Availability of Supplemental Corn for Sandhill Cranes, Light Geese, and Dabbling Ducks Wintering in New Mexico

Natural resource organizations (e.g., U.S. Fish and Wildlife Service, Migratory Bird Habitat Joint Ventures, and State Wildlife Agencies) in North America are regularly tasked with establishing habitat objectives for wintering migratory birds that are implemented at regional scales (U.S. Fish and Wildlife Service 2012; Williams et al. 2014). Focused on waterfowl, these initiatives concentrate on linking energetic requirements of a species and the energy base provided from food resources within winter habitats (Fino et al. 2017; Guillemain et al. 2017). Consideration of food resources in the context of energy supply and demand has great practical significance for managing and conserving populations of migratory gamebirds. Determining whether energy availability in food supplies is in equilibrium with population demands can guide habitat management practices, which can be modified to meet population objectives (Davis et al. 2014; Petrie et al. 2016).

Limitation of winter food is an important driver of population dynamics in migratory birds (Sherry et al. 2005; Danner et al. 2013). Access to high-quality food resources during winter can not only influence physiological condition and survival during this period (Duriez et al. 2012; Cooper et al. 2015; Rakhimberdiev et al. 2015) but also dictate timing and successful completion of spring migration (Cooper et al. 2015; Paxton and Moore 2015). Furthermore, energetic gains during winter can be a deciding factor in reproductive performance during the breeding season (Newton 2004; Inger et al. 2010; Sedinger and Alisauskas 2014). Essentially, the potential variation in habitat quality and quantity (i.e., available forage) can influence population dynamics of migratory birds, such as sandhill cranes, ducks, and geese (Sedinger and Alisauskas 2014). Therefore, potential for such cross-seasonal effects requires careful planning and consideration of the availability and quality of food for wintering migratory birds.

In the Middle Rio Grande Valley (MRGV) of central New Mexico, a multiagency program supplements food to wintering migratory gamebirds by growing corn Zea mays on public lands that is left standing until manipulated (i.e., mowed) during critical wintering months (December–February). With reductions in food supplies in the MRGV because of changes in agricultural practices (e.g., increased alfalfa production and less corn), drought (e.g., wetland extent loss), and loss of habitat from anthropogenic activities (Pacific Flyway Council and Central Flyway Council 2016; Boggie et al. 2018b), public lands provide crucial habitat for wintering migratory gamebirds. The principal motivation for the supplemental corn program was to provide energetic and nutritional support to the Rocky Mountain Population (RMP) of greater sandhill cranes Antigone canadensis tabida that overwinter in the MRGV and reduce crop depredation issues on adjacent private lands (Taylor 1999; U.S. Department of Agriculture 2009; Sanchez and Cline 2020). Approximately 80% of the RMP winter in the MRGV, making it the single most important wintering area for this population (Pacific Flyway Council and Central Flyway Council 2016).

Additionally, the MRGV supports thousands of wintering light geese (lesser snow geese Anser caerulescens and Ross's geese A. rossii) and a variety of dabbling duck species (i.e., Anas spp.), including mallard Anas platyrhynchos and northern pintail A. acuta. Additionally, an influx of lesser sandhill cranes A. c. canadensis from the Midcontinent Population (Krapu et al. 2014; Pacific Flyway Council and Central Flyway Council 2016) has placed greater pressure on food resources, in particular supplemental corn. A general planning structure devised and currently implemented in the MRGV to manage migratory gamebirds is in place (Taylor 1999; Sanchez and Cline 2020). Management actions detailed in the plan have proven successful; however, incorporation of species-specific energetic demands and energy available from corn produced on public lands is lacking.

An evaluation of whether current crop management (e.g., production, depletion, and energetic content) for supplemental corn provides the necessary population energy requirements for each guild (sandhill cranes, light geese, and dabbling ducks) is necessary. We drew from literature to estimate daily energy expenditures and relied on survey data collected seasonally by state and federal agencies to establish winter population estimates for each guild. Our objective was to evaluate whether current crop management for corn produces a sufficient energy base for each of these guilds that overwinter in the MRGV through total corn production, daily depletion rates, and energetic content of grown corn.

