Nutritional ecology of large mammals has been of primary concern for scientists trying to understand the underlying drivers of population dynamics. Selection of forages to support reproduction is an important component of how individuals interact with their environment (Barboza et al., 2009; Parker et al., 2009). Nutritional interactions are present at all levels of an ecosystem and result in the population characteristics we observe in real time (Parker et al., 2009). For instance, individuals that obtain high-quