https://cpw.cvlcollections.org/files/original/1e377a0c08baeb247a61484ef6ac984b.pdf 187628da4ffe89a7524fb3ec28cac7e2 PDF Text Text COLORADO PARKS AND WILDLIFE - AVIAN RESEARCH PROGRAM Progress Report October 28, 2016 TITLE: Foraging Ecology of Nonbreeding Ducks and Other Waterbirds in the South Platte River Basin AUTHOR: Adam C. Behney PROJECT PERSONNEL: Brian Sullivan, Jim Gammonley Period Covered: 1 September 2015 – 30 September 2016 All information in this report is preliminary and subject to further evaluation. Information MAY NOT BE PUBLISHED OR QUOTED without permission of the author. Manipulation of these data beyond that contained in this report is discouraged. EXTENDED ABSTRACT Attracting and holding large populations of waterfowl are goals of habitat management for nonbreeding waterfowl. Currently, many habitat planners use bioenergetics approaches to guide habitat planning for nonbreeding waterfowl and shorebirds. In their simplest form, these bioenergetics models predict the amount of habitat needed to support a population goal based on the energy requirements of that goal and the productivity of the habitat. Many of these models assume that energy availability is the only factor affecting duck use of sites. However, recent evidence suggests that although energy availability is important for predicting wetland use by ducks, there are many other factors that influence duck use of sites and utility of those used sites. Therefore, more complex models have been developed but it is unclear how complex these models need to be and what specific factors should be incorporated into to accurately predict carrying capacity or habitat needs. Regardless of what form these models take, given the demonstrated importance of energy availability, estimates of food availability are necessary for the different wetland types in which nonbreeding ducks and shorebirds forage. Most dabbling and many diving ducks primarily consume benthic seeds during winter and migrations but transition to diets higher in invertebrates prior to nesting in spring. Shorebird diet consists of almost exclusively invertebrates. The lower South Platte River corridor in northeastern Colorado (Fig. 1) is considered a waterfowl conservation priority area for migrating and wintering ducks and is important in terms of recreation. However, relatively little research has been conducted to determine the effectiveness of habitat management actions or food availability in the region. My first objective is to estimate duck and shorebird food abundance in various wetland types in the South Platte River basin throughout the nonbreeding season. Depletion of energy in wetlands is most likely a result of consumption by ducks and other wildlife or decomposition. Therefore, wetlands that exhibit the most depletion are also likely the most used during the nonbreeding season. Based on previous food availability research in other regions and the attractiveness of moist-soil units to ducks, we predict that moist soil units will produce the most energy and exhibit the greatest amount of energy depletion throughout the nonbreeding period. Secondly, we will assess the 1 �relationship between duck and shorebird use of wetlands, food abundance, and habitat structure characteristics in an effort to inform management treatments and better understand what form the energetic habitat planning models need to take. Sampling food availability for ducks and shorebirds in wetlands is very time-intensive. Therefore, my last objective is to evaluate a rapid visual assessment procedure to estimate duck habitat quality by comparing the calculated score with actual habitat quality metrics (i.e., food availability and duck use of sites). During September – October (fall) 2015, we sampled 38 sites (Table 1), collecting a total of 264 core samples (avian food samples), and conducted a rapid habitat visual assessment procedure at each. During February – March (winter) and May-June (summer) 2016, we sampled 26 sites, collecting 182 core samples, and conducted the rapid assessment at all sites during both sampling occasions. To date, I’ve processed avian food samples from the fall 2015 and winter 2016 sampling occasions. During fall, moist soil wetlands contained the most energy per unit area (6472 DED/ha), followed by other emergent wetlands (6092.3 DED/ha), sloughs (2869 DED/ha), playas (2138 DED/ha), recharge ponds (913 DED/ha), and reservoirs (565 DED/ha). Depletion of energy from fall to winter was greatest in moist soil wetlands (lost 1129 DED/ha), followed by sloughs (lost 1072 DED/ha), playas (lost 794 DED/ha), emergent (lost 647 DED/ha), and reservoirs (lost 278 DED/ha). We conducted weekly bird counts at 22 sites during spring 2016 and these data are currently being analyzed. Rapid assessments were carried out at every sampling site during fall, winter, and spring. During fall and winter, moist soil units and emergent wetlands had the greatest habitat quality score. The rapid assessment procedures explained about 45% of the variability in energy availability of sites (Fig. 2) and about 68% of the variability in mean duck density on sites during spring (Fig 4.). These estimates of food abundance can be directly incorporated into bioenergetic planning models for northeastern Colorado and based on energy depletion of sites and initial analyses of duck use, it appears that factors other than food availability are working to shape where ducks feed during the nonbreeding season. Furthermore, avian food sampling in wetlands is a very time intensive process and we demonstrate that a simple rapid assessment procedure is positively related to food biomass and duck use. However, depending on the desired accuracy of food availability estimates, the rapid visual assessment procedure we tested may not be sufficient. INTRODUCTION Attracting and holding large populations of waterfowl are goals of habitat management for nonbreeding waterfowl (Soulliere et al. 2007). Migrating and wintering waterfowl provide recreation opportunities for hunters and birdwatchers who stimulate local economies (USFWS 2012) and provide conservation funding through the sale of licenses and donations to conservation organizations (Hvenegaard 2002, Vrtiska et al. 2013). Information that can assist planners to acquire or manage habitat in a way that provides the maximum value to waterfowl and therefore, provide additional recreational opportunity, is extremely valuable. The North American Waterfowl Management Plan (U.S. Department of the Interior and Environment Canada 1986:14) tasked Joint Ventures (regional partnerships of government agencies, non-profit organizations, corporations, tribes, and individuals involved in habitat conservation) with “planning, funding and implementation of projects to preserve or enhance waterfowl habitat”. Currently, many Joint Ventures use bioenergetics approaches to guide habitat planning for nonbreeding waterfowl (Central Valley Joint Venture 2006, Soulliere et al. 2 �2007, Playa Lakes Joint Venture 2008). There are different types of bioenergetics models but in their simplest form (e.g., daily ration models) they predict the amount of habitat (energy) needed to support a population goal based on the energy requirements of that goal and the productivity of the habitat. Input parameters may include energy demands of the birds, energy supply (food production) of the habitat, and the population goal (Williams et al. 2014). These simple daily ration models assume that energy availability is the only factor affecting duck use of sites. However, recent evidence suggests that although energy availability is important for predicting wetland use by ducks, it explains a relatively low percentage of the variability in duck use (<7%, Brasher 2010; <15%, O’Shaughnessy 2014) and factors like vegetation structure can influence the ability of ducks to feed efficiently (Behney 2014). Therefore, more complex spatial depletion models (Sutherland and Allport 1994) and agent-based models (Miller et al. 2014) have been developed to incorporate spatial and/or temporal components into predicting animal use of certain areas. It is unclear how complex these models need to be and what specific factors should be incorporated into the spatial or temporal components of the models to accurately predict carrying capacity or habitat needs. In addition to waterfowl, nonbreeding shorebirds rely on wetlands for foraging and joint ventures have begun using bioenergetics models (Playa Lakes Joint Venture 2008) for shorebird habitat planning. Many shorebirds migrate long distances and require substantial food resources along the migration corridor to build enough energy reserves for successful migration and subsequent reproduction (Fellows et al. 2001, Brown et al. 2001). Shorebirds have a relatively high basal metabolic rate (Kersten and Piersma 1987) and combined with their long distance migrations necessitates foraging habitat with abundant food during migration to build and maintain energy reserves. Regardless of how complex the waterfowl and shorebird bioenergetics models become, estimates of food (energy) availability are necessary for the different wetland types in which nonbreeding ducks and shorebirds forage. Most dabbling and many diving ducks primarily consume benthic seeds during winter and migrations but transition to diets higher in invertebrates prior to nesting in spring (Hitchcock 2009, Tidwell et al. 2013) and shorebirds primarily consume invertebrates (Davis and Smith 1998). A variety of methods have been developed for estimating food biomass in wetlands including core sampling (Kross et al. 2008, Stafford et al. 2011, Hagy and Kaminski 2012a), visual assessment (Naylor et al. 2005), vacuum sampling (Penny et al. 2006), and clipping vegetation (Haukos and Smith 1993, Gray et al. 2009). All these methods have certain drawbacks (Williams et al. 2014), but core sampling is commonly used and can produce precise and unbiased estimates if enough samples are collected, although it is a very time intensive method (Behney et al. 2014). Due to this time commitment, rapid visual assessment procedures have been developed to assess habitat quality more quickly (Naylor et al. 2005, Ortega 2013). Naylor et al. (2005) found that the estimates from their rapid assessment procedure explained 88% of the variation in estimates of seed production obtained through core sampling. Stafford et al. (2011) found the same procedure accounted for 65% of the variation in seed biomass (estimated through core sampling). Rapid assessment procedures for a variety of focal species in the lower South Platte River basin were developed (Ortega 2013) and could benefit from a comparison with estimates of food production and avian use. The lower South Platte River corridor in northeastern Colorado is considered a waterfowl conservation priority area for migrating and wintering ducks (Colorado Parks and Wildlife 2011). About 60% of Colorado duck hunters hunt along the Lower South Platte River and about half of the ducks harvested in Colorado are harvested in this area (Runge and Gammonley 2012). 3 �There are 26 State Wildlife Areas along the lower South Platte River offering opportunities for public hunting. Because of these factors, the lower South Platte River corridor represents an important area in terms of recreation opportunity. Recently, research has been conducted on hunting regulations and hunter satisfaction in the area (South Platte River Blue Ribbon Panel 2007), however, relatively little research has been conducted to determine the effectiveness of habitat management actions in the region. Additionally, the South Platte River basin supports a diverse and abundant suite of migrating shorebirds. Cariveau and Risk (2007) found at least 24 shorebird species using the region during spring, including six species of conservation concern: Wilson’s phalarope (Phalaropus tricolor), piping plover (Charadrius melodus), Baird’s sandpiper (Calidris bairdii), least sandpiper (Calidris minutilla), long-billed dowitcher (Limnodromus scolopaceus), and stilt sandpiper (Calidris himantopus). These species forage in the same types of wetlands