Our study area was the MRGV of central New Mexico (Figure 1). The MRGV is an area of ∼726 km², as defined by the Rio Grande and associated floodplain, and spans north of Albuquerque, New Mexico, and approximately 200 km downstream south of Bosque del Apache National Wildlife Refuge (Bosque del Apache NWR). The valley floor has a mean elevation of 1,470 m. Mean precipitation during winter 2014–2017 was 19 mm (range = 0.8–49.8 mm), with a mean temperature of 8°C (range = 1.6–17.3°C). Outward from the riparian zone along the Rio Grande, the area is characterized by Rio Grande cottonwood Populus deltoides wislizeni galleries along with dominant native and nonnative vegetation assemblages, including willows (Gooding willow Salix gooddingii and coyote willow S. exigua), salt cedar Tamarisk chinensis, New Mexico privet Forestiera neomexicana, and Russian olive Elaeagnus angustifolia. The remaining floodplain is a matrix of suburban and urban areas with irrigated agricultural lands consisting of primarily alfalfa Medicago sativa, pastures and grass hay, and some small grain silage for livestock (Howe et al. 1991; Swanson et al. 2011). In addition, New Mexico Department of Game and Fish state (Ladd S. Gordon Complex) and federal (Bosque del Apache NWR) properties support sandhill cranes, light geese, and multiple species of dabbling ducks through a matrix of the aforementioned floodplain habitats along with moist-soil-managed wetlands and corn fields (Boggie et al. 2018a).

Multiple modeling techniques have been developed and implemented for estimating daily energy expenditures of waterfowl. The foundation of these techniques is estimation of resting metabolic rate from respirometry studies for a particular species or generalized linear models that explain variation in resting metabolic rate as a function of body mass across multiple species (Williams et al. 2014; Ringleman et al. 2018). Bioenergetic models, which derive energy expenditures of an individual, account for resting metabolic rate and metabolic cost of daily activities experienced under free-living conditions. However, complexity increases by incorporating other components, such as thermoregulatory costs (McKinney and McWilliams 2005; Ladin et al. 2011; Livolsi et al. 2015; Aagaard et al. 2018) or costs of growth and lipid production (Pearse et al. 2010). More complex bioenergetic models can offer greater insight into the energetic envelope of a species. Transferability of the models to different taxa and data limitations (e.g., zone of thermal neutrality or lower and upper critical temperatures for a particular species) led to use of more simplistic models, albeit with more assumptions and caveats, to determine energetic expenditures for a species and associated energetic carrying capacity of their habitat (Williams et al. 2014; Petrie et al. 2016). For our purposes of approximating a general baseline for energetic carrying capacity of the MRGV, we elected to use more simplistic models to estimate the energetic expenditures of the three guilds of wintering species to maintain consistency between the three guilds due to sandhill cranes not having an established true metabolic energy value. We believe these are suitable given the objectives of this study and maintain consistency among the three guilds of birds.

Our central objective was estimating energy requirements for the populations of each guild through winter. Thus, we scaled individual daily energy expenditures to estimated population abundance for each guild. We used surveys flown by New Mexico Department of Game and Fish in fixed-winged aircraft during November, December, and January combined with biweekly ground surveys conducted by U.S. Fish and Wildlife Service on Bosque del Apache NWR to determine mean winter population abundance for sandhill cranes, light geese, and dabbling ducks during the study period (New Mexico Department of Game and Fish and U.S. Fish and Wildlife, unpublished data). We used mean peak abundance indices between the two surveys (monthly aerial and biweekly ground surveys) to represent population abundance, because these values more accurately depicted the number of individuals placing demands on resources through the winter based on our personal observations in the field. We assumed that the proportion of the RMP of greater sandhill cranes wintering in the MRGV was ∼80% (Pacific Flyway Council and Central Flyway Council 2016) of the 3-y average from fall population surveys on staging areas that spanned our study period (Thorpe et al. 2016). We subtracted this number from the total number of sandhill cranes counted in surveys conducted in the MRGV to provide an estimate of lesser sandhill cranes that belong to the Midcontinent Population and winter in the MRGV.

