Research Paper

INTRODUCTION

In light of rapid rates of ecological change and a broadened stakeholder base, grassland management priorities have expanded from a focus on forage and livestock production to include such concepts as biodiversity and natural disturbance regimes (Fuhlendorf et al. 2006, Toombs and Roberts 2009). Grassland disturbances and biodiversity are both synonymous with patchiness and vegetation heterogeneity, and as a result, contemporary multi-use management often advocates for heterogeneity-based grazing strategies that mimic natural disturbances to create a shifting mosaic of vegetation structure (Fuhlendorf et al. 2017). Areas managed with heterogeneity-based grazing strategies theoretically contain patches of vegetation for wildlife species specializing on all points along a gradient of vegetation structural complexity (Hovick et al. 2015). However, wildlife species that utilize grasslands with low levels of disturbance and structural heterogeneity may not be compatible with management practices with frequent disturbance (Swengel and Swengel 2001, Arnold et al. 2007). To date, few studies have focused on the specific response of wildlife species that utilize dense vegetation structure to heterogeneity-based grazing strategies.

Patch-burn grazing (PBG) is one heterogeneity-based grazing strategy that can simultaneously promote grassland wildlife diversity and cattle production (Fuhlendorf and Engle 2001, Allred et al. 2014, Scasta et al. 2016). PBG divides a pasture into patches of near equal area and burns an individual patch annually to increase forage quality and livestock grazing in that patch. Patches with greater year(s) since fire (hereafter YSF) have reduced grazing activity and increased vegetation structure, resulting in structurally distinct patches (Fuhlendorf et al. 2009, Allred et al. 2011). The spatiotemporal distribution of fire and grazing using a PBG framework can provide greater niche differentiation, thereby providing for a more diverse grassland wildlife community (Hovick et al. 2015, Davis et al. 2016, Ricketts and Sandercock 2016). Additionally, livestock performance and pasture conditions under PBG can be similar or improved compared to management practices with uniform disturbances (Limb et al. 2011, Allred et al. 2014, Capozzelli et al. 2020, Spiess et al. 2020). However, such an approach may pose risks to grassland birds that rely on higher measures of vegetation structure during the breeding season, such as upland-nesting ducks (Augustine and Derner 2015).

Grassland managers prioritizing upland-nesting ducks are advised to use strategies with low levels of uniform disturbances that increase vegetation structure during the breeding season to better conceal nests from predators (Arnold et al. 2007, Warren et al. 2008, Bloom et al. 2013). For this reason, duck conservation is viewed as less compatible with multi-use management practices that include concepts like natural disturbance regimes, biodiversity, and forage quality (Naugle et al. 2000, Grant et al. 2011). However, periodic fire was once common in grasslands, and upland-nesting ducks can adjust nest site selection behaviors to fire return intervals that occurred pre-industrialized settlement when adequate vegetation structure is available (Grant et al. 2011). Additionally, livestock overall can have a negative effect on duck nest densities but grazing during both the growing and dormant season at low to moderate stocking rates can support acceptable levels of nesting activity (Bloom et al. 2013, Rischette et al. 2021). Contrary to a “rest is best” mentality, fire and grazing can be suitable management strategies for upland-nesting ducks when considering vegetation structure needed during the nesting season (Grant et al. 2011, Bloom et al. 2013, Rischette et al. 2021) but insight regarding their response to PBG is lacking.

Upland-nesting ducks are an important wildlife resource and certain species (i.e., Mallard [Anas platyrhynchos] and Northern Pintail [A. acuta]) are prioritized for wildlife conservation in the northern Great Plains (Cox et al. 2000, Northern Great Plains Joint Venture Technical Committee 2012). Landscapes with a focus on upland-nesting duck conservation overlap with the ranges of many declining non-game grassland bird species, which offers an opportunity to identify multi-use management practices that can support both groups. Grassland birds are adapted to patchy habitat, with different species utilizing areas with varying levels of disturbance and vegetation structure (Knopf 1996, Holcomb et al. 2014, Rosche et al. 2021). Research suggests that pastures managed with PBG can support a more diverse grassland bird community by creating a mosaic of distinct habitat patches along a postfire and grazing gradient resulting from successional stages of vegetation structure (Fuhlendorf et al. 2006, Hovick et al. 2015, Davis et al. 2016). Although the patch most recently disturbed may lack suitable structure for most nesting ducks, the greater structure provided in patches with greater YSF and less grazing pressure may provide adequate or preferred nesting structure.

Disturbance and multi-use management practices can maximize diversity and improve grassland condition but the effects of spatially and temporally varying fire and livestock grazing within a PBG framework on upland-nesting ducks are unknown. To assess the suitability of multi-use management practices for upland-nesting ducks we quantified duck nest site selection and nest survival on grasslands managed with PBG in southwestern and central North Dakota. We predicted differences in duck nest site selection and survival between YSF patches but that selection and survival would generally increase with greater YSF. Information regarding disturbance and multi-use management practices that can support upland-nesting duck populations will be essential for projected grassland management and policy in the northern Great Plains. This study will provide insight regarding disturbance frequencies for several important upland-nesting duck species in the northern Great Plains and may help to find balance between private lands livestock production and conservation.