as duck in the South Platte Basin although they use areas with shallower water (or no standing water; Cariveau and Risk 2007). Despite the importance of the basin to nonbreeding ducks and shorebirds, we am unaware of any published estimates of duck or shorebird food availability for this area and relying on estimates from other regions may bias predictions of energetic models for the lower South Platte River corridor. Colorado Parks and Wildlife’s (CPW’s) Wetland Wildlife Conservation Program was initiated in 1997 and provides funding for wetland creation, restoration, enhancement, and easements, conducts wildlife and aquatic inventories, participates in education and outreach, and conducts wetland project monitoring and evaluation. As of 2011, the program has protected, enhanced, or created more than 80,937 ha of wetlands and associated uplands and over 1,126 km of riparian habitat much of which in the South Platte River basin (Colorado Parks and Wildlife 2011). Colorado Parks and Wildlife has invested about $25 million on over 750 individual wetland projects. Other conservation programs in Colorado’s South Platte River basin include the U.S. Fish and Wildlife Service’s Partners for Fish and Wildlife program, which has worked to restore over 13,118 ha of wetlands and 516 km of riparian/stream habitat in Colorado since 1988. Ducks Unlimited has conserved about 38,849 ha of wetlands, spending $26,167,184, in Colorado as of 2012. The U.S. Department of Agriculture’s Wetland Reserve Program has about 258 ha of wetland easements in the South Platte River basin. Many of these conservation programs are specifically targeted to provide waterfowl and other waterbird habitat, however formal evaluations of the effectiveness of these programs and their habitat management strategies, in terms of bird response, are lacking. The first goal listed in the Strategic Plan for the Wetland Wildlife Conservation Program (hereafter ‘the plan’, Colorado Parks and Wildlife 2011) is to improve the distribution and abundance of ducks, and opportunities for public waterfowl hunting. To achieve these goals, the plan outlines biological planning strategies including “identify limiting factors and appropriate management treatments” (BP4) and “develop and apply models to help understand the relationship of species populations to limiting factors” (BP5). The plan also identifies research strategies evaluating the assumptions of the models (R1) and habitat treatments (R2). We plan to address these strategies in the lower South Platte River basin by measuring habitat characteristics and duck use throughout migration and winter. Specifically, the first objective of this study is to estimate duck and shorebird food abundance in the South Platte River basin by sampling during three occasions throughout the nonbreeding season in various wetland types thought to provide food for ducks and shorebirds. Emphasis will be placed on sampling food available to dabbling ducks because 6 out of 7 duck species identified as priority species in the plan (Colorado Parks and Wildlife 2011) are dabbling 4 �ducks and much of the energetic information used in the duck bioenergetics models are based on dabbling ducks (Prince 1979). However, we will also estimate food availability for shorebirds that use similar feeding behaviors in the same types of wetlands by limiting data to invertebrates from shallow samples (<15 cm; Fellows et al. 2001). Sampling during three occasions (fall, winter, summer) will allow me to estimate food depletion during each period. Depletion of energy in wetlands is most likely a result of consumption by ducks and other wildlife or decomposition (Hagy and Kaminski 2012b). Therefore, wetlands that exhibit the most depletion are also likely the most used during the nonbreeding season. Based on the attractiveness of moist-soil units to ducks because of their structure and high food production, we predict that moist soil units will produce the most energy and exhibit the greatest amount of energy depletion throughout the nonbreeding period. My second objective is to assess the relationship between duck and shorebird use of wetlands, food abundance, and habitat structure characteristics. This will allow me to identify limiting factors in terms of duck and shorebird habitat and inform management treatments, better understand what form the energetic models need to take (do ducks distribute themselves solely based on food abundance?). We will also evaluate assumptions regarding food availability in the models and habitat treatments in relation to bird use. Lastly, we will perform the rapid visual assessment procedure for ducks (Ortega 2013) at each wetland and compare the calculated index of habitat quality with food biomass and duck use of sites. This research represents the first phase of wetland habitat research associated with CPW’s Wetland Wildlife Conservation Program. We plan to build on these findings in the future by assessing the probability that wetlands are actually inundated during migration and winter (wetland availability), evaluating more specific habitat management actions in terms of food abundance and bird use, as well as incorporating marked birds to assess habitat selection and the ultimate destinations of birds migrating through the South Platte River basin. OBJECTIVES 1. Estimate duck and shorebird food abundance in the South Platte River basin by sampling during three occasions throughout the nonbreeding season in various wetland types thought to provide food for ducks and shorebirds. 