Each winter, land managers at Bosque del Apache NWR and Ladd S. Gordon Complex estimate production (i.e., yield) for all corn fields. We used these estimates to represent winter corn production on state and federal properties from 2014 to 2017. We used procedures to sample corn yield within fields following standardized methods (Lauer 2002; Nielsen 2004; Thomison 2015). In general, depending on field size and row spacing, we randomly sampled two to six 4.5- to 5.3-m² sections of standing corn (corresponds to 1/1,000 acre) within a field. Within these sections, we counted the total number of ears and sampled a subset of ears. We then counted the number of kernels on sampled ears and determined the moisture content of kernels through a moisture analyzer or oven drying and weighing to constant mass. We estimated total dry mass of corn within sample sections and extrapolated to estimate field-specific corn production (dry weight [kg/ha]). We derived the total corn production (dry weight [kg/ha]) at each property for the total area planted per site (Bernardo Wildlife Area, mean = 50 ha; Bosque del Apache NWR, mean = 100 ha). We used analysis of variance (ANOVA) to determine if corn production varied by year.

In waterfowl bioenergetic studies, true metabolizable energy (amount of energy in a food resource after accounting for metabolic fecal and endogenous urinary losses) is often used as the unit of energy of food resources available to consumers (Miller and Reinecke 1984; Brasher et al. 2007; Williams et al. 2014; Guillemain et al. 2017). To our knowledge, there are no taxa-specific estimates of true metabolizable energy of food resources for cranes, and estimates for the closest species in body mass are geese (Storey and Allen 1982; Petrie et al. 1998). Therefore, rather than assume that these estimates are germane to sandhill cranes and corn varieties grown in the MRGV, we estimated and used gross energy of different varieties of corn grown on state and federal properties to represent energetic availability of corn resources. State and federal agencies have different guidelines governing what types of corn they grow. Specifically, Bosque del Apache NWR is restricted to grow only nongenetically modified crops (NON-GMC) of corn, whereas New Mexico Department of Game and Fish is not. Moreover, throughout the study period, Bosque del Apache NWR grew strains of heirloom corn historically farmed in the MRGV, with the objective to determine if these strains are more resilient to lower soil nutrients and drought than other NON-GMC varieties. During the first year of the study, we sampled all varieties grown on state and federal properties, which included all varieties available to grow throughout the study period. We classified each variety as genetically modified crop (GMC), NON-GMC, or heirloom. We sent samples to Washington State University Wildlife Habitat and Nutrition Laboratory to analyze each variety for gross energy by bomb calorimetry. For each variety, we applied mean caloric content (kcal/g dry mass) to fields growing that variety to estimate total energy availability (kcal) from corn production on state and federal properties. We used ANOVA to determine if energy content differed among corn varieties.

In many studies investigating availability and depletion of corn as it relates to consumption by migratory birds, corn availability is residual waste grain from postharvested fields usually owned by private producers (Pearse et al. 2010; Anteau et al. 2011; Sherfy et al. 2011). In our study area, standing corn is available to birds gradually by mechanical manipulation (i.e., tractors pulling commercial rotary mowers) throughout the season. Of note, bumping (knocked down at about 0.75 m off the ground; Post et al. 1998) of corn was historically a management option that was not used during this study period. We were particularly interested in estimating depletion rates of mowed corn; that is, once an area of standing corn was mowed, how long does residual corn from that mowing event persist before it is completely consumed? We could apply such estimates toward mowing schedules used by land managers, and these estimates could provide useful information in determining availability of corn biomass given the area of standing corn mowed while considering energetic demands of migratory gamebird populations that rely on the resource throughout winter.