METHODS

Study site

We used data from two locations within the North Dakota Agriculture Experiment Station’s network of research extension centers (REC): Central Grasslands REC (46°42’ N, 99°27’ W) and Hettinger REC (46°00’ N, 102°38’ W; Appendix 1). The Central Grasslands REC is located in Kidder and Stutsman Counties, North Dakota, in the Missouri Coteau ecoregion. The Hettinger REC is located in Adams County, North Dakota, in the Missouri Slope ecoregion.

The experimental pastures at Central Grasslands REC are northern-mixed grass prairie and were managed for livestock prior to the station’s inception in 1981. Vegetation cover is dominated by non-native Kentucky bluegrass (Poa pratensis L.) and smooth brome (Bromus inermis) interspersed with native grasses, such as western wheatgrass (Pascopyrum smithii [Rydb.] À. Löve), needlegrasses (Nassella viridula, Hesperostipa comata), and blue grama (Bouteloua gracilis [Willd. ex Kunth] Lag. ex Griffiths). The experimental pastures at Hettinger REC were enrolled in the Conservation Reserve Program (CRP) from 1988 through 2006, and managed for wildlife and livestock after their removal from the program (Geaumont et al. 2017). Grasslands under the original CRP contract were established with intermediate wheatgrass (Thinopyrum intermedium), smooth brome (Bromus inermis), crested wheatgrass (Agropyron cristatum), alfalfa (Medicago sativa), and sweetclover (Melilotus officinalis). These species continue to dominate the plant community within experimental pastures.

Study design

Central grasslands REC

Research was carried out on eight pastures, each 65 ha in size (Duquette et al. 2020). Pasture boundaries were delineated with a perimeter fence. We randomly assigned one of two treatments to four pastures (replicates) each. The original study design was interested in comparing PBG pastures that received two prescribed fires annually (a combination of a dormant season and a late-growing season burn) with pastures that received one dormant season fire annually. The PBG treatment with two fires was designed to have the late-growing season fire stimulate new growth and elevate forage quality during a time when cool-season grass quality typically begins to decline. Data were collected on both PBG treatments. Late-growing season fires were conducted post breeding season and would be structurally similar to dormant season fires from the year before. In addition, a preliminary survival analysis suggests fire seasonality has no effect on duck nests (Appendix 3). Therefore, we do not distinguish between growing and dormant season fires for the selection and survival analysis. We conducted dormant season prescribed fires each spring between April and May. Growing season burns occurred in late August to early September. We delineated experimental patches with mineral soil firebreaks. We grazed pastures season-long from May to October with Angus cow-calf (Bos taurus) pairs. We targeted a moderate stocking rate of 2.26–2.31 animal unit months (AUM)/ha.

Hettinger REC

We conducted research at the Hettinger REC in six experimental pastures, each 65 ha in size. We delineated pastures with perimeter fencing and divided each pasture into four equal quarters (patches) using mineral soil firebreaks. We carried out annual prescribed fires in one patch per pasture during the dormant season (late September to October). We grazed pastures season-long from late May through September. In three pastures (replicates), we randomly assigned sheep (Ovis aries) as the grazing species while the other three pastures were grazed with cattle. Pastures were grazed with assigned animals throughout the duration of the study. Both cow-calf pairs and bred Rambouillet ewes were moderately stocked at approximately 1.6 AUM/ha. With regard to ducks, we were not interested in comparing the effects of grazer type on nest selection. In addition, we found no evidence that grazer type affected duck nest survival in our preliminary analysis (Appendix 3). Therefore, we do not include grazer type in the selection or survival analysis.

Prescribed fires were first conducted in spring 2017 at Central Grasslands REC and fall 2016 at Hettinger REC, with grazing beginning in spring 2017 at both sites. These initial fires represent the beginning of the implementation of PBG at each site. For instance, experimental pastures in Hettinger by mid-May 2017 consisted of one patch classified as recently burned (0 YSF) and three classified as unburned (UNB). By May 2018, experimental pastures consisted of one 0 YSF patch (fall 2017), one patch classified as one year since fire (1 YSF), and two UNB patches. By the end of the four-year burn interval (Hettinger REC: fall 2019; Central Grasslands REC: summer 2020), pastures consisted of 0, 1, 2, and 3 YSF and unburned patches (Table 1). Our chronosequence of prescribed fire resulted in uneven classifications among patches with respect to YSF, leading to differences in cover available for nesting ducks each year.

Nest searches

We collected nest data on four upland-nesting duck species: Blue-winged Teal (Spatula discors), Gadwall (Mareca strepera), Mallard (Anas platyrhynchos), and Northern Pintail (A. acuta). Species selected for this study were chosen on the basis of their prevalence at our study sites and their use in the literature for assessing the effects of grazing strategies on duck nest site selection and survival (Warren et al. 2008, Bloom et al. 2013). We searched for upland duck nests 3-4 times per season using a chain and rope dragging method at both sites (Higgins et al. 1969). Observers used a hand-held global positioning system (GPS) to track movements and ensure systematic and complete coverage of sites during nest searches. Nest searches took place during the peak nesting season, from 6 May to 15 July. Most searches occurred between 0700 and 1300 hours to increase the probability of encountering an incubating hen (Klett et al. 1986). When a nest was located, we recorded the Universal Transverse Mercator coordinate using a GPS and flagged the nest using orange flagging tape 3 m north and south of the nest bowl. We candled eggs to determine the anticipated hatch date (Weller 1956). We revisited all nests at 2- to 5-day intervals, or until nest fate was determined. During each nest visit, we recorded hen presence/absence, time, date, number of eggs, and the stage of the nest. We considered a nest successful if ≥ 1 egg hatched from the nest.

Vegetation sampling

We assessed vegetation obstruction, litter cover, and litter depth at the nest bowl within two days of a successful nest hatch or after the estimated hatch date for unsuccessful nests to reduce bias in differential vegetation phenology. We measured vegetation obstruction (VOR) by reading the highest strata (dm) on a Robel pole that was at least 50% obscured from a distance of 4 m and a height of 1 m at each cardinal direction (Robel et al. 1970). We estimated percent canopy cover of litter using a 1 m × 1 m quadrat centered on the nest (Daubenmire 1959). We measured litter depth (mm) at each corner of the quadrat using a measuring tape or wooden ruler.

Site selection analysis

We combined nest data from both RECs into one data set for all analyses. We used a selection ratio with known proportions of resources to assess the strength of nest site selection of upland nesting waterfowl by patch YSF (Manly et al. 2002). We used a Design 1 framework to account for encountering the same hen in multiple years from repeated sampling efforts (Manly et al. 2002). Availability was defined as the entire delineated area of both Central Grasslands and Hettinger REC. The selection ratio (w_i) was calculated as

(1)

with o_i representing the proportion of nests in a specific YSF and π_i as the proportion of a specific YSF on the landscape (Manly et al. 2002, Duquette et al. 2020). The standard error of w_i was calculated

(2)

with u representing the total number of nests in the sample and presented as 85% CIs (Duquette et al. 2020). We averaged YSF proportions across sites and years (Seavy et al. 2012). Selection ratio values > 1 suggest disproportionate selection for a resource compared to its availability, whereas values < 1 suggest avoidance. Additionally, we included patch level vegetation data for VOR, litter cover, and litter depth to illustrate patch level differences in vegetation structure (Fig. 1). Patch level vegetation data were collected along predefined transects within each patch and were not directly associated with waterfowl nests or included in the nest selection or survival analysis.

Survival analysis

We used logistic-exposure method to assess the influence of litter cover, litter depth, vegetation obstruction, and YSF at nest sites on the daily survival rates (hereafter DSR) of upland-nesting ducks (Shaffer 2004, Shew et al. 2019). This formulation expresses the probability of a nest surviving one interval between nest visits (i.e., exposure periods) as a function of covariates, with each duration assigned a binary fate (success or failure; Shew et al. 2019). Vegetation variables were selected on the basis of a priori knowledge of upland-nesting waterfowl nest survival. Year(s) since fire was treated as a factor variable with 0 YSF used as the reference level for betas of other YSFs. Prior to fitting models, we assessed vegetation variables for multicollinearity using Pearson’s correlation (r > |0.7|; Coppedge et al. 2008). Correlated variables were not included in the same model. Our a priori model selection approach included all single variable models, all 2- and 3-way additive combination models, an intercept (null) model, and a global model (Appendix 2). All models included a unique nest identifier as a random effect to account for repeated measures. Models also included plot id and year as a random effect to account for annual and spatial variability, respectively. Models were ranked based on Akaike’s information criterion values corrected for small sample size (AIC_c) and in relation to the null model (Burnham and Anderson 2002). Models within 2 AIC_c units with > ±1 parameter of the top-ranked model were further evaluated to determine variable significance (Burnham and Anderson 2002). We considered variables with coefficient estimates that included zero in the 85% CIs to be uninformative (Arnold 2010). Parameter coefficients were reported from model-averaged top models (i.e., with 2 AIC_c of top models). The survival analysis was conducted by using lme4 package in the R statistical computing environment (R Core Team 2020).

RESULTS

We found 478 duck nests from 2017 to 2020. Our data consisted of 230 Blue-winged Teal, 72 Gadwalls, 71 Mallards, and 105 Northern Pintails. The number of nests varied by patch YSF, with 189 nests in unburned patches, 78 nests in 0 YSF patches, 77 nests in 1 YSF patches, 82 nests in 2 YSF patches, and 52 nests in 3 YSF patches.