2. Assess how food abundance and/or habitat structure characteristics are related to duck and shorebird use of wetlands. This will facilitate identification of limiting factors in terms of duck and shorebird habitat, inform management treatments, and better understand what form energetic planning models need to take. 3. Test a rapid visual assessment procedure for ducks (Ortega 2013) at each wetland and compare the calculated index of habitat quality with food abundance and duck use of sites. METHODS Study area and site selection This study is being conducted in the South Platte River basin in northeastern Colorado. Different wetland types are distributed differently across the basin; therefore, we used two study areas: a larger area encompassing the whole basin (for playas) and a smaller area directly around 5 �the South Platte River (for other wetland types; Fig. 1). Based on the classification and wetland abundance described in Lemly et al. (2015) and information needs of collaborators, we used 6 wetland type classifications: 1) moist-soil units: actively managed specifically for duck food production, 2) other emergent wetlands, 3) sloughs, 4) playas, 5) recharge ponds (not including moist-soil units), and 6) reservoirs. We used a multi-stage sampling strategy to select sites (Stafford et al. 2006, described below) and took 7 core samples per site. For playas, moist-soil units, other emergent wetlands, and recharge ponds less than 2 ha, we randomly distributed 7 core samples throughout the wetland/unit. For wetlands larger than 2 ha, we randomly selected a half or quarter (whichever will be closer to 2 ha) and distributed 7 core sample points within the 2 ha section. For sloughs and reservoirs, we randomly selected a 300 m stretch of the slough or reservoir shoreline and randomly distributed 7 core sample points along that 300 m stretch. Field methods Starting in late September/early October, we sampled all study wetlands to estimate the amount of food available for nonbreeding ducks and shorebirds at the beginning of fall migration. We chose these sampling dates because they are at the beginning of migration in northeastern Colorado and it is prior to duck hunting season, when access to wetlands was restricted. Cores were 5 cm diameter (Behney et al. 2014) and 5 cm deep (Evans-Peters 2010). If the random core sample location was dry and greater than 5 m from water (i.e., not a mudflat associated with the wetland) or if the water was deeper than 50 cm, the conditions were noted, but no core sample was collected because these are not conditions associated with duck foraging, and a new random core location selected. We chose 50 cm as a cutoff because Behney (2014) found that the majority of mallard (Anas platyrhynchos) feeding occurred at water depths less than 50 cm. Cores were washed through a 500 µm sieve bucket in the field to wash away soil and the remaining material was placed in a bag with ≥ 70% ethanol and transported to the Fort Collins lab for processing. At each core location, we recorded water depth and estimated vegetation density using a modified robel pole technique for which we noted the number of 10 cm bands on a 1.5 m tall PVC (2.54 cm diameter) pole, that were completely visible from 5 m in the 4 cardinal directions. Lastly, we conducted a rapid-assessment procedure created by the CNHP (Ortega 2013) at each site. As soon as wetlands began to thaw (February/March), we took 7 core samples from a smaller subsample of study wetlands to estimate how much food depletion occurred during fall migration and how much food was available for spring migrants. Again, in early summer, when we observed a dramatic decrease in regional duck and shorebird abundance (after most of spring migration, May/June), we collected an additional 7 core samples from the same subsample of study wetlands to assess depletion during spring migration. After most water bodies thaw and ducks started arriving in the area (February/March), we began conducting bird counts on the smaller subsample of wetlands. We visited each wetland once per week to estimate wetland use by all duck and shorebird species throughout the spring. At each wetland, we first conducted a vantage count, noting the number of each species present. Directly following this vantage count, we approached the wetland, flushed birds and estimated the number of each species present. Identification of flushing shorebirds can be challenging, therefore, we grouped shorebirds into the lowest taxonomic level possible (e.g., small sandpipers: genus Calidris, yellowlegs: genus Tringa). We also conducted aerial surveys when 6 �possible, during which, we counted the number of each species of duck and if possible shorebirds. Lab methods In the lab, cores were washed through a series of sieves (2000 µm [#10], 35 µm [#45]) to separate different size particles. We visually searched through the material and picked out any seeds, tubers, or invertebrates and sorted seeds to genus and invertebrates to phylum, class, or order, if possible. We dried all material at 60°C to a constant mass (about 48 hours) and weighed to the nearest 0.00001 g. If processing time per core for certain sites regularly exceeded 3 – 4 hours, we subsampled other cores from those sites by spreading the material retained by the fine sieve out on a tray with 100 equal sized cells drawn onto it and randomly selecting 25 cells to sort through (25%). We then multiplied the biomass of avian food retained by the fine sieve by 4. Subsampling is a commonly used practice to reduce processing time (Williams et al. 2014). It is unlikely that subsampling substantially increases the variance in biomass estimates because most of the biomass in a core sample comes from large seeds, which are not subsampled because they are retained by the course sieve (Reinecke and Hartke 2005, Kross et al. 2008, Hagy et al. 2011). Statistical analyses To estimate duck and shorebird food biomass (kg/ha) in study wetlands, we summed the biomass of all seeds, tubers, and invertebrates. We excluded cocklebur (Xanthium spp.) because they have not been reported in the diet of ducks (Hagy and Kaminski 2012b) and are very large, potentially introducing substantial bias into estimates of food biomass. We will use linear mixed effects models with site as a random effect and wetland type as a fixed effect (package lme4: Bates et al. 2015) in R (R Core Team 2015) to estimate the mean food biomass for each wetland type based on the multiple samples per site, sampling strategy. To convert estimates of food biomass to duck energy days (DED), we used true metabolizable