 0.02 (0.50) 0.28 (0.77) 8 878.7 4.7 0.0646 10.5 0.26 (0.78) 8 878.9 4.9 0.0588 11.5 0.05 (0.51) 0.27 (0.77) 9 879.6 5.6 0.0420 16.1 0.32 (0.77) 5 881.7 7.7 0.0145 46.6 0.32 (0.77) 6 883.4 9.4 0.0063 108.2 0.12 (0.54) 0.08 (0.75) 11 890.0 16.0 0.0002 2977.1 0.08 (0.49) 0.14 (0.77) 9 897.3 23.2 <0.0001 109701.9 0.04 (0.50) 0.15 (0.76) 10 898.5 24.4 <0.0001 203056.6 0.20 (0.77) 6 901.9 27.9 <0.0001 >106 0.21 (0.76) 7 902.7 28.6 <0.0001 >106 0.06 (0.48) 0.10 (0.76) 8 916.3 42.2 <0.0001 >106 0.07 (0.47) 0.12 (0.76) 8 918.5 44.5 <0.0001 >106 0.15 (0.76) 5 919.6 45.6 <0.0001 >106 0.19 (0.77) 5 924.2 50.1 <0.0001 >106 0.21 (0.76) 3 926.0 52.0 <0.0001 >106 Averageb 0.06 (0.40) -0.42 (0.35) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 80 Domestic & Exotic Waterfowl Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -6.20 (2.12) 1.93 (0.44) 1.61 (1.26) 0.89 (0.36) -5.20 (2.76) 1.95 (0.60) 1.62 (1.30) 0.84 (0.43) -6.08 (1.57) 2.90 (0.41) -6.26 (2.19) 1.95 (0.45) 1.62 (1.27) 0.89 (0.37) -3.80 (1.04) 2.67 (0.38) 0.19 (0.82) -5.26 (2.83) 1.97 (0.60) 1.63 (1.31) 0.84 (0.43) -6.13 (1.64) 2.92 (0.42) -3.81 (1.13) 2.68 (0.38) 0.19 (0.82) 5.62 (1.46) -22.66 (4236.70) 4.81 (1.97) -22.22 (5087.10) 0.50 (0.83) 5.61 (1.50) -24.80 (0.00) 4.80 (2.00) -22.57 (0.00) 0.50 (0.83) -2.97 (0.99) 1.17 (0.65) -2.98 (1.04) 1.18 (0.65) -1.98 (0.18) -4.00 (1.13) 1.05 (0.69) -4.01 (1.17) 1.05 (0.69) Averageb -5.94 (2.17) 0.00 (0.00) 2.03 (0.47) 0.01 (0.03) 1.46 (1.15) 0.80 (0.34) 81 Domestic & Exotic Waterfowl Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -2.01 (1.55) -2.59 (2.50) -0.48 (1.01) 1.64 (1.23) -2.00 (1.55) 0.34 (0.77) -0.07 (0.76) -1.78 (1.62) -2.57 (2.51) -0.48 (1.01) 0.33 (0.76) -0.06 (0.76) 1.66 (1.23) 0.29 (0.71) -0.03 (0.72) -1.79 (1.63) 0.28 (0.72) -0.03 (0.72) -11.13 (2.25) -2.81 (0.99) -10.90 (2.24) -2.79 (0.99) -11.14 (2.25) -2.80 (0.99) 0.18 (0.59) -0.01 (0.60) -10.91 (2.24) -2.79 (0.99) 0.18 (0.59) -0.01 (0.60) -2.53 (0.97) -2.54 (0.97) 0.15 (0.51) -0.01 (0.52) 0.47 (0.51) 0.47 (0.51) 0.14 (0.50) -0.01 (0.51) Averageb -1.52 (1.18) -0.45 (0.43) -0.08 (0.17) 0.11 (0.08) 0.02 (0.04) 0.00 (0.04) 82 Domestic & Exotic Waterfowl Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip -1.05 (1.80) 7 269.2 0.0 0.6914 1.0 -1.04 (1.80) 8 272.0 2.9 0.1645 4.2 -0.68 (1.74) 5 273.9 4.8 0.0634 10.9 -0.19 (0.80) 0.43 (0.26) 10 275.0 5.9 0.0369 18.7 -0.73 (1.75) 6 275.4 6.3 0.0300 23.1 -0.19 (0.80) -0.91 (1.81) 11 278.0 8.8 0.0085 80.9 -0.25 (0.75) -0.62 (1.73) 8 279.7 10.5 0.0036 190.9 -0.25 (0.75) -0.65 (1.74) 9 281.2 12.1 0.0017 417.0 -0.89 (1.75) 6 309.1 39.9 <0.0001 >106 -0.88 (1.75) 7 310.8 41.7 <0.0001 >106 -0.14 (0.61) -0.85 (1.75) 9 315.1 46.0 <0.0001 >106 -0.14 (0.62) -0.84 (1.75) 10 316.9 47.7 <0.0001 >106 -0.69 (1.71) 5 367.2 98.0 <0.0001 >106 -0.13 (0.53) -0.67 (1.71) 8 373.2 104.0 <0.0001 >106 -0.70 (1.72) 3 376.6 107.4 <0.0001 >106 -0.69 (1.72) 5 377.1 107.9 <0.0001 >106 -0.12 (0.52) -0.67 (1.71) 8 383.1 113.9 <0.0001 >106 Averageb -0.01 (0.04) -0.96 (1.74) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 83 Dabbling Ducks Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV 3.40 (--) -33.07 (--) -0.62 (2.11) 3.31 (0.97) -0.01 (1.00) -0.38 (0.47) -0.80 (2.19) 3.44 (0.98) 0.00 (1.02) -0.39 (0.48) 2.60 (--) -31.46 (--) 0.49 (--) 3.52 (1.41) -16.56 (578.25) 2.67 (1.86) -19.86 (3595.01) 0.53 (0.78) -5.88 (1.08) 2.93 (0.48) 2.13 (0.70) -7.34 (1.80) 3.56 (0.60) -7.89 (1.97) 3.84 (0.70) -6.18 (1.19) 3.11 (0.53) 2.18 (0.72) -3.88 (1.29) 3.72 (0.67) 0.38 (0.86) -0.65 (0.36) -4.10 (1.38) 3.92 (0.71) 0.36 (0.88) -0.68 (0.37) -4.95 (1.07) 0.84 (0.63) -5.02 (1.11) 0.85 (0.63) -2.96 (0.85) 1.18 (0.57) -1.55 (0.15) -3.01 (0.89) 1.19 (0.57) Averageb 1.55 (1.12) -16.70 (210.78) 1.42 (0.41) 0.00 (0.00) 0.08 (0.46) -0.16 (0.20) 84 Dabbling Ducks Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -14.99 (--) 0.51 (--) -12.54 (2.72) 1.69 (0.89) -12.94 (2.80) 1.73 (0.91) 0.98 (0.65) 0.00 (0.69) -14.77 (--) 0.53 (--) -15.60 (2.65) 0.51 (0.84) 0.70 (0.58) -0.03 (0.57) -15.37 (2.61) 0.53 (0.84) 0.71 (0.58) -0.03 (0.57) 2.57 (0.91) 3.43 (1.32) 3.66 (1.39) 0.97 (0.65) -0.07 (0.68) 2.65 (0.93) 0.89 (0.62) -0.05 (0.64) 0.77 (0.55) 0.81 (0.55) 0.87 (0.60) -0.04 (0.64) 1.87 (0.54) 1.89 (0.54) 0.38 (0.43) -0.01 (0.45) -0.85 (0.53) -0.85 (0.53) 0.36 (0.42) -0.01 (0.44) Averageb 0.00 (0.00) -14.06 (1.53) 1.02 (0.50) 0.00 (0.00) 0.25 (0.18) 0.00 (0.19) 85 Dabbling Ducks Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip -0.76 (--) 6 396.8 0.0 0.3042 1.0 -0.90 (1.24) 8 397.1 0.2 0.2705 1.1 -0.47 (0.73) -0.84 (1.24) 11 398.2 1.4 0.1519 2.0 -0.75 (--) 7 398.5 1.7 0.1320 2.3 -0.45 (0.59) -0.76 (1.36) 9 399.1 2.3 0.0985 3.1 -0.46 (0.60) -0.75 (1.36) 10 400.8 3.9 0.0428 7.1 -0.91 (1.26) 6 422.7 25.9 <0.0001 413310.5 -0.90 (1.23) 5 422.7 25.9 <0.0001 416917.6 -0.47 (0.72) -0.84 (1.23) 8 423.6 26.8 <0.0001 653527.3 -0.45 (0.67) -0.85 (1.25) 9 424.1 27.2 <0.0001 809607.7 -0.88 (1.24) 7 430.6 33.8 <0.0001 >106 -0.44 (0.68) 0.33 (0.21) 10 432.0 35.1 <0.0001 >106 -0.73 (1.38) 5 491.0 94.2 <0.0001 >106 -0.31 (0.47) -0.72 (1.38) 8 494.9 98.1 <0.0001 >106 -0.72 (1.38) 5 502.9 106.1 <0.0001 >106 -0.74 (1.38) 3 505.6 108.7 <0.0001 >106 -0.31 (0.46) -0.71 (1.38) 8 506.8 110.0 <0.0001 >106 Averageb -0.14 (0.20) -0.81 (0.72) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 86 Doves Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -0.08 (1.50) -0.02 (0.59) -0.30 (0.66) 0.22 (1.21) 0.34 (0.49) -0.47 (0.68) -0.49 (0.36) -1.48 (0.99) -0.13 (0.63) -1.32 (0.79) 0.27 (0.30) -0.82 (0.99) 0.21 (0.67) -0.31 (1.04) 0.79 (0.49) -0.23 (0.72) -0.66 (0.37) -0.59 (0.63) 0.23 (0.30) -0.06 (0.48) 0.77 (1.45) -0.21 (0.72) 0.06 (0.99) 0.23 (0.54) 0.48 (1.34) 0.20 (0.54) -0.28 (0.60) 1.13 (1.11) 0.34 (0.52) -0.44 (0.63) -0.46 (0.34) -0.50 (0.88) -0.11 (0.59) 0.36 (0.19) -0.26 (0.68) 0.25 (0.28) 0.18 (0.85) 0.19 (0.60) 0.63 (0.92) 0.81 (0.51) -0.20 (0.64) -0.62 (0.34) 0.36 (0.53) 0.23 (0.28) -0.04 (0.43) Averageb -0.24 (1.26) -0.01 (0.23) 0.16 (0.21) 0.00 (0.01) -0.30 (0.62) -0.19 (0.14) 87 Doves Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -2.78 (1.16) 0.86 (0.73) 1.85 (0.54) 2.04 (0.52) -2.71 (1.17) 0.85 (0.52) 1.86 (0.53) 2.03 (0.51) 0.91 (0.48) 1.72 (0.49) 1.99 (0.51) 0.48 (0.63) 1.83 (0.53) 2.01 (0.52) -0.41 (0.48) 1.84 (0.54) 2.07 (0.55) -0.19 (0.49) 1.87 (0.54) 2.04 (0.53) -0.38 (0.60) 1.86 (0.55) 2.06 (0.55) -3.84 (1.50) 0.24 (0.94) 37.67 (0.00) 2.40 (0.69) -2.36 (1.03) 1.12 (0.67) -2.44 (1.05) 1.14 (0.67) -2.52 (1.08) 0.91 (0.48) 0.93 (0.45) 0.45 (0.57) -0.22 (0.44) -0.02 (0.46) -0.17 (0.54) Averageb -0.02 (0.04) -2.08 (0.88) 0.75 (0.54) 0.02 (0.03) 2.25 (0.52) 2.03 (0.52) 88 Doves Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip 0.45 (0.43) -3.85 (0.94) 10 954.2 0.0 0.3852 1.0 0.46 (0.44) -3.90 (0.95) 11 954.3 0.2 0.3565 1.1 0.40 (0.42) -3.84 (0.95) 8 956.4 2.2 0.1276 3.0 0.42 (0.42) -3.79 (0.95) 8 958.5 4.3 0.0452 8.5 0.41 (0.41) -3.82 (0.94) 8 959.1 4.9 0.0334 11.5 0.43 (0.42) -0.26 (0.15) 10 959.6 5.4 0.0254 15.2 0.42 (0.42) -3.81 (0.94) 9 960.6 6.5 0.0152 25.3 0.52 (0.47) -3.80 (0.74) 9 961.2 7.0 0.0115 33.5 -3.89 (0.93) 6 974.1 19.9 <0.0001 21480.9 -3.89 (0.93) 7 976.0 21.8 <0.0001 54909.1 -3.91 (0.93) 8 976.2 22.1 <0.0001 61577.3 -3.86 (0.93) 5 978.2 24.0 <0.0001 164375.3 -3.85 (0.93) 3 978.7 24.5 <0.0001 211761.6 -3.81 (0.93) 5 981.4 27.2 <0.0001 799269.4 -3.85 (0.93) 5 982.5 28.3 <0.0001 >106 -3.86 (0.93) 7 982.6 28.5 <0.0001 >106 -3.83 (0.93) 6 983.9 29.8 <0.0001 >106 Averageb 0.45 (0.43) -3.77 (0.92) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 89 Flycatchers Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -3.27 (1.26) 5.57 (1.89) 1.11 (0.73) -2.46 (0.81) -3.83 (1.31) 5.58 (1.89) 1.06 (0.73) -2.48 (0.82) -1.81 (0.97) 7.18 (1.92) 1.25 (0.73) -3.09 (0.83 -2.33 (1.03) 7.33 (1.95) 1.20 (0.74) -3.16 (0.85) -1.09 (1.41) -1.75 (0.55) 1.05 (0.67) 0.54 (0.98) -1.83 (0.55) -1.65 (1.43) -1.77 (0.55) 1.00 (0.66) -0.09 (1.01) -1.86 (0.55) 1.49 (0.55) 0.89 (0.35) -1.35 (0.44) 1.96 (0.76) 1.06 (0.46) 1.53 (0.80) 1.00 (0.44) 1.00 (0.60) 0.85 (0.34) -1.39 (0.45) -3.17 (1.04) 1.12 (0.64) -3.77 (1.09) 1.09 (0.63) -1.45 (0.85) 1.47 (0.64) -1.95 (0.89) 1.43 (0.63) 0.26 (0.18) Averageb -3.07 (1.21) -0.03 (0.01) 5.83 (1.86) 0.00 (0.00) 1.11 (0.73) -2.56 (0.80) 90 Flycatchers Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer 1.59 (1.19) 0.90 (0.55) 1.59 (1.19) 0.93 (0.55) 0.62 (0.50) 0.65 (0.50) -0.39 (0.60) -0.34 (0.60) 0.57 (0.49) 0.59 (0.49) 0.83 (0.95) 0.02 (0.68) 0.56 (0.94) -0.01 (0.67) 0.83 (0.95) 0.04 (0.68) 0.68 (0.46) 0.63 (0.45) 0.58 (0.95) 0.01 (0.68) 0.70 (0.46) 0.62 (0.45) -1.51 (0.55) -1.81 (0.62) -1.91 (0.64) 0.63 (0.45) 0.55 (0.44) -1.53 (0.56) 0.61 (0.44) 0.52 (0.43) 1.75 (0.51) 1.78 (0.51) 0.68 (0.45) 0.65 (0.44) -0.98 (0.45) -0.96 (0.45) 0.60 (0.43) 0.59 (0.42) Averageb -0.08 (0.13) 1.23 (0.93) 0.70 (0.43) 0.00 (0.00) 0.19 (0.15) 0.19 (0.15) 91 Flycatchers Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip -1.64 (0.81) 8 957.9 0.0 0.5327 1.0 1.10 (0.52) -1.66 (0.81) 11 959.6 1.7 0.2322 2.3 -1.64 (0.82) 7 960.4 2.5 0.1535 3.5 1.07 (0.52) -0.44 (0.13) 10 962.2 4.3 0.0624 8.5 -1.69 (0.82) 7 966.8 8.9 0.0064 83.7 -1.71 (0.82) 6 967.4 9.5 0.0046 115.6 1.09 (0.49) -1.72 (0.82) 10 967.7 9.8 0.0040 132.1 1.10 (0.48) -1.73 (0.82) 9 968.1 10.1 0.0033 159.5 -1.64 (0.83) 6 973.2 15.3 0.0003 2078.9 -1.64 (0.82) 5 973.5 15.6 0.0002 2455.6 1.10 (0.49) -1.66 (0.82) 8 974.2 16.3 0.0002 3485.9 1.02 (0.46) -1.66 (0.83) 9 974.3 16.4 0.0001 3556.6 -1.67 (0.83) 5 975.7 17.8 0.0001 7398.8 1.07 (0.47) -1.68 (0.83) 8 976.4 18.5 0.0001 10249.4 -1.59 (0.84) 5 986.0 28.1 <0.0001 >106 0.97 (0.45) -1.60 (0.84) 8 987.2 29.3 <0.0001 >106 -1.61 (0.84) 3 991.5 33.6 <0.0001 >106 Averageb 0.33 (0.16) -1.57 (0.77) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 92 Kingfishers Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -2.35 (1.03) 9.64 (1.96) -3.89 (1.09) 9.90 (2.18) 0.18 (0.86) -2.37 (1.16) 9.75 (2.00) -6.39 (3.03) 9.05 (2.45) -0.46 (1.39) -0.23 (1.27) -2.64 (1.60) 10.44 (2.44) -0.82 (1.20) -0.35 (1.01) -5.67 (2.60) 8.32 (2.63) -0.90 (1.26) -1.00 (1.14) -2.74 (1.72) 10.66 (2.49) -0.82 (1.21) -0.43 (1.03) -1.76 (0.86) 4.25 (1.26) -0.90 (0.67) -0.18 (1.00) -3.27 (1.08) 0.04 (1.04) -3.29 (1.08) -0.69 (1.36) -3.25 (1.08) 0.38 (0.69) -0.48 (1.41) -3.28 (1.08) 0.38 (0.70) -4.57 (1.06) 0.66 (0.64) -4.41 (1.11) 0.68 (0.65) -2.21 (0.86) 1.17 (0.58) -2.01 (0.90) 1.20 (0.59) -1.21 (0.16) Averageb -2.95 (1.27) 0.00 (0.00) 9.66 (2.08) 0.02 (0.13) -0.12 (0.22) -0.07 (0.19) 93 Kingfishers Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -1.70 (0.98) -1.16 (0.92) -1.77 (1.00) -0.86 (0.87) 0.64 (0.72) 1.17 (1.60) 2.09 (1.09) -1.19 (0.82) 0.84 (1.51) 2.49 (1.04) -1.32 (0.91) 0.44 (0.71) -1.24 (0.83) -0.88 (0.91) 0.68 (0.74) -1.40 (0.86) -1.61 (0.91) 0.27 (0.69) -3.96 (1.20) 0.74 (0.72) -4.05 (1.22) 0.79 (0.73) -0.96 (0.50) -0.21 (0.46) -3.93 (1.21) 0.70 (0.72) -4.01 (1.23) 0.74 (0.73) -0.96 (0.50) -0.21 (0.46) 2.08 (0.55) 2.13 (0.56) -0.93 (0.49) -0.21 (0.43) -2.34 (0.74) -2.38 (0.75) -0.90 (0.48) -0.18 (0.43) Averageb -0.27 (0.21) 0.10 (0.14) 0.20 (0.10) -1.16 (0.67) -0.17 (0.16) 0.11 (0.13) 94 Kingfishers Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip 0.73 (1.04) 5 453.9 0.0 0.5424 1.0 0.70 (1.04) 6 456.5 2.6 0.1484 3.7 0.08 (0.77) 0.76 (1.04) 8 456.7 2.8 0.1331 4.1 0.72 (1.04) 8 458.0 4.1 0.0685 7.9 0.70 (1.04) 7 458.2 4.3 0.0644 8.4 0.06 (0.71) 0.71 (1.05) 11 460.3 6.4 0.0223 24.3 0.13 (0.80) -0.63 (0.17) 10 461.0 7.1 0.0157 34.5 0.12 (0.63) 0.72 (1.08) 9 463.2 9.3 0.0052 104.4 0.54 (1.15) 6 521.5 67.6 <0.0001 >106 0.06 (0.46) 0.53 (1.15) 9 522.6 68.7 <0.0001 >106 0.55 (1.15) 7 523.3 69.4 <0.0001 >106 0.06 (0.46) 0.53 (1.15) 10 524.4 70.6 <0.0001 >106 0.53 (1.15) 5 548.0 94.1 <0.0001 >106 0.00 (0.43) 0.53 (1.15) 8 549.2 95.3 <0.0001 >106 0.55 (1.14) 5 551.0 97.1 <0.0001 >106 0.01 (0.42) 0.53 (1.14) 8 552.4 98.5 <0.0001 >106 0.53 (1.15) 3 565.2 111.3 <0.0001 >106 Averageb 0.01 (0.13) 0.70 (1.03) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 95 Long-tailed Passerines Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -1.87 (0.98) 6.39 (1.63) 1.52 (0.70) -2.79 (0.69) -3.24 (1.20) 6.46 (1.73) 1.03 (0.66) -2.93 (0.75) -0.55 (0.84) 5.90 (1.52) 1.21 (0.64) -2.59 (0.65) -1.83 (1.06) 6.07 (1.64) 0.80 (0.62) -2.74 (0.70) 1.96 (0.70) 1.21 (0.49) -1.41 (0.50) -2.34 (0.96) 2.43 (0.71) -3.61 (1.01) 1.53 (0.61) 2.64 (0.66) 1.15 (0.48) -1.31 (0.48) -2.69 (1.46) -0.81 (0.53) 1.43 (0.62) -0.23 (1.00) -1.06 (0.51) 1.02 (0.78) 1.41 (0.55) -2.37 (0.87) 1.31 (0.57) -1.15 (0.83) 2.09 (0.63) -1.35 (1.36) -0.83 (0.50) 1.21 (0.59) 0.71 (0.96) -1.06 (0.50) 1.74 (0.74) 1.36 (0.54) 1.12 (0.16) Averageb -2.35 (1.05) 0.00 (0.00) 6.42 (1.67) 0.00 (0.00) 1.35 (0.69) -2.84 (0.71) 96 Long-tailed Passerines Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -1.13 (0.52) 1.83 (0.57) 1.49 (0.50) 2.22 (1.27) 0.72 (0.48) 1.85 (0.55) 1.38 (0.47) -0.98 (0.49) 2.06 (1.22) 0.64 (0.45) -2.34 (0.61) 1.63 (0.53) 1.32 (0.48) -1.79 (0.46) 1.68 (0.56) 1.38 (0.50) 1.73 (0.45) 1.77 (0.58) 1.21 (0.45) -2.18 (0.58) 0.67 (1.01) 0.95 (0.66) 1.73 (0.55) 1.20 (0.45) 0.22 (1.00) 0.81 (0.63) 1.62 (0.53) 1.16 (0.45) -0.95 (0.61) 1.48 (0.49) 1.20 (0.45) 1.59 (0.42) -1.61 (0.42) 0.54 (0.98) 0.80 (0.62) 0.14 (0.98) 0.69 (0.61) -0.91 (0.60) Averageb -0.73 (0.34) 0.79 (0.45) 0.26 (0.17) 0.00 (0.00) 1.82 (0.56) 1.44 (0.49) 97 Long-tailed Passerines Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip 0.79 (0.45) 0.33 (0.11) 10 1247.8 0.0 0.6416 1.0 0.81 (0.45) -1.29 (0.67) 11 1249.0 1.2 0.3521 1.8 -1.38 (0.67) 7 1258.0 10.3 0.0038 168.7 -1.38 (0.67) 8 1258.9 11.1 0.0025 260.4 0.57 (0.43) -1.29 (0.67) 9 1270.2 22.5 <0.0001 74994.4 0.55 (0.41) -1.26 (0.67) 8 1277.7 29.9 <0.0001 >106 0.62 (0.42) -1.33 (0.67) 8 1277.7 29.9 <0.0001 >106 -1.37 (0.66) 6 1278.6 30.8 <0.0001 >106 0.59 (0.42) -1.32 (0.67) 10 1278.6 30.8 <0.0001 >106 0.57 (0.42) -1.35 (0.67) 9 1282.2 34.4 <0.0001 >106 0.57 (0.41) -1.32 (0.67) 8 1283.8 36.0 <0.0001 >106 -1.41 (0.66) 5 1286.7 38.9 <0.0001 >106 -1.38 (0.67) 5 1287.0 39.2 <0.0001 >106 -1.40 (0.66) 7 1287.2 39.4 <0.0001 >106 -1.42 (0.66) 6 1289.5 41.7 <0.0001 >106 -1.39 (0.66) 5 1291.0 43.2 <0.0001 >106 -1.36 (0.66) 3 1306.1 58.3 <0.0001 >106 Averageb 0.80 (0.45) -0.25 (0.31) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 98 Raptors Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -4.72 (2.72) 18.57 (19.26) 3.09 (2.04) -7.67 (8.27) -5.54 (2.19) 9.87 (6.58) 2.46 (1.58) -3.79 (2.80) -5.04 (2.27) 8.87 (5.92) 2.42 (1.56) -3.35 (2.51) -4.52 (2.69) 12.81 (14.43) 2.77 (1.74) -5.18 (6.27) -0.27 (0.82) 1.98 (0.95) -1.43 (0.72) 0.04 (0.86) 1.94 (0.74) -1.29 (0.70) -1.22 (1.53) 4.14 (2.58) -0.94 (1.47) 2.04 (1.04) -3.63 (2.22) -1.21 (0.83) 2.68 (1.20) -6.53 (2.24) 3.01 (1.35) -4.96 (1.69) 3.12 (1.37) -2.98 (2.27) -1.22 (0.83) 2.48 (1.18) -5.87 (2.08) 2.76 (1.20) 0.51 (1.45) -1.37 (0.81) -4.37 (1.63) 2.83 (1.18) 1.04 (1.61) -1.41 (0.84) -0.77 (0.51) Averageb -4.99 (2.47) 0.00 (0.00) 14.20 (13.17) -0.01 (0.01) 2.76 (1.79) -5.72 (5.64) 99 Raptors Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -5.65 (5.58) 0.33 (1.36) 0.33 (1.50) 0.35 (1.35) -0.01 (0.91) -1.12 (1.00) -4.05 (4.58) 0.55 (1.24) 0.02 (0.97) -0.97 (1.03) -0.84 (0.89) -0.60 (0.90) -0.17 (0.72) -1.12 (0.78) -0.31 (1.24) 0.05 (1.15) -0.20 (0.73) -1.30 (0.82) -3.32 (1.87) 0.30 (1.02) 1.28 (0.94) -0.93 (0.82) -3.27 (1.86) 0.33 (1.03) -0.11 (0.69) -0.69 (0.72) 1.26 (0.89) 0.05 (0.70) -0.60 (0.73) -3.48 (1.75) 0.29 (1.05) -0.73 (0.76) -0.01 (0.67) -0.59 (0.70) -3.64 (1.84) 0.42 (1.13) -0.15 (0.74) -1.01 (0.76) Averageb 0.14 (0.66) -3.02 (3.01) 0.19 (0.74) 0.00 (0.01) 0.00 (0.09) -0.10 (0.10) 100 Raptors Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip -0.18 (1.46) 8 320.7 0.0 0.5035 1.0 -0.15 (1.47) 7 321.2 0.5 0.3885 1.3 -0.86 (0.97) -2.01 (0.29) 10 325.3 4.6 0.0492 10.2 -0.74 (0.98) -0.21 (1.48) 11 325.6 5.0 0.0422 11.9 -0.18 (1.48) 6 329.4 8.8 0.0062 80.9 -1.53 (0.85) -0.18 (1.52) 9 330.7 10.1 0.0032 155.1 -0.29 (1.41) 5 331.1 10.4 0.0028 181.8 -1.56 (0.87) -0.26 (1.50) 8 332.3 11.7 0.0015 345.1 0.09 (1.58) 7 332.8 12.1 0.0012 429.7 0.13 (1.55) 5 334.2 13.5 0.0006 859.2 0.14 (1.56) 5 335.2 14.6 0.0003 1448.1 -1.26 (0.75) 0.13 (1.61) 10 335.4 14.7 0.0003 1570.2 -1.22 (0.76) 0.17 (1.59) 8 336.6 16.0 0.0002 2947.4 -0.07 (1.53) 6 337.7 17.1 0.0001 5120.2 -1.13 (0.73) 0.18 (1.60) 8 338.2 17.5 0.0001 6410.9 -1.40 (0.80) -0.08 (1.55) 9 339.1 18.4 <0.0001 10116.0 -0.02 (1.54) 3 342.5 21.8 <0.0001 54439.5 Averageb -0.08 (0.09) -0.26 (1.40) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 101 Shorebirds Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -3.00 (1.34) 0.10 (0.44) 0.10 (0.71) 1.24 (0.34) -2.94 (1.03) 0.46 (0.42) 0.34 (0.67) 1.10 (0.33) -0.85 (0.81) 1.53 (0.39) -2.33 (1.25) -0.05 (0.39) 0.13 (0.69) 1.20 (0.33) -2.21 (0.93) 0.29 (0.36) 0.36 (0.64) 1.05 (0.32) -2.92 (0.76) 2.33 (1.38) 0.31 (0.47) 1.49 (1.01) -2.68 (0.66) 2.15 (1.36) -2.71 (0.66) -0.46 (0.63) -0.16 (0.67) 1.28 (0.30) 1.87 (0.94) -2.53 (0.63) 2.51 (1.29) -2.56 (0.64) -0.44 (0.61) -2.01 (0.54) 1.37 (0.43) 0.35 (0.43) -1.80 (0.84) 0.31 (0.54) -2.73 (0.94) -0.01 (0.57) -2.14 (0.86) -0.02 (0.56) -1.26 (0.76) 0.28 (0.52) -1.18 (0.15) Averageb -2.91 (1.27) 0.00 (0.00) 0.20 (0.44) 0.00 (0.00) 0.14 (0.69) 1.18 (0.33) 102 Shorebirds Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -2.40 (1.31) 1.01 (0.55) 1.60 (0.51) 0.72 (0.51) -0.75 (0.57) 1.59 (0.51) 0.72 (0.51) -1.56 (0.65) 1.59 (0.50) 0.68 (0.50) -2.23 (1.24) 1.00 (0.53) -0.72 (0.55) -0.19 (0.67) 1.71 (0.57) 0.70 (0.56) -4.28 (1.24) -1.01 (0.66) 1.40 (0.46) 0.66 (0.46) -4.33 (1.23) -0.99 (0.66) 1.40 (0.46) 0.66 (0.46) -1.50 (0.60) -3.95 (1.16) -0.94 (0.63) -4.01 (1.16) -0.92 (0.64) -0.24 (0.63) -1.06 (0.53) 1.20 (0.42) 0.59 (0.43) 0.93 (0.44) 1.20 (0.42) 0.59 (0.43) 0.89 (0.43) -1.01 (0.51) Averageb -0.14 (0.11) -1.89 (1.04) 0.80 (0.44) -0.04 (0.02) 1.57 (0.50) 0.71 (0.50) 103 Shorebirds Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip 0.08 (0.54) -2.31 (1.17) 11 556.9 0.0 0.7752 1.0 0.07 (0.55) 0.25 (0.21) 10 559.8 2.9 0.1818 4.3 0.07 (0.53) -2.33 (1.15) 8 563.9 7.0 0.0237 32.7 -2.31 (1.16) 8 565.1 8.2 0.0131 59.1 -2.28 (1.16) 7 568.1 11.2 0.0029 270.7 -0.07 (0.65) -2.16 (1.12) 9 569.9 12.9 0.0012 647.9 0.13 (0.49) -2.32 (1.18) 9 569.9 13.0 0.0012 654.4 0.13 (0.49) -2.33 (1.18) 10 571.5 14.6 0.0005 1460.1 -2.30 (1.15) 5 572.5 15.5 0.0003 2379.0 -2.35 (1.17) 6 576.1 19.2 0.0001 14771.8 -2.36 (1.17) 7 577.7 20.8 <0.0001 32105.1 -2.26 (1.14) 6 579.0 22.1 <0.0001 63439.5 0.12 (0.46) -2.37 (1.19) 8 601.7 44.7 <0.0001 >106 0.13 (0.46) -2.39 (1.19) 8 601.7 44.8 <0.0001 >106 -2.40 (1.18) 5 606.3 49.4 <0.0001 >106 -2.39 (1.18) 5 606.3 49.4 <0.0001 >106 -2.40 (1.18) 3 606.9 49.9 <0.0001 >106 Averageb 0.08 (0.54) -1.85 (0.99) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 104 Small Forest Passerines Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -2.70 (0.96) 1.71 (0.74) 2.23 (0.67) -2.62 (0.91) -3.22 (1.15) 1.74 (0.75) 2.06 (0.64) -2.60 (0.93) -3.13 (0.93) 1.81 (0.75) 2.29 (0.68) -2.72 (0.92) -3.59 (1.12) 1.85 (0.77) 2.11 (0.65) -2.73 (0.96) -3.57 (0.88) 2.55 (0.59) -4.07 (0.99) 2.43 (0.59) -4.77 (1.30) 0.06 (0.49) 2.53 (0.59) -3.93 (0.85) 2.54 (0.59) -4.42 (0.96) 2.42 (0.58) -5.10 (1.26) 0.05 (0.48) 2.52 (0.59) 1.01 (0.53) -0.24 (0.23) -0.85 (0.38) 0.46 (0.70) -0.23 (0.23) -0.71 (0.88) -0.23 (0.47) 0.61 (0.46) -0.24 (0.23) -0.83 (0.38) -0.48 (0.16) 0.05 (0.64) -0.23 (0.23) -1.07 (0.82) -0.22 (0.46) Averageb -2.90 (1.00) 0.00 (0.00) 1.74 (0.74) 0.00 (0.00) 2.20 (0.67) -2.63 (0.92) 105 Small Forest Passerines Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -0.49 (0.52) -1.11 (0.48) -0.71 (0.46) 0.65 (1.01) 0.25 (0.44) -1.12 (0.47) -0.71 (0.45) -0.52 (0.51) 0.60 (0.99) 0.24 (0.43) -0.59 (0.45) -1.06 (0.43) -0.70 (0.41) 0.43 (0.40) -1.06 (0.43) -0.70 (0.41) 1.25 (0.89) 0.45 (0.57) -1.06 (0.43) -0.70 (0.41) -0.59 (0.44) 0.41 (0.39) 1.22 (0.87) 0.43 (0.56) -1.00 (0.51) -1.04 (0.42) -0.70 (0.40) -0.40 (0.55) -1.04 (0.42) -0.69 (0.40) 0.68 (0.87) 0.37 (0.56) -1.04 (0.42) -0.71 (0.40) -0.99 (0.50) -0.39 (0.53) 0.64 (0.84) 0.36 (0.55) Averageb -0.38 (0.39) 0.16 (0.25) 0.06 (0.11) 0.00 (0.00) -0.91 (0.39) -0.58 (0.37) 106 Small Forest Passerines Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip 0.12 (0.45) -0.60 (0.18) 10 745.1 0.0 0.6105 1.0 0.13 (0.45) 1.12 (0.95) 11 747.3 2.2 0.2030 3.0 1.23 (0.94) 7 748.0 2.9 0.1432 4.3 1.22 (0.94) 8 750.4 5.3 0.0432 14.1 0.05 (0.41) 1.08 (0.97) 8 763.4 18.4 0.0001 9684.0 0.05 (0.41) 1.09 (0.96) 8 764.1 19.0 <0.0001 13273.0 0.06 (0.41) 1.11 (0.96) 10 766.4 21.3 <0.0001 42306.1 1.23 (0.96) 5 766.9 21.9 <0.0001 55814.0 1.23 (0.96) 5 767.7 22.7 <0.0001 84018.8 1.23 (0.96) 7 770.0 24.9 <0.0001 257175.7 0.05 (0.40) 1.12 (0.96) 9 781.5 36.4 <0.0001 >106 0.03 (0.39) 1.11 (0.96) 8 784.8 39.7 <0.0001 >106 0.05 (0.40) 1.08 (0.96) 9 785.2 40.1 <0.0001 >106 1.21 (0.96) 6 785.3 40.2 <0.0001 >106 1.22 (0.96) 3 786.1 41.0 <0.0001 >106 1.22 (0.97) 5 788.7 43.6 <0.0001 >106 1.20 (0.96) 6 789.2 44.1 <0.0001 >106 Averageb 0.10 (0.37) 0.09 (0.48) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 107 Sparrows Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -1.14 (1.26) 0.11 (0.46) 2.82 (0.94) -0.66 (0.37) -1.22 (1.17) 2.63 (0.86) -1.08 (1.15) 2.19 (0.74) -1.10 (1.32) 0.10 (0.45) 2.32 (0.82) -0.62 (0.35) -1.73 (1.60) 0.28 (0.64) 2.26 (0.76) 3.20 (1.01) -0.46 (0.27) -0.41 (0.54) 2.73 (1.05) -0.45 (0.27) 2.05 (1.31) -0.05 (0.65) -2.67 (1.14) 0.14 (0.43) 3.13 (0.98) -0.68 (0.35) -2.49 (1.02) 2.74 (0.85) -2.34 (1.21) 0.12 (0.43) 2.50 (0.84) -0.63 (0.34) -2.19 (1.00) 2.21 (0.71) -2.90 (1.49) 0.39 (0.60) 2.30 (0.73) 1.91 (0.63) -0.37 (0.23) -0.63 (0.48) 0.79 (0.20) 1.65 (0.80) -0.35 (0.23) 0.66 (0.95) 0.04 (0.54) Averageb -1.15 (1.23) 0.00 (0.00) 0.08 (0.32) 0.00 (0.00) 2.74 (0.91) -0.46 (0.26) 108 Sparrows Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -1.41 (0.61) -0.87 (0.66) -2.66 (0.65) -1.39 (0.61) -0.84 (0.65) -2.61 (0.64) -0.18 (0.52) -0.67 (0.61) -2.45 (0.58) -0.05 (1.16) 0.01 (0.57) -0.69 (0.63) -2.53 (0.60) 0.66 (1.17) 0.01 (0.76) -0.66 (0.61) -2.44 (0.57) -1.39 (0.68) -1.01 (0.82) -2.82 (0.84) -0.43 (0.72) -0.85 (0.75) -2.68 (0.74) 0.58 (1.36) -0.22 (0.77) -0.85 (0.75) -2.64 (0.73) -1.32 (0.55) -1.23 (0.54) -0.35 (1.01) 0.13 (0.55) -0.08 (0.47) 0.31 (1.00) 0.25 (0.72) -1.27 (0.60) -0.62 (0.64) 0.02 (1.03) 0.09 (0.63) Averageb -1.34 (0.58) 0.00 (0.03) 0.00 (0.02) 0.00 (0.00) -0.85 (0.65) -2.64 (0.64) 109 Sparrows Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip -0.66 (0.66) -0.33 (0.13) 10 1117.3 0.0 0.6906 1.0 -0.53 (0.66) -0.53 (0.68) 8 1119.2 1.9 0.2637 2.6 -0.45 (0.63) -0.52 (0.69) 8 1124.2 6.9 0.0218 31.7 -0.57 (0.64) -0.45 (0.70) 11 1124.6 7.2 0.0186 37.2 -0.42 (0.64) -0.49 (0.69) 10 1128.1 10.7 0.0032 214.6 -0.85 (0.83) -0.56 (0.68) 9 1129.7 12.4 0.0014 485.4 -0.72 (0.77) -0.56 (0.68) 8 1131.3 13.9 0.0006 1065.3 -0.56 (0.79) -0.55 (0.68) 9 1136.0 18.7 0.0001 11283.9 -0.18 (0.70) 7 1143.4 26.1 <0.0001 455917.3 -0.27 (0.69) 5 1146.3 28.9 <0.0001 >106 -0.21 (0.70) 8 1151.0 33.7 <0.0001 >106 -0.28 (0.70) 5 1151.5 34.2 <0.0001 >106 -0.25 (0.70) 7 1155.2 37.9 <0.0001 >106 -0.32 (0.70) 6 1159.2 41.9 <0.0001 >106 -0.35 (0.70) 3 1159.6 42.3 <0.0001 >106 -0.34 (0.70) 5 1160.8 43.5 <0.0001 >106 -0.35 (0.70) 6 1165.8 48.5 <0.0001 >106 Averageb -0.62 (0.66) -0.39 (0.30) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ?i ?i) ?AICc = AICci - AICmin ?i = 110 Waders Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -7.59 (2.01) 17.06 (4.66) 2.79 (1.25) 1.28 (0.98) -6.58 (1.70) 7.36 (2.19) 0.28 (0.86) 1.61 (0.81) -4.56 (1.05) 14.28 (4.01) 0.85 (0.61) -3.48 (1.14) 11.61 (2.57) -5.18 (1.61) 12.80 (3.10) 2.35 (1.09) 0.98 (0.80) -2.78 (0.74) 11.81 (2.66) 0.74 (0.55) -4.34 (1.43) 6.23 (1.89) 0.11 (0.78) 1.32 (0.71) -2.00 (1.00) 10.16 (2.64) 0.08 (0.95) -2.94 (0.61) 0.57 (1.37) -2.97 (0.61) -0.34 (0.69) 0.84 (0.86) -2.65 (0.56) 1.32 (1.26) -2.67 (0.56) -0.34 (0.64) -4.26 (1.03) 0.30 (0.60) -3.14 (0.90) 0.22 (0.57) -1.83 (0.83) 0.96 (0.56) -0.99 (0.74) 0.87 (0.53) -0.26 (0.14) Averageb -6.77 (1.76) 0.00 (0.00) 14.68 (4.04) 0.15 (0.11) 1.78 (0.93) 1.08 (0.75) 111 Waders Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -3.76 (1.41) 2.25 (0.88) 2.69 (0.75) 0.55 (1.30) 1.85 (0.70) 2.00 (0.69) 2.59 (0.64) -1.21 (1.15) 1.99 (0.73) 2.58 (0.68) -0.50 (0.75) 2.04 (0.71) 2.60 (0.66) -2.86 (1.08) -1.04 (1.00) 0.21 (1.15) 1.71 (0.64) -0.28 (0.73) -3.02 (0.98) 0.48 (0.66) 1.17 (0.45) 1.99 (0.50) -3.11 (1.00) 0.51 (0.66) 1.18 (0.45) 1.99 (0.50) -2.69 (0.92) 0.47 (0.62) -2.78 (0.93) 0.50 (0.62) 2.48 (0.50) 1.05 (0.42) 1.66 (0.43) 2.31 (0.48) -1.59 (0.48) 0.95 (0.39) 1.53 (0.40) -1.47 (0.46) Averageb -2.53 (1.07) 0.10 (0.23) 0.33 (0.13) -0.01 (0.02) 2.15 (0.81) 2.65 (0.72) 112 Waders Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip -0.07 (0.75) 0.06 (0.13) 10 765.0 0.0 0.6178 1.0 0.46 (0.63) -2.68 (0.86) 11 767.4 2.5 0.1808 3.4 0.22 (0.77) -2.73 (0.86) 9 767.5 2.5 0.1747 3.5 0.44 (0.70) -2.70 (0.86) 8 771.3 6.3 0.0266 23.2 -2.68 (0.85) 7 786.2 21.2 <0.0001 40544.3 -2.74 (0.86) 6 788.2 23.2 <0.0001 108669.1 -2.70 (0.86) 8 788.4 23.4 <0.0001 120221.9 -2.72 (0.86) 5 791.9 26.9 <0.0001 694975.6 0.43 (0.43) -2.39 (0.90) 9 843.4 78.4 <0.0001 >106 0.44 (0.43) -2.40 (0.90) 10 845.3 80.3 <0.0001 >106 -2.58 (0.89) 6 859.7 94.8 <0.0001 >106 -2.59 (0.89) 7 861.6 96.6 <0.0001 >106 0.42 (0.41) -2.47 (0.91) 8 879.4 114.4 <0.0001 >106 -2.63 (0.90) 5 892.7 127.7 <0.0001 >106 0.36 (0.39) -2.40 (0.91) 8 899.4 134.4 <0.0001 >106 -2.55 (0.90) 5 911.9 146.9 <0.0001 >106 -2.55 (0.90) 3 922.1 157.1 <0.0001 >106 Averageb 0.09 (0.73) -0.99 (0.41) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i = 113 Warblers Model Parameter Estimates (S.E.)a ?0 Type Area OW Irreg OW:EV -2.57 (1.15) 0.35 (0.50) 3.17 (0.85) -1.20 (0.50) -3.06 (1.35) 0.35 (0.52) 2.83 (0.78) -1.17 (0.50) -3.21 (1.05) 3.18 (0.75) -3.63 (1.15) 2.95 (0.72) -4.92 (1.58) 0.55 (0.65) 3.13 (0.74) 2.82 (0.82) -0.54 (0.29) -1.10 (0.51) 2.18 (1.24) -0.59 (0.33) 0.21 (1.05) 0.10 (0.60) -3.84 (1.10) 0.47 (0.48) 3.11 (0.88) -1.15 (0.48) -4.08 (1.22) 0.48 (0.48) 2.56 (0.73) -1.09 (0.47) -4.16 (0.99) 3.01 (0.74) -4.46 (1.06) 2.68 (0.67) -5.20 (1.39) 0.23 (0.52) 2.76 (0.68) 1.25 (0.59) -0.38 (0.25) -1.03 (0.45) -0.24 (0.18) 0.27 (0.70) -0.34 (0.24) -0.82 (0.87) -0.08 (0.48) Averageb -2.62 (1.17) 0.00 (0.00) 0.35 (0.50) 0.00 (0.00) 3.14 (0.84) -1.18 (0.49) 114 Warblers Model Parameter Estimates (S.E.)a Isol Slope Veg Land Spring Summer -1.15 (0.60) -1.45 (0.53) -3.52 (0.65) 0.21 (1.14) 0.49 (0.55) -1.54 (0.55) -3.56 (0.67) -1.01 (0.55) -1.30 (0.48) -3.22 (0.59) 0.41 (0.48) -1.37 (0.50) -3.25 (0.60) 1.23 (1.08) 0.83 (0.74) -1.42 (0.52) -3.35 (0.63) -1.66 (0.67) -1.47 (0.55) -3.29 (0.64) -0.39 (0.73) -1.73 (0.80) -3.54 (0.89) 0.52 (1.09) 0.66 (0.70) -1.42 (0.56) -3.21 (0.64) -1.21 (0.57) 0.29 (1.00) 0.35 (0.46) -1.04 (0.51) 0.35 (0.42) 0.92 (0.94) 0.52 (0.62) -1.51 (0.58) -0.33 (0.58) 0.34 (0.92) 0.42 (0.58) Averageb -1.06 (0.55) 0.02 (0.09) 0.04 (0.04) 0.00 (0.00) -1.46 (0.53) -3.52 (0.65) 115 Warblers Model Parameter Estimates (S.E.)a K AICc ?AICc ?i Evidence Ratio Fall Precip -1.25 (0.52) -0.45 (0.17) 10 732.3 0.0 0.9120 1.0 -1.33 (0.55) -1.86 (0.95) 11 737.3 5.0 0.0747 12.2 -1.09 (0.48) -1.84 (0.95) 8 741.3 9.0 0.0099 91.9 -1.15 (0.49) -1.82 (0.94) 8 744.0 11.7 0.0026 350.6 -1.22 (0.52) -1.82 (0.94) 10 746.4 14.2 0.0008 1190.1 -1.23 (0.55) -1.78 (0.93) 9 757.5 25.2 <0.0001 294685.9 -1.52 (0.79) -1.80 (0.92) 8 762.4 30.1 <0.0001 >106 -1.21 (0.56) -1.78 (0.93) 9 766.6 34.3 <0.0001 >106 -1.57 (0.95) 7 770.2 38.0 <0.0001 >106 -1.56 (0.95) 8 776.8 44.6 <0.0001 >106 -1.57 (0.95) 5 778.1 45.8 <0.0001 >106 -1.57 (0.95) 5 782.0 49.7 <0.0001 >106 -1.58 (0.95) 7 785.1 52.8 <0.0001 >106 -1.56 (0.95) 6 795.8 63.5 <0.0001 >106 -1.57 (0.95) 3 800.3 68.0 <0.0001 >106 -1.58 (0.95) 5 802.2 69.9 <0.0001 >106 -1.59 (0.95) 6 804.7 72.4 <0.0001 >106 Averageb -1.26 (0.52) -0.57 (0.24) a (--) denotes an inestimable standard error. b Parameter averages were calculated following Burnham and Anderson (2002), where ?av = ?(?1 ?1? + ? i ?i) ?AICc = AICci - AICmin ?i =