We coordinated with land managers on state and federal properties to sample mowed corn following their normal mowing schedules beginning in December through February to estimate daily depletion of postmowed corn. We stratified our sampling across December, January, and February. Depending on how many fields still had standing corn, we sampled two to four fields on both state and federal properties within each month. Once land managers identified fields for each sampling occasion, we followed tractors to a field where they mowed 12 rows of corn using two passes within that field. Immediately following mowing and before any granivory, we sampled corn within the mowed area. We used 1-m² quadrats to sample corn and sampled 25 quadrats/ha (10 quadrats/acre). We estimated number of hectares per row within each field and then area mowed by multiplying ha/row by number of rows mowed and allocated sampling quadrats proportionally to the mowed area. We used a vehicle-mounted global positioning system antenna (Garmin GPS18x USB) connected to a laptop running a geographic information system (ArcGIS 10.2, ESRI, Redlands, CA) equipped with multispectral high-resolution imagery (1-m² pixels) to randomly generate sampling locations for quadrats within the mowed area in real time. We removed all corn (kernels and any intact ears) within each quadrat. For the subsequent 2 d following initial sampling, we repeated the protocol within the mowed area to determine daily change in corn availability. Locations of quadrats within the mowed area were never spatially coincident with quadrats sampled on previous days. Following field collection, we dried and weighed (i.e., corn kernels removed from ears) each sample to a constant mass to the nearest gram. We summed dry weights of corn from all quadrats sampled on the same day within the same field. We then estimated proportional change in amount of corn between subsequent days within each sampling occasion for each field.

To relate daily proportional change in amount of corn to total amount of corn available within the mowed area, we used corn production estimates for each field and field area to derive density of corn (kg/ha) for each field and estimate kg/mowed area on the first day of sampling. We used this as our starting density of corn within the mowed area and to estimate total corn production within the study area. For each subsequent day, we applied the corresponding proportional change between sample days to estimate amount of corn depleted daily. Our estimate of daily depletion rates relied on several assumptions. First, values of standing corn production remain constant throughout the season, and there is little depletion of corn when standing. We felt that this is a fair assumption for our sampling circumstances, as the structural characteristics of an average fully matured corn field (e.g., height and density of corn stalks within a field) in our study area discourages feeding of standing corn by birds. While there is a presence of elk Cervus canadensis and mule deer Odocoileus hemionus, they typically consume corn before it is made available to the target birds (i.e., cranes, ducks, and geese). Second, changes in amount of corn found in quadrats sampled between days were proportional to changes of depletion of total amount of corn within the mowed area, which we believe is a reasonable assumption given that corn is dispersed in areas that were mowed. We used the lme4 package to fit linear mixed-effects models to explore what factors might explain variation in daily depletion rates (Bates et al. 2015). We square root transformed the response variable to meet assumptions of normality and homogeneity of variance. We considered fields nested within season as random effects, and covariates included day of season, month, and minimum temperature. We performed all statistical analyses in Program R, and we used an α of 0.07 as the threshold to determine statistical significance to capture what we felt were biologically relevant parameter estimates (R Core Team 2017).

Fall survey estimates during this study estimated 22,087 RMP sandhill cranes (3-y average; Thorpe et al. 2016). Assuming ∼80% of the RMP winters in the MRGV, 17,670 sandhill cranes from the RMP relied on forage resources on state and federal properties. Winter surveys of sandhill cranes in the MRGV conducted by the New Mexico Department of Game and Fish and U.S. Fish and Wildlife Service from 2014 to 2017 estimated a mean population abundance of 20,608 sandhill cranes. After accounting for the 17,670 sandhill cranes from the RMP, we estimated that 2,938 lesser sandhill cranes from the Midcontinent Population wintered in the MRGV. We estimated the mean duration of stay for the RMP in the MRGV from sandhill cranes tagged with satellite transmitters as ∼88 d (mean arrival of November 23 ± 2 d, n = 22; mean departure of February 18 ± 1 d, n = 35; Boggie et al. 2018a). Assuming a winter season of 88 d, we estimated the seasonal energetic requirement for the RMP and Midcontinent Population from monthly daily energetic expenditures as 1.14 × 10⁹ kcal (95% CI = 1.07 × 10⁹ to 1.21 × 10⁹ kcal).