energy (TME) values of 2.87 kcal/g for most food items (Soulliere et al. 2007) and 0.57 kcal/g for Mollusca (mean of Gastropoda and Bivalvia in Cramer 2009). We separated Mollusca (Gastropoda and Bivalvia) from other invertebrates because they were abundant in our samples, they are heavy, but contain little energy per mass compared to other soft bodied invertebrates (Jorde and Owen 1988). We used 356.8 kcal/day as the daily energy requirement of a mallard (Miller and Eadie 2006, Soulliere et al. 2007). To assess factors that influence duck use of sites during spring, we will model duck and shorebird counts conducted during late-winter/spring. The specific family of models we will use to predict duck or shorebird use will depend on the nature of the variability we find in the data. Although the response variable is a count, assuming a normal distribution can sometimes be appropriate (McDonald and White 2010). Therefore, we will attempt to use linear mixed models (normal error distribution) to predict duck or shorebird use. If the assumptions of linear models are severely violated, we will use generalized linear mixed models (GLMMs) with poisson or negative binomial error distribution (log link). Predictor variables will include wetland area to account for differences in size, wetland type, food biomass, water depth, vegetation density, and percentage open water. Wetland will be included as a random effect. The food biomass variable will be linearly interpolated between sampling occasions. Estimating variability in food biomass based on interpolated estimates is problematic. There are multiple ways to deal with this (e.g., including SE as a covariate, using same SE as pre and post sampling occasions), and the best 7 �way will depend on the data, but we will attempt to account for it in some way. We will compare models using ΔAICc, model weights, and R2 as described in Nakagawa and Schielzeth (2013). If we find evidence of overdispersion (residual deviance substantially greater than residual degrees of freedom), we will use QAICc to compare models. If overdispersion is substantial due to many zeros in the dataset, we will use zero-inflated poisson or negative binomial models. To determine the amount of variation in duck counts explained by the rapid assessment procedures we will again, use duck counts as the response variable in an analysis similar to above but with the rapid assessment score as the predictor variable and report R2. Similarly, we will use simple linear regression to model food biomass as a function of rapid assessment scores and report the amount of variation explained (R2). Food biomass (kg/ha) and duck density was log transformed in the analysis to reduce heteroskedasticity. RESULTS AND DISCUSSION During September – October (fall) 2015, we sampled 38 sites (Table 1), collecting a total of 264 core samples (avian food samples), and conducted the rapid assessment procedure at each. During February – March (winter) and May-June (summer) 2016, we sampled 26 sites, collecting 182 core samples, and conducted the rapid assessment at all sites during both sampling occasions. Food biomass During fall sampling (Sep-Oct 2015), emergent and moist soil wetlands contained the most energy per unit area for ducks and were very similar to each other (Table 2). Sloughs contained a relatively high biomass of total food, however, this was strongly influenced by molluscs which contain little energy. Similarly, during winter sampling, emergent wetlands contained the greatest amount of energy, followed by moist soil, and sloughs (Table 3). Our estimates of food abundance are on the high end of what has been reported in the literature. In moist soil wetlands in the Mississippi Alluvial Valley, reported seed biomasses range from 396 – 750 kg/ha (Kross et al. 2008, Hagy and Kaminsky 2012a, Olmstead et al. 2013). In the Upper Mississippi River and Great Lakes Region, Straub et al. (2012) found mean food biomass ranged from 52 – 208 kg/ha depending on wetland type. My preliminary estimates may appear high because we only sampled at shallow depths within each water body. Our estimates only apply to portions of the water bodies with ≤ 50 cm water depth. For example, on large reservoirs, our estimates only apply to the shallow ring around the edge (a small percentage of the reservoir), not the entire reservoir. Based on preliminary analyses of the distance from the shore of reservoirs to where water depth reaches 50 cm, 5.9% (range 3.3 – 14.7%) of reservoirs are less than 50 cm deep (shallow enough for a dabbling duck to feed). Overall, Polygonum, Amaranthus, and Echinochloa were the most frequently detected seed genera and (Table 4) and also made up the greatest overall percent biomass during fall sampling (Table 5). Diptera, Gastropoda, and Annelida were the most frequently observed invertebrate taxa (Table 4) and made up the greatest invertebrate percent biomass during fall (Table 5). Based on the subsample of water bodies that were sampled during both fall and winter sampling, depletion of energy was greatest in moist soil wetlands (lost 1129 DED/ha), followed by sloughs (lost 1072 DED/ha), playas (lost 794 DED/ha), emergent (lost 647 DED/ha), and 8 �reservoirs (lost 278 DED/ha). Estimates of DED/ha in recharge ponds actually increased from fall to winter sampling and this is most likely due to sampling error (Hagy and Kaminski 2012b). Moist soil wetlands are known to be preferred foraging habitats for ducks due to food availability, water depth, and plant structure and composition, which, consistent with our predictions and observations of moist soil wetlands supporting the greatest duck density (see below), likely explains why there was the most depletion in these wetlands. Bird counts Due to some wetlands drying up during the spring, we were only able to conduct weekly bird counts on 22 sites. Averaging duck counts across weeks within sites, moist soil units had the greatest mean duck density (25.1 ducks/ha), followed by playas (20.2 ducks/ha), other emergent wetlands (17.0 ducks/ha), sloughs (4.1 