Blue-winged Teal

Nest site selection

Blue-winged Teal avoided nesting in 0 YSF (w = 0.70, 85% CI = 0.61 to 0.78), 1 YSF (w = 0.84, 85% CI = 0.75 to 0.93) and unburned (w = 0.84, 85% CI = 0.76 to 0.91) patches, and selected to nest in 2 YSF (w = 1.63, 85% CI = 1.54 to 1.71) and 3 YSF (w = 2.70, 85% CI = 2.62 to 2.79) patches (Fig. 2).

Nest survival

The top-ranked nest survival model for Blue-winged Teal included litter depth (Table 2). Three additional models fell within 2 AIC units of the top model, including VOR, percent litter, and YSF. Nest survival decreased as litter depth increased (β = -0.01, 85% CI = -0.02 to -0.003) and increased when a nest was located in 2 YSF (β = 0.11, 85% CI = 0.10 to 1.08) and unburned (β = 0.16, 85% CI = 0.33 to 1.30) patches (Fig. 3). Model averaged coefficients for VOR (β = -0.01, 85% CI = -0.21 to 0.10), percent litter (β = -0.0002, 85% CI = -0.01 to 0.01), 1 YSF (β = 0.06, 85% CI = -0.18 to 0.82), and 3 YSF (β = 0.10, 85% CI = -0.04 to 1.06) patches had 85% CI that included zero.

Gadwall

Nest site selection

Gadwall avoided nesting in 0 YSF (w = 0.48, 85% CI = 0.32 to 0.64) patches, and selected to nest in 2 YSF (w = 1.66, 85% CI = 1.50 to 1.81) and 3 YSF (w = 2.03, 85% CI = 1.87 to 2.19) patches (Fig. 2). We found no evidence of nest site selection or avoidance in 1 YSF (w = 1.03, 85% CI = 0.87 to 1.18) and unburned (w = 0.96, 85% CI = 0.83 to 1.09) patches (Fig. 2).

Nest survival

The top-ranked nest survival model for Gadwall included YSF (Table 2). Four additional models fell within 2 AIC units of the top model, including VOR, percent litter, and litter depth. Nest survival decreased as litter depth increased (β = -0.001, 85% CI = -0.02 to -0.002) and decreased when a nest was located in 3 YSF (β = -2.16, 85% CI = -3.68 to -0.64) patches (Fig. 3). Model-averaged coefficients for VOR (β = 0.05, 85% CI = -0.10 to 0.41), percent litter (β = -0.001, 85% CI = -0.04 to 0.02), 1 YSF (β = 0.11, 85% CI = -1.16 to 1.38), 2 YSF (β = 0.95, 85% CI = -0.39 to 2.29), and unburned patches (β = -0.86, 85% CI = -2.22 to 0.50) had 85% CIs that included zero.

Mallard

Nest site selection

Mallards avoided nesting in 0 YSF (w = 0.61, 85% CI = 0.45 to 0.77) and 1 YSF (w = 0.72, 85% CI = 0.56 to 0.88) patches, and selected to nest in 2 YSF (w = 1.20, 85% CI = 1.04 to 1.36), 3 YSF (w = 1.29, 85% CI = 1.12 to 1.45), and unburned (w = 1.24, 85% CI = 1.12 to 1.36) patches (Fig. 2).

Nest survival

The top-ranked nest survival model for Mallards was the null model (Table 2).

Northern Pintail

Nest site selection

Northern Pintails avoided nesting in 3 YSF (w = 0.87, 85% CI = 0.73 to 1.00) patches, and selected to nest in 1 YSF (w = 1.14, 85% CI = 1.01 to 1.26) and 2 YSF (w = 1.13, 85% CI = 1.00 to 1.27) patches. We found no evidence of nest site selection or avoidance in 0 YSF (w = 0.95, 85% CI = 0.82 to 1.07) and unburned (w = 0.95, 85% CI = 0.84 to 1.06) patches (Fig. 2).

Nest survival

The top-ranked nest survival model for Northern Pintails was the null model (Table 2).

DISCUSSION

Our study results indicate that PBG may be a compatible multi-use management strategy for four common upland-nesting ducks within two regions of the northern Great Plains. In general, ducks avoided early YSF patches and selected for later YSF patches over unburned areas. These results suggest that PBG can provide attractive vegetation structure for nesting ducks in patches with greater YSF despite the presence of burning and livestock grazing. Patch-burn grazing had varying impacts on the survival of duck nests, with Gadwalls experiencing low survival in later YSF patches. Patch-burn grazing has been promoted throughout the Great Plans in part because of its ability to increase grassland heterogeneity and biodiversity (Scasta et al. 2016). Our results demonstrate that patches with greater YSF can provide nesting sites for ducks and may even be selected for in greater frequency to unburned patches, but other factors may be at play when it comes to nest survival.