Winter surveys of light geese and dabbling duck populations in the MRGV conducted by the New Mexico Department of Game and Fish and U.S. Fish and Wildlife Service from 2014 to 2017 estimated a mean population abundance of 20,608 light geese and 50,935 dabbling ducks. We assumed the same winter period of 88 d for light geese and dabbling ducks for consistency. The estimated mean population-level energy requirement for light geese wintering in the MRGV was 1.61 × 10⁹ kcal (95% CI = 1.39 × 10⁹ to 1.84 × 10⁹ kcal), and the estimated mean population-level energy requirement for dabbling ducks wintering in the MRGV was 1.48 × 10⁹ kcal (95% CI = 1.35 × 10⁹ to 1.62 × 10⁹ kcal).

Total corn production on state and federal properties ranged from 967,278 to 1,165,408 kg through the winters of 2014 to 2017, with corn acreage planted ranging from 47 to 132 ha between state and federal properties. Mean corn production between state and federal properties over the three winters was 1,052,006 ± 58,965 kg. Estimated caloric energy differed among the three varieties of corn (GMC, heirloom, and NON-GMC) grown on state and federal properties (F_2,31 = 25.7 and P < 0.001). Estimated caloric content for heirloom varieties (4.74 ± 0.03 kcal/g, n = 15) was slightly greater than NON-GMC (4.52 ± 0.02 kcal/g, n = 5) and GMC varieties (4.57 ± 0.01 kcal/g, n = 15). Although caloric content of corn varieties differed, mean energy availability from corn produced on state and federal properties did not differ by season (F_2,78 = 1.69 and P = 0.19). Using the mean energy available across all corn varieties (4.6 kcal/g ± 0.019), total energy available from corn production across years was 4.82 × 10⁹ kcal ± 2.72 × 10⁸.

We sampled corn fields on state and federal properties in December (n = 13), January (n = 12), and February (n = 8) throughout the study period to estimate monthly rates of daily depletion of mowed corn. There was considerable variation in depletion rates within month (Figure 2). We fit several linear mixed-effects models to explain variability in estimates of daily depletion rates. Results suggested that minimum temperature had little effect on depletion rates, and monthly differences in depletion rates were marginal, but differences were detected between December and February (Table 1). Day of season appeared to be the most important effect in explaining depletion rates, which increased as the season progressed (Table 1; Figure 3).

The seasonal corn requirements using the mean gross energy content of all varieties was 245,434 kg (95% CI = 230,481–261,543 kg), 347,242 kg (95% CI = 299,610–396,467 kg), 319,137 kg (95% CI = 290,729–348,884 kg) for sandhill cranes, light geese, and dabbling ducks, respectively. We estimated the energetic requirements of all populations combined to be 4.23 × 10⁹ kcal (95% CI = 3.81 × 10⁹ to 4.53 × 10⁹ kcal), which translates to 911,813 kg of corn (95% CI = 820,821–1,006,894 kg) required to support energetic demands of these populations throughout winter. At this estimated requirement, mean corn production in the MRGV from 2014 to 2017 (1,052,006 kg) would support joint energetic needs of wintering migratory gamebirds for 101 d (95% CI = 91–113 d).

Management of supplemental corn on state and federal properties in the MRGV currently seems to produce sufficient food energy to support energetic requirements of sandhill cranes, light geese, and dabbling ducks from early November to the onset of spring migration in mid-February. Corn has substantial caloric value to migratory birds because of its high energy content (i.e., rich source of carbohydrates) and digestibility (Petrie et al. 1998). Our results suggest that heirloom varieties of corn contain about 150 to 200 kcal/kg more than GMC and NON-GMC varieties. Although this increased caloric content might seem small, NON-GMC corn is already required on Bosque del Apache NWR. Cultivating heirloom varieties may be warranted if the benefits of these varieties (e.g., they are more adapted to arid climate conditions and require less water) outweigh financial costs of purchasing seeds compared with other NON-GMCs. Regardless, maintaining corn production at current levels across state and federal properties should provide sufficient food resources to support current population levels of migratory gamebirds for the duration of the wintering period.

Our calculations of seasonal energetic expenditures used peak abundance of sandhill cranes, light geese, and dabbling ducks in the MRGV. We used peak abundance to avoid underestimation of energetic demands and to establish a corn production objective that accommodates sandhill cranes, light geese, and dabbling ducks at their largest levels during winter. Moreover, many migratory gamebirds use the MRGV as a stopover site at variable times throughout winter (Donnelly et al. 2021), which may place greater energetic demands on corn resources. While using peak abundance to calculate population-level demands might be overly cautious, underestimating corn production may have severe consequences. Unforeseen losses due to crop failure from pests, suboptimal soil conditions (e.g., pH and nutrients), and unaccounted demands placed on corn by other species (e.g., elk and deer) that are not necessarily the primary target for the resource could all reduce or limit corn availability for migratory gamebirds (Sherry et al. 2005; Danner et al. 2013; DeVore et al. 2016). Thus, total corn production objectives within a season should incorporate these potential losses.

Although our estimates of daily energy expenditures for sandhill cranes were similar to previously reported values (Tacha et al. 1987; Rainwater Basin Joint Venture 2013), our estimates did not incorporate energy expenditures related to the synthesis of lipids. Several previous studies have estimated metabolic costs of lipid production to account for the importance of building lipid reserves for migration (Krapu et al. 1985; Tacha et al. 1987; Pearse et al. 2010). Data to estimate lipid production require measurement of lipid content on collected individuals (Iverson and Vohs 1982; Krapu et al. 1985; Afton and Ankney 1991) or prediction of lipid content through body condition indices (Ardia 2005; Labocha et al. 2014). After adjusting for body size and metabolic efficiencies of lipid production, we can determine changes and metabolic cost of lipid accumulation. Although these estimates do not exist for the RMP, establishing baseline estimates of lipid deposition by sandhill cranes of the RMP may be valuable to include in seasonal energetic costs and to track any changes in physiological condition that might be related to corn quality and availability (Krapu et al. 2005).

The high temporal variability in daily depletion rates of corn is likely due to several confounding factors. First, we could not control for number of birds using supplemental corn resources at any given time, which greatly influences depletion rates. Second, we could not account for the time it takes sandhill cranes, light geese, and dabbling ducks to relocate and begin foraging in a newly mowed area. This is further conflated by how much existing corn remains unconsumed in previously mowed areas before sampling, which could be a product of mowing too much corn within a day given the daily population demands, switching to other available food resources, or changing foraging rates related to weather conditions. Minimum temperature, for example, regulates time spent foraging by migratory birds (Gates et al. 2001; Jónsson and Afton 2009). Although our results suggest that this was not influential in describing variation in daily depletion rates, it may be important and may operate beyond the resolution of our data. Interestingly, Boggie et al. (2018a) found that sandhill crane time–activity budgets indicated a temporal increase in foraging through winter, which corresponded with increasing daily depletion rates of postmowed corn. Increased food intake through winter may be a signal of premigratory fattening where sandhill cranes are steadily acquiring energy reserves needed for spring migration (Rappole 2013; Cooper et al. 2015). This insight could be incorporated into management actions by matching the amount of corn made available to both population abundances and increased monthly dietary requirements.

State and federal agencies do not traditionally begin mowing corn until December 1, which is reflected in our sampling dates. Boggie et al. (2018a) reported the RMP begin to arrive in the MRGV in late November (November 23 ± 2 d) and that the current initiation date of corn mowing likely satisfies their monthly energetic requirements. However, Boggie et al. (2018a) did have sandhill cranes tagged with satellite transmitters that arrived as early as November 5. Bearing this in mind, managers might want to consider beginning corn mowing in mid-to-late November depending on population abundances and availability of alternative food resources on state and federal properties.