ducks/ha), recharge ponds (0.4 ducks/ha), and reservoirs (0.1 ducks/ha). Mallard, northern shoveler, redhead, gadwall, and ring-necked duck were the most common ducks observed. American avocet, Wilson’s phalarope, killdeer, least sandpiper, and Baird’s sandpiper were the most common shorebirds observed and identified. Rapid assessments In Sep/Oct 2015, moist soil wetlands had the greatest mean rapid assessment score (77.0), followed by emergent wetlands (75.0), playas (67.2), sloughs (57.2), recharge ponds (33.7), and reservoirs (22.0). The model using total rapid assessment score to predict energy availability per unit area (DED/ha) was more parsimonious than the null model (∆AICc = 20.7, wi = 0.99) and the rapid assessment explained 45.5% of the variability in energy availability per unit area (Fig. 2). On average for a one unit increase in rapid assessment score, energy availability was 3.6% greater. Similarly, in Feb/Mar 2016, moist soil wetlands had the greatest mean rapid assessment score (77.9), followed by emergent wetlands (70.0), sloughs (64.8), playas (61.8), recharge ponds (31.6), and reservoirs (25.5). The model using total rapid assessment score to predict energy availability per unit area (DED/ha) was more parsimonious than the null model (∆AICc = 13.4, wi = 0.99) and the rapid assessment explained 47.3% of the variability in energy availability per unit area (Fig. 3). On average for a one unit increase in rapid assessment score, energy availability was 4.2% greater. A model predicting duck density (ducks/ha) averaged over the spring from total rapid assessment score was more parsimonious than the null model (∆AICc = 22.3, wi = 0.99) and the rapid assessment explained 67.9% of the variability in duck density (Fig. 4). On average, for a one unit increase in rapid assessment score, duck density increased by 8.4%. Estimating avian food availability through core sampling or conducting repeated bird counts to assess habitat quality is very time intensive. The simple rapid assessment procedure we tested was positively related to food biomass and duck use. However, depending on the desired accuracy of food availability estimates, the rapid visual assessment procedure we tested may not be sufficient. CURRENT PROGRESS We collected core samples and conducted rapid assessments during summer (May-Jun) and fall (Sep-Oct) 2016 sampling occasions and am currently processing these samples in the 9 �lab. This study is expected to continue for one more year encompassing two more sampling periods: Feb-Mar 2017 and May-Jun 2017, as well as bird counts during spring 2017. LITERATURE CITED Bates, D., M. Maechler, B. Bolker, and S. Walker. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67:1-48. Behney, A. C. 2014. Foraging behavior of ducks during spring migration in the Wabash River Region, Illinois. Dissertation, Southern Illinois University, Carbondale, Illinois, USA. Behney, A. C., R. O’Shaughnessy, M. W. Eichholz, J. D. Stafford. 2014. Influence of item distribution pattern and abundance on efficiency of benthic core sampling. Wetlands 34:1109-1121. Brasher, M. G. 2010. Duck use and energetic carrying capacity of actively and passively managed wetlands in Ohio during autumn and spring migration. Dissertation, The Ohio State University, Columbus, Ohio, USA. Brown, S., C. Hickey, B. Harrington, and R. Gill, eds. 2001. The U.S. Shorebird Conservation Plan, 2nd ed. Manomet Center for Conservation Sciences, Manomet, Massachusetts, USA. Cariveau, A. B. and B. Risk. 2007. Colorado wetlands monitoring and evaluation project: migratory bird habitat in the South Platte River corridor. Rocky Mountain Bird Observatory, Brighton, Colorado, USA. Central Valley Joint Venture. 2006. Central Valley Joint Venture Implementation Plan – Conserving Bird Habitat. U.S. Fish and Wildlife Service, Sacramento, California, USA. Colorado Parks and Wildlife. 2011. Statewide strategies for wetland and riparian conservation: strategic plan for the Wetland Wildlife Conservation Program. Cramer, D. M. 2009. Estimating habitat carrying capacity for American black ducks wintering in southern New Jersey. Thesis, University of Delaware, Newark, Delaware, USA. Davis, C. A. and L. M. Smith. 1998. Ecology and management of migrant shorebirds in the playa lakes region of Texas. Wildlife Monographs 140. Evans-Peters, G. R. 2010. Assessing biological values of Wetland Reserve Program wetlands for wintering waterfowl. Thesis, Oregon State University, Corvallis, Oregon, USA. Fellows, S., K. Stone, S. Jones, N. Damude, S. Brown. 2001. Central plains/playa lakes regional shorebird conservation plan, version 1., U.S. Shorebird Conservation Plan. Gray, M. J., M. A. Foster, L. A. P. Peniche. 2009. New technology for estimating seed production of moist-soil plants. Journal of Wildlife Management 73:1229-1232. Hagy, H. M., J. N. Straub, and R. M. Kaminski. 2011. Estimation and correction of seed recovery bias from moist-soil cores. Journal of Wildlife Management 75:959-966. Hagy, H. M. and R. M. Kaminski. 2012a. Winter waterbird and food dynamics in autumnmanaged moist-soil wetlands in the Mississippi Alluvial Valley. Wildlife Society Bulletin 36:512-523. Hagy, H. M. and R. M. Kaminski. 2012b. Apparent seed use by ducks in moist-soil wetlands of the Mississippi Alluvial Valley. Journal of Wildlife Management 76:1053-1061. Haukos, D. A. and L. M. Smith. 1993. Moist-soil management of playa lakes for migrating and wintering ducks. Wildlife Society Bulletin 21:288-298. Hitchcock, A. N. 2009. Diets of spring-migrating waterfowl in the Upper Mississippi River and Great Lakes Region. Thesis, Southern Illinois University, Carbondale, Illinois, USA. 10 �Hvenegaard, G. T. 2002. Birder specialization differences in conservation involvement, demographics, and motivations. Human Dimensions of Wildlife 7:21-36. Jorde, D. G. and R. B. Owen, Jr. 1988. Efficiency of nutrient use by American black ducks wintering in Maine. Journal of Wildlife Management 52:209-214. Kersten, M. and T. Piersma. 1987. High levels of energy expenditure in shorebirds; metabolic adaptations to an energetically expensive way of life. Ardea 75:175-187. Kross, J., R. M. Kaminski, K. J. Reinecke, E. J. Penny, and A. T. Pearse. 2008. Moist soil seed abundance in managed wetlands in the Mississippi Alluvial Valley. Journal of Wildlife Management 72:707-714. Lemly, J., L. Gilligan, and G. Smith. 2015. Wetlands of the lower South Platte River basin: extent, condition and habitat quality. Colorado Natural Heritage Program, Colorado State University, Fort Collins, Colorado, USA. McDonald, T. L. and G. C. White. 2010. A comparison of regression models for small counts. Journal of Wildlife Management 74:514-521. Miller, M. L., K. M. Ringelman, J. C. Schank, and J. M. Eadie. 2014. SWAMP: An agent-based model for wetland and waterfowl conservation and management. Simulation 90:52-68. Miller, M. R. and J. M. Eadie. 2006. The allometic relationship between resting metabolic rate and body mass in wild waterfowl (Anatidae) and an application to estimation of winter habitat requirements. The Condor 108:166-177. Nakagawa, S. and H. Schielzeth. 2013. A general and simple method for obtaining R2 from generalized linear mixed-effects models. Methods in Ecology and Evolution 4:133-142. Naylor, L. W., J. M. Eadie, W. D. Smith, M. Eichholz, and M. J. Gray. 2005. A simple method to predict seed yield in moist soil habitats. Wildlife Society Bulletin 33:1335-1341. O’Shaughnessy, R. 2014. Factors determining habitat selection by spring migrating waterfowl along the Wabash River, Illinois. Dissertation, Southern Illinois University, Carbondale, Illinois, USA. Olmstead, V. G., E. B. Webb, and R. W. Johnson. 2013. Moist-soil seed biomass and species richness on wetland reserve program easements in the Mississippi Alluvial Valley. Wetlands 33:197-206. Ortega, C. P. 2013. Habitat quality for wetland-dependent priority wildlife species in the lower South Platte River basin, Colorado: species assessments and monitoring protocols. Report, Submitted to: Colorado Natural Heritage Program, Colorado Parks and Wildlife, Environmental Protection Agency. Penny, E. J., R. M. Kaminski, and K. J. Reinecke. 2006. A new device to estimate abundance of moist-soil plant seeds. Wildlife Society Bulletin 34:186-190. Playa Lakes Joint Venture. 2008. Area Implementation Plan for the Shortgrass Prairie Bird Conservation Region (18) of Colorado. Prince, H. H. 1979. Bioenergetics of postbreeding dabbling ducks. Pages 103-117 in Waterfowl and wetlands – an integrated review. (T. A. Bookhour, editor). La Crosse, Printing Co. Inc. La Crosse, Wisconsin, USA. R Core Team. 2015. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Reinecke, K. J. and K. M. Hartke. 2005. Estimating moist-soil seeds available to waterfowl with double sampling for stratification. Journal of Wildlife Management 69:794-799. Runge, J. P. and J. H. Gammonley. 2012. Evaluating relationships between hunting regulations, habitat conditions, and duck hunting quality on state wildlife areas in northeastern 11 �Colorado. Colorado Division of Parks and Wildlife – Avian Research Program, Progress Report. Soulliere, G. J., B. A. Potter, J. M. Coluccy, R. C. Gatti., C. L. Roy, D. R. Luukkonen, P. W. Brown, and M. W. Eichholz. 2007. Upper Mississippi River and Great Lakes Region Joint Venture Waterfowl Habitat Conservation Strategy. U.S. Fish and Wildlife Service, Fort Snelling, Minnesota, USA. South Platte River Blue Ribbon Panel. 2007. South Platte waterfowl hunting blue ribbon panel summary report. Fort Collins, Colorado, USA. Stafford, J. D., K. J. Reinecke, R. M. Kaminski, and P. D. Gerard. 2006. Multi-stage sampling for large scale natural resources surveys: A case study of rice and waterfowl. Journal of Environmental Management 78:353-361. Stafford, J. D., A. P. Yetter, C. S. Hine, R. V. Smith, and M. M. Horath. 2011. Seed abundance for waterfowl in wetlands managed by the Illinois Department of Natural Resources. Journal of Fish and Wildlife Management 2:3-11. Straub, J. N., R. J. Gates, R. D. Schultheis, T. Yerkes, J. M. Coluccy, and J. D. Stafford. 2012. Wetland food resources for sping-migrating ducks in the Upper Mississippi River and Great Lakes Region. Journal of Wildlife Management 76:768-777. Sutherland, W. J. and G. A. Allport. 1994. A spatial depletion model of the interaction between bean geese and wigeon with the consequences for habitat management. Journal of Animal Ecology 63:51-59. Tidwell, P. R., E. B. Webb, M. P. Vrtiska, and A. A. Bishop. 2013. Diets and food selection of female mallards and blue-winged teal during spring migration. Journal of Fish and Wildlife Management 4:63-74. U.S. Department of the Interior and Environment Canada. 1986. North American Waterfowl Management Plan, Washington, D.C., USA. U.S. Fish and Wildlife Service [USFWS]. 2012. 2011 National survey of fishing, hunting, and wildlife-associated recreation. U.S. Department of the Interior, Washington, D.C., USA. Vrtiska, M. P., J. H. Gammonley, L. W. Naylor, and A. H. Raedeke. 2013. Economic and conservation ramifications from the decline of waterfowl hunters. Wildlife Society Bulletin 37:380-388. Williams, C. K., B. D. Dugger, M. G. Brasher, J. M. Coluccy, D. M. Cramer, J. M. Eadie, M. J. Gray, H. M. Hagy, M. Livolsi, S. R. McWilliams, M. Petrie, G. J. Soulliere, J. M. Tirpak, and E. B. Webb. 2014. Estimating habitat carrying capacity for migrating and wintering waterfowl: considerations, pitfalls and improvements. Wildfowl Special Issue 4:407-435. 