Duck nest site selection was influenced by patch YSF, with selection generally being low in 0 and 1 YSF and greater in 2 and 3 YSF compared to unburned patches. Previous studies that have evaluated the impact of fire in the absence of grazing on the nest site selection of upland-nesting ducks in the northern Great Plains have reported similar results, with lower nest densities in early YSF grasslands, but differed slightly from our results in that selection generally stabilized or decreased ≥ 3 YSF (Devries and Armstrong 2011, Grant et al. 2011). Perhaps livestock’s preference for recently burned areas within a PBG framework resulted in our findings differing from previous studies because this interaction does alter vegetation structural characteristics (McGranahan et al. 2012). By preferentially grazing new growth in the most recent burn patch, livestock keep vegetation structure low. Low vegetation accumulation in recent burn patches may result in different structural attributes for ducks than those found in similar YSF grasslands devoid of livestock. The potential structural differences post-fire between grasslands with and without livestock grazing may have resulted in the differences we observed in duck nest site selection in relation to YSF compared to previous studies (Devries and Armstrong 2011, Grant et al. 2011).

Differences we observed in nest site selection of ducks were likely driven by individual species preference for nesting vegetation structure (Higgins 1986, Grant et al. 2011). Blue-winged Teal, Gadwall, and Mallard prefer denser and taller vegetation structure for nest sites, whereas Northern Pintails utilize a wider gradient of structure (Fondell et al. 2004). Our nest site selection results for ducks that prefer greater measures of vegetation structure for nesting was consistent with patch scale vegetation in that selection and vegetation structure increased as YSF increased and was similar or greater in 2 and 3 YSF than in unburned patches (Figs. 1 and 2). Selection for later YSF patches over unburned patches suggests that PBG can provide attractive nesting areas for certain species of upland ducks despite increased levels of disturbance. However, we emphasize that fire frequency within our PBG framework was tailored to ensure that at least a portion of each pasture was burned annually to attract grazing livestock (Bragg 1995). Implementation of PBG in other regions of the Great Plains will require similar forethought in order to provide attractive nesting cover for species that prefer denser vegetation structure in later YSF grasslands (Grant et al. 2011, Augustine and Derner 2015).

Nest survival was greater for Blue-winged Teal in 2 YSF and unburned patches, lower for Gadwall in 3 YSF patches, and decreased as litter depth increased for both species (Fig. 3). We speculate that higher patch-level measures of litter depth in 3 YSF in comparison to 2 YSF and unburned patches might be influencing nest survival among duck species (Voorhees and Cassel 1980, Devries and Armstrong 2011; Fig. 1). Areas with greater amounts of residual vegetation can have low duck nest survival as a result of increased foraging efforts by nest predators to those areas because of greater prey abundance (i.e., small mammals, insects, and avian nests; Maile 2003, Devries and Armstrong 2011). We did not assess relationships between vegetation structure and prey abundance, but Blue-winged Teal and Gadwall did experience lower nest survival in areas with greater litter depth (Figs. 1 and 3). Additionally, Blue-winged Teal had greater nest survival in 2 YSF and unburned patches. Two YSF and unburned patches had low to moderate amounts of litter depth compared to 3 YSF patches. We found areas with greater litter accumulation and low nest survival were generally confined to 3 YSF patches, especially for Gadwalls. Therefore, PBG might not limit duck production with adjacent 1 and 2 YSF patches having vegetative characteristics that could support upland nesting ducks.

Patch-burn grazing is effective in creating heterogeneous vegetation structure, which subsequently influenced selection and survival of upland-nesting ducks (Fuhlendorf et al. 2009, Allred et al. 2014). However, the contrasting effects of patch YSF between duck nest selection and survival shown by certain species suggest further evaluation of PBG for duck conservation might be necessary. For example, Gadwall had greater selection but low survival in 3 YSF patches. This could be problematic if Gadwalls show avoidance and low survival in other YSF patches, but this was not the case in our study. Gadwalls still showed high selection and survival in 2 YSF patches relative to early YSF and unburned patches, which could buffer lower production in 3 YSF patches. In other instances, Blue-winged Teal had high nest survival in unburned patches, which would no longer be available after the completion of the first burn cycle within a PBG framework, suggesting the potential for a slight decrease in projected annual survival following the first four-year burn rotation. However, Blue-winged Teal did not select for unburned patches, and mean nest survival exceeded levels deemed necessary to sustain their population in 0 YSF, 1 YSF, and 3 YSF patches (i.e., DSR ≥ 0.948; Klett et al. 1988). Nest site selection and survival of ducks and other birds that utilize dense nesting cover is influenced by the amount of vegetation regrowth following fire (Grant et al. 2011, Augustine and Derner 2015). Rates of vegetation regrowth in PBG pastures fluctuate by annual precipitation, regional climatic conditions, local site productivity, and stocking rate. This may mean that additional management strategies are required to ensure patches of dense nesting cover are available for nesting ducks (Derner and Hart 2007, Augustine and Derner 2015).

Growing environmental uncertainty combined with a broadening stakeholder interest means that grassland managers must carefully consider a wide range of strategies to meet diverse objectives. Although waterfowl management has long been associated with strategies designed to limit disturbance and preserve maximum levels of nesting cover, our data show that the shifting mosaic of vegetation structure created by PBG is largely compatible with upland-nesting duck conservation. Therefore, managers with an interest in duck production are not precluded from the many demonstrated benefits of PBG, such as improved forage quality, cattle gains, drought resistance, and increased biodiversity (Fuhlendorf et al. 2006, Allred et al. 2014, Scasta et al. 2016). However, as is often the case, responses to PBG are species-specific; care must be taken to provide adequate cover for species that may be particularly sensitive to disturbance. Our research provides an important step towards integrating waterfowl management plans with broader rangeland production and biodiversity objectives.

CONCLUSION

Our findings suggest that PBG, in combination with moderate stocking rates, can provide attractive areas for upland-nesting ducks in the northern Great Plains. However, Gadwalls may not be compatible with the interactive disturbances of fire and grazing if there are limited areas with attractive nesting cover. Managers will need to ensure that > 1 YSF patches consist of vegetation characteristics that will attract duck nesting activity. Stochastic events can limit vegetative regrowth (e.g., drought, hail, wildfires) and may require managers to adjust fire frequency and/or stocking rates in order to provide attractive nesting cover in multiple YSF patches. Wildlife managers might consider PBG as an alternative to uniform applications of prescribed fire when disturbing grasslands because of PBG’s ability to provide productive nesting areas for ducks and the economic gains associated with keeping livestock on the land.

LITERATURE CITED

Allred, B. W., S. D. Fuhlendorf, D. M. Engle, and R. D. Elmore. 2011. Ungulate preference for burned patches reveals strength of fire-grazing interaction. Ecology and Evolution 1(2):132-144. https://doi.org/10.1002/ece3.12

Allred, B. W., J. D. Scasta, T. J. Hovick, S. D. Fuhlendorf, and R. G. Hamilton. 2014. Spatial heterogeneity stabilizes livestock productivity in a changing climate. Agriculture, Ecosystems & Environment 193:37-41. https://doi.org/10.1016/j.agee.2014.04.020

Arnold, T. W., L. M. Craig-Moore, L. M. Armstrong, D. W. Howerter, J. H. Devries, B. L. Joynt, R. B. Emery, and M. G. Anderson. 2007. Waterfowl use of dense nesting cover in the Canadian parklands. Journal of Wildlife Management 71(8):2542-2549. https://doi.org/10.2193/2005-568

Arnold, T. W. 2010. Uninformative parameters and model selection using Akaike’s information criterion. Journal of Wildlife Management 74(6):1175-1178. https://doi.org/10.1111/j.1937-2817.2010.tb01236.x

Augustine, D. J., and J. D. Derner. 2015. Patch‐burn grazing management, vegetation heterogeneity, and avian responses in a semi‐arid grassland. Journal of Wildlife Management 79(6):927-936. https://doi.org/10.1002/jwmg.909

Bragg, T. B. 1995. Climate, soils and fire: the physical environment of North American grasslands. Pages 49-81 in A. Joern and K. H. Keeler, editors. The changing prairie: North American grasslands. Oxford University Press, New York, New York, USA.

Bloom, P. M., D. W. Howerter, R. B. Emery, and L. M. Armstrong. 2013. Relationships between grazing and waterfowl production in the Canadian prairies. Journal of Wildlife Management 77(3):534-544. https://doi.org/10.1002/jwmg.497

Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference: a practical information-theoretic approach. Second edition. Springer-Verlag, New York, New York, USA.

Capozzelli, J. F., J. R. Miller, D. M. Debinski, and W. H. Schacht. 2020. Restoring the fire-grazing interaction promotes tree-grass coexistence by controlling woody encroachment. Ecosphere 11(2):e02993. https://doi.org/10.1002/ecs2.2993

Coppedge, B. R., S. D. Fuhlendorf, W. C. Harrell, and D. M. Engle. 2008. Avian community response to vegetation and structural features in grasslands managed with fire and grazing. Biological Conservation 141(5):1196-1203. https://doi.org/10.1016/j.biocon.2008.02.015

Cox, R. R., Jr., D. H. Johnson, M. A. Johnson, R. E. Kirby, J. W. Nelson, and R. E. Reynolds. 2000. Waterfowl research priorities in the northern Great Plains. Wildlife Society Bulletin 28(3):558-564.

Daubenmire, R. 1959. A canopy-coverage method of vegetational analysis. Northwest Science 33:39-64.