While corn is a dominant dietary component of sandhill cranes (Boggie et al. 2018a) and probably the most important forage energetically, state and federal agencies manage an extensive system of moist-soil impoundments mimicking once naturally occurring wetlands. These moist-soil-managed wetlands support a diversity of natural wetland plants and invertebrate communities that provide essential sources of protein and carbohydrates for sandhill cranes, light geese, and dabbling ducks (Taylor and Smith 2003, 2005a, 2005b). Wetland resources likely fulfill any nutritional requirements beyond corn. Contemporary estimates of wetland food densities should be generated and incorporated into the total energy availability for migratory birds in the MRGV. Our study established winter energetic demands of three guilds of migratory game birds, which we suggest should be used in future conservation planning for the population. Finally, our results demonstrated that current corn production aligns with flyway population objectives for the RMP (Pacific Flyway Council and Central Flyway Council 2016), and supplemental corn in the MRGV can likely meet energetic needs of the RMP and other wintering migratory gamebirds.

Active management and manipulation of supplemental corn in the MRGV makes the food resource amendable to support the current RMP while simultaneously serving a multispecies function. Demands on the forage resource are not static and will vary through time with changes in winter habitat condition and population sizes. We encourage Federal and State land managers in the MRGV to maintain interagency coordination for flexibility in applied management actions (i.e., amount of corn grown and mowing schedules as well as variety selection), track population trends of focal species, and consider daily energetic expenditures to meet and link habitat and population objectives.

Please note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author for the article.

Reference S1.Pacific Flyway Council and Central Flyway Council. 2016. Pacific and Central Flyways Management plan for the Rocky Mountain Population of greater sandhill cranes. Vancouver, Washington: Pacific Flyway Council and Central Flyway Council, care of the U.S. Fish and Wildlife Service's Pacific Flyway Representative.

Available: https://doi.org/10.3996/JFWM-22-013.S1 (746 KB PDF)

Reference S2.Rainwater Basin Joint Venture. 2013. The Rainwater Basin Joint Venture implementation plan. Grand Island, Nebraska: Rainwater Basin Joint Venture.

Available: https://doi.org/10.3996/JFWM-22-013.S2 (1.527 MB PDF)

Reference S3. Sanchez J, Cline M. 2020. A plan for the management of waterfowl, sandhill cranes, and other migratory birds in the Middle Rio Grande Valley of New Mexico. Albuquerque, New Mexico: U.S. Fish and Wildlife Service.

Available: https://doi.org/10.3996/JFWM-22-013.S3 (2.898 MB PDF)

Reference S4. Taylor JP. 1999. A plan for the management of waterfowl, sandhill cranes, and other migratory birds in the Middle Rio Grande Valley of New Mexico. Albuquerque, New Mexico: U.S. Fish and Wildlife Service.

Available: https://doi.org/10.3996/JFWM-22-013.S4 (2.208 MB PDF)

Reference S5. Thorpe PP, Donnelly JP, Collins DP. 2016. September 2016 survey of the RMP of greater sandhill cranes. Lakewood, Colorado: Special Report in the files of the Central Flyway Representative.

Available: https://doi.org/10.3996/JFWM-22-013.S5 (152 KB PDF)

Reference S6.U.S. Department of Agriculture. 2009. Environmental assessment of crop damage management in New Mexico. Santa Fe: U.S. Department of Agriculture Animal Plant and Health Inspection Service, New Mexico Wildlife Services Program.

Available: https://doi.org/10.3996/JFWM-22-013.S6 (3.930 MB PDF)

Reference S7.U.S. Fish and Wildlife Service. 2012. North American Waterfowl Management Plan. Washington, D.C.: U.S. Fish and Wildlife Service.

Available: https://doi.org/10.3996/JFWM-22-013.S7 (2.676 MB PDF)

This study was funded by the U.S. Fish and Wildlife Service, Division of Migratory Birds (cooperative agreement F13AC00868). We thank the U.S. Fish and Wildlife Service Webless Migratory Game Bird Program, U.S. Fish and Wildlife Service Division of Migratory Birds Southwest Region, and Bosque del Apache NWR. We are also greatly appreciative of the New Mexico Department of Game and Fish and the staff at Ladd S. Gordon Waterfowl Complex for field and technical support. We thank the technicians from New Mexico State University who helped in collecting and processing data. We thank two anonymous reviewers and Associate Editor for their thorough reviews.

Any use of trade, product, website, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U.S. Government.