12 �Table 1. Number of sites sampled during each sampling period. Site type Emergent1 Moist soil Playa Reservoir Recharge pond Slough 1 Sept-Oct 2015 6 8 5 6 5 8 Feb-Mar 2016 5 4 2 5 3 7 May-Jun 2016 5 4 2 5 3 7 At one emergent site in Sept-Oct 2015, only 5 core samples were collected. Table 2. Food availability (kg/ha) ± standard error and duck energy days per hectare (DED/ha) for each site type sampled during fall 2015. Site type Emergent Moist soil Playa Recharge Reservoir Slough Seed biomass 659.2 ± 159.1 719.8 ± 137.0 242.9 ± 173.3 85.3 ± 173.3 64.0 ± 158.2 205.9 ± 137.0 Invert biomass 20.5 ± 11.9 33.6 ± 10.2 22.5 ± 12.9 8.9 ± 12.9 2.7 ± 11.8 43.1 ± 10.2 Mollusca biomass 391.3 ± 198.7 257.7 ± 168.9 2.3 ± 213.6 97.2 ± 213.6 18.2 ± 195.0 542.5 ± 168.9 Total biomass DED/ha 1070.4 ± 273.8 6092.3 1011.1 ± 234.3 6472.0 267.7 ± 296.4 2138.5 191.5 ± 296.4 913.4 84.8 ± 270.5 565.3 791.5 ± 234.3 2869.5 Table 3. Food availability (kg/ha) ± standard error and duck energy days per hectare (DED/ha) for each site type sampled during winter 2016. Site type Emergent Moist soil Playa Recharge Reservoir Slough Seed biomass 592.0 ± 143.7 261.2 ± 160.6 140.1 ± 227.2 201.7 ± 185.5 32.1 ± 143.7 141.7 ± 131.2 Invert biomass 27.7 ± 17.4 35.3 ± 19.4 68.4 ± 27.5 11.2 ± 22.4 4.2 ± 17.4 82.0 ± 15.9 Mollusca biomass 644.2 ± 212.7 430.1 ± 237.9 1.1 ± 336.4 58.4 ± 274.6 49.0 ± 212.7 477.3 ± 194.2 Total biomass DED/ha 1264.0 ± 251.0 6014.5 726.6 ± 280.6 3071.9 209.6 ± 396.9 1678.9 271.3 ± 324.0 1805.7 85.3 ± 251.0 370.3 700.9 ± 229.1 2561.5 Table 4. Percentage of fall samples containing most common seed and invertebrate food taxa. 13 �Taxa Moist Overall Emergent soil Playa Reservoir Recharge Slough Seeds Polygonum Amaranthus Echinochloa Chenopodium Rumex Pascopyrum Carex Cyperus Helianthus Panicum Poa Verbena Zannichellia Inverts Diptera Gastropoda Annelida Coleoptera Odonata Arachnida Bivalvia Entognatha Ephemeroptera Hemiptera Trichoptera 61.7 49.6 43.6 37.1 35.2 65.0 62.5 55.0 65.0 73.2 51.8 57.1 62.5 55.4 77.1 57.1 57.1 20.0 20.0 40.5 54.8 33.3 45.7 54.3 64.3 26.8 64.3 22.9 23.8 26.8 20.0 28.6 20.0 20.0 20.0 74.3 52.5 71.2 47.3 35.6 17.0 12.9 65.0 97.5 35.0 28.6 91.1 73.2 21.4 33.9 35.0 20.0 91.4 20.0 22.9 22.9 50.0 16.7 26.2 4.8 4.8 23.2 51.4 22.9 31.4 20.0 11.4 11.4 71.4 53.6 69.6 8.9 12.5 8.9 14.3 8.9 8.9 14 �Table 5. Mean percent mass of each fall sample that most common seed and invertebrate food taxa comprise. Moist Taxa Overall Emergent soil Playa Reservoir Recharge Slough Seeds Polygonum 19.4 21.2 20.5 17.8 20.3 18.1 18.3 Echinochloa 15.8 8.8 16.9 5.0 16.9 33.4 Amaranthus 7.9 7.7 17.8 14.7 Rumex 5.6 6.7 12.3 3.9 4.1 Schoenoplectus 5.3 12.0 11.4 Agrostis 4.3 Chenopodium 5.5 4.4 6.7 5.0 Cyperus 4.7 Eleocharis 8.5 Lemna 7.4 Panicum 5.1 Verbena 36.3 Inverts Gastropoda 44.5 85.6 86.8 16.6 19.4 37.2 Diptera 28.1 4.2 27.8 56.7 30.0 28.3 26.0 Annelida 10.6 3.1 6.2 5.4 16.0 12.1 18.4 Coleoptera 4.1 6.7 10.9 6.0 1.5 Odonata 2.2 3.3 7.4 4.3 Arachnida 2.0 3.7 Bivalvia 6.6 Ephemeroptera 2.3 Hemiptera 3.3 Hymenoptera 2.4 15 �Figure 1. Study site along the South Platte River corridor in northeastern Colorado. 16 �Figure 2. The relationship between rapid assessment score and energy availability (DED/ha) during fall (Sep-Oct) 2015 sampling. The solid line represents predicted values from a linear model using rapid assessment score to predict energy availability. Energy availability was natural log transformed in analysis to reduce heteroskedasticity but values were backtransformed in the figure. Dotted lines represent ± one standard error. Figure 3. The relationship between rapid assessment score and energy availability (DED/ha) during winter (Feb-Mar) 2016 sampling. The solid line represents predicted values from a linear model using rapid assessment score to predict energy availability. Energy availability was natural log transformed in analysis to reduce heteroskedasticity but values were backtransformed in the figure. Dotted lines represent ± one standard error. Figure 4. The relationship between rapid assessment score (winter 2016) and mean duck density (ducks/ha) during spring (Mar-Jun 2016). The solid line represents predicted values from a linear model using rapid assessment score to predict mean duck density. Duck density was natural log transformed in analysis to reduce heteroskedasticity but values were backtransformed in the figure. Dotted lines represent ± one standard error. 17 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Reports Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Foraging ecology of nonbreeding ducks and other waterbirds in the South Platte River Basin Description An account of the resource Attracting and holding large populations of waterfowl are goals of habitat management for nonbreeding waterfowl. Currently, many habitat planners use bioenergetics approaches to guide habitat planning for nonbreeding waterfowl and shorebirds. In their simplest form, these bioenergetics models predict the amount of habitat needed to support a population goal based on the energy requirements of that goal and the productivity of the habitat. Many of these models assume that energy availability is the only factor affecting duck use of sites. However, recent evidence suggests that although energy availability is important for predicting wetland use by ducks, there are many other factors that influence duck use of sites and utility of those used sites. Therefore, more complex models have been developed but it is unclear how complex these models need to be and what specific factors should be incorporated into to accurately predict carrying capacity or habitat needs. Regardless of what form these models take, given the demonstrated importance of energy availability, estimates of food availability are necessary for the different wetland types in which nonbreeding ducks and shorebirds forage. Most dabbling and many diving ducks primarily consume benthic seeds during winter and migrations but transition to diets higher in invertebrates prior to nesting in spring. Shorebird diet consists of almost exclusively invertebrates. The lower South Platte River Creator An entity primarily responsible for making the resource Behney, Adam C. Subject The topic of the resource Waterfowl Wildlife management South Platte River Basin Extent The size or duration of the resource. 17 pages Date Created Date of creation of the resource. 2016-10-28 Rights Information about rights held in and over the resource <a href="http://rightsstatements.org/vocab/NoC-NC/1.0/">No Copyright - Non-Commercial Use Only</a> Type The nature or genre of the resource Text Format The file format, physical medium, or dimensions of the resource application/pdf Language A language of the resource English