Davis, C. A., R. T. Churchwell, S. D. Fuhlendorf, D. M. Engle, and T. J. Hovick. 2016. Effect of pyric herbivory on source-sink dynamics in grassland birds. Journal of Applied Ecology 53(4):1004-1012. https://doi.org/10.1111/1365-2664.12641

Derner, J. D., and R. H. Hart. 2007. Grazing-induced modifications to peak standing crop in northern mixed-grass prairie. Rangeland Ecology and Management 60(3):270-276. https://doi.org/10.2111/1551-5028(2007)60[270:GMTPSC]2.0.CO;2

Devries, J. H., and L. M. Armstrong. 2011. Impact of management treatments on waterfowl use of dense nesting cover in the Canadian parklands. Journal of Wildlife Management 75(6):1340-1349. https://doi.org/10.1002/jwmg.164

Duquette, C. A., T. J. Hovick, R. F. Limb, D. A. McGranahan, and K. K. Sedevic. 2020. Restored fire and grazing regimes influence nest selection and survival in Brewer’s Blackbirds Euphagus cyanocephalus. Acta Ornithologica 54(2):171-180. https://doi.org/10.3161/00016454AO2019.54.2.004

Fondell, T. F., and I. J. Ball. 2004. Density and success of bird nests relative to grazing on western Montana grasslands. Biological Conservation 117(2):203-213. https://doi.org/10.1016/S0006-3207(03)00293-3

Fuhlendorf, S. D., and D. M. Engle. 2001. Restoring heterogeneity on rangelands: ecosystem management based on evolutionary grazing patterns: we propose a paradigm that enhances heterogeneity instead of homogeneity to promote biological diversity and wildlife habitat on rangelands grazed by livestock. BioScience 51(8):625-632. https://doi.org/10.1641/0006-3568(2001)051[0625:RHOREM]2.0.CO;2

Fuhlendorf, S. D., D. M. Engle, J. A. Y. Kerby, and R. Hamilton. 2009. Pyric herbivory: rewilding landscapes through the recoupling of fire and grazing. Conservation Biology 23(3):588-598. https://doi.org/10.1111/j.1523-1739.2008.01139.x

Fuhlendorf, S. D., R. W. Fynn, D. A. McGranahan, and D. Twidwell. 2017. Heterogeneity as the basis for rangeland management. Pages 169-196 in D. Briske, editor. Rangeland systems: processes, management and challenges. Springer, Cham, Switzerland. https://doi.org/10.1007/978-3-319-46709-2_5

Fuhlendorf, S. D., W. C. Harrell, D. M. Engle, R. G. Hamilton, C. A. Davis, and D. M. Leslie, Jr. 2006. Should heterogeneity be the basis for conservation? Grassland bird response to fire and grazing. Ecological Applications 16(5):1706-1716. https://doi.org/10.1890/1051-0761(2006)016[1706:SHBTBF]2.0.CO;2

Geaumont, B. A., K. K. Sedivec, and C. S. Schauer. 2017. Ring-necked pheasant use of post-conservation reserve program lands. Rangeland Ecology & Management 70(5):569-575. https://doi.org/10.1016/j.rama.2017.04.003

Grant, T. A., T. L. Shaffer, E. M. Madden, and G. B. Berkey. 2011. Ducks and passerines nesting in northern mixed‐grass prairie treated with fire. Wildlife Society Bulletin 35(4):368-376. https://doi.org/10.1002/wsb.65

Higgins, K. F. 1986. Interpretation and compendium of historical fire accounts in the northern Great Plains. U.S. Fish and Wildlife Service, Department of the Interior, Washington, D.C., USA.

Higgins, K. F., L. M. Kirsch, and I. J. Ball, Jr. 1969. A cable-chain device for locating duck nests. Journal of Wildlife Management 33(4):1009-1011. https://doi.org/10.2307/3799339

Holcomb, E. D., C. A. Davis, and S. D. Fuhlendorf. 2014. Patch-burn management: implications for conservation of avian communities in fire-dependent sagebrush ecosystems. Journal of Wildlife Management 78(5):848-856. https://doi.org/10.1002/jwmg.723

Hovick, T. J., R. D. Elmore, S. D. Fuhlendorf, D. M. Engle, and R. G. Hamilton. 2015. Spatial heterogeneity increases diversity and stability in grassland bird communities. Ecological Applications 25(3):662-672. https://doi.org/10.1890/14-1067.1

Klett, A. T., H. F. Duebbert, C. A. Faanes, and K. F. Higgins. 1986. Techniques for studying nest success of ducks in upland habitats in the prairie pothole region. U.S. Fish and Wildlife Service, Department of Interior, Washington D.C., USA.

Klett, A. T., T. L. Shaffer, and D. H. Johnson. 1988. Duck nest success in the prairie pothole region. Journal of Wildlife Management 52(3):431-440. https://doi.org/10.2307/3801586

Knopf, F. L. 1996. Prairie legacies—birds. Pages 135-148 in F. B. Samson and F. L. Knopf, editors. Prairie conservation: preserving North America’s most endangered ecosystem. Island, Washington, D.C., USA.

Limb, R. F., S. D. Fuhlendorf, D. M. Engle, J. R. Weir, R. D. Elmore, and T. G. Bidwell. 2011. Pyric-herbivory and cattle performance in grassland ecosystems. Rangeland Ecology & Management 64(6):659-663. https://doi.org/10.2111/REM-D-10-00192.1

Maile, B. A. 2003. Effects of planted cover management on buffer prey and their role in upland waterfowl nest success. Thesis. University of Alberta, Edmonton, Alberta, Canada.

Manly B. F., L. L. McDonald, D. L. Thomas, T. L. McDonald, and W. P. Erickson. 2002. Resource selection by animals: statistical design and analysis for field studies. Second edition. Kluwer Academic Publishers, Dordrecht, Netherlands.

McGranahan, D. A., D. M. Engle, S. D. Fuhlendorf, S. J. Winter, J. R. Miller, and D. M. Debinski. 2012. Spatial heterogeneity across five rangelands managed with pyric‐herbivory. Journal of Applied Ecology 49(4):903-910. https://doi.org/10.1111/j.1365-2664.2012.02168.x

Naugle, D. E., K. F. Higgins, and K. K. Bakker. 2000. A synthesis of the effects of upland management practices on waterfowl and other birds in the northern Great Plains of the U.S. and Canada. College of Natural Resources, University of Wisconsin-Stevens Point, Stevens Point, Wisconsin, USA.

Northern Great Plains Joint Venture Technical Committee. 2012. Northern Great Plains joint venture, priority birds. Report to the Northern Great Plains Joint Venture Management Board. August 2012.

R Core Team. 2020. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/

Ricketts, A. M., and B. K. Sandercock. 2016. Patch‐burn grazing increases habitat heterogeneity and biodiversity of small mammals in managed rangelands. Ecosphere 7(8):e01431. https://doi.org/10.1002/ecs2.1431

Rischette, A. C., B. A. Geaumont, R. D. Elmore, C. S. Schauer, K. K. Sedivec, and T. J. Hovick. 2021. Duck nest density and survival in post‐conservation reserve program lands. Wildlife Society Bulletin 45(4):630-637. https://doi.org/10.1002/wsb.1233

Robel, R. J., J. N. Briggs, A. D. Dayton, and L. C. Hulbert. 1970. Relationships between visual obstruction measurements and weight of grassland vegetation. Journal of Range Management 23(4):295-297. https://doi.org/10.2307/3896225

Rosche, S. B., C. E. Moorman, A. J. Kroeger, K. Pacifici, J. G. Jones, and C. S. Deperno. 2021. Effects of prescribed fire on Northern Bobwhite nesting ecology. Wildlife Society Bulletin 45(2):249-257. https://doi.org/10.1002/wsb.1183

Scasta, J. D., E. T. Thacker, T. J. Hovick, D. M. Engle, B. W. Allred, S. D. Fuhlendorf, and J. R. Weir. 2016. Patch-burn grazing (PBG) as a livestock management alternative for fire-prone ecosystems of North America. Renewable Agriculture and Food Systems 31(6):550-567. https://doi.org/10.1017/S1742170515000411

Seavy, N. E., R. D. Burnett, and P. J. Taille. 2012. Black‐backed Woodpecker nest‐tree preference in burned forests of the Sierra Nevada, California. Wildlife Society Bulletin 36(4):722-728. https://doi.org/10.1002/wsb.210

Shaffer, T. L. 2004. A unified approach to analyzing nest success. Auk 121(2):526-540.

Shew, J. J., C. K. Nielsen, and D. W. Sparling. 2019. Finer-scale habitat predicts nest survival in grassland birds more than management and landscape: a multi-scale perspective. Journal of Applied Ecology 56(4):929–945. https://doi.org/10.1111/1365-2664.13317

Spiess, J. W., D. A. McGranahan, B. A. Geaumont, K. Sedivec, M. Lakey, M. Berti, T. J. Hovick, and R. F. Limb. 2020. Patch-burning buffers forage resources and livestock performance to mitigate drought in the northern Great Plains. Rangeland Ecology & Management 73(4):473-481. https://doi.org/10.1016/j.rama.2020.03.003

Swengel, A. B., and S. R. Swengel. 2001. Effects of prairie and barrens management on butterfly faunal composition. Biodiversity & Conservation 10:1757-1785. https://doi.org/10.1023/A:1012051510584

Toombs, T. P., and M. G. Roberts. 2009. Are natural resources conservation service range management investments working at cross-purposes with wildlife habitat goals on western United States rangelands? Rangeland Ecology & Management 62(4):351-355. https://doi.org/10.2111/08-027.1

Voorhees, L. D., and J. F. Cassel. 1980. Highway right-of-way: mowing versus succession as related to duck nesting. Journal of Wildlife Management 44(1):155-163. https://doi.org/10.2307/3808361

Warren, J. M., J. Rotella, and J. Thompson. 2008. Contrasting effects of cattle grazing intensity on upland-nesting duck production at nest and field scales in the Aspen Parkland, Canada. Avian Conservation and Ecology 3(2):6. https://doi.org/10.5751/ACE-00264-030206

Weller, M. W. 1956. A simple field candler for waterfowl eggs. Journal of Wildlife Management 20(2):111-113. https://doi.org/10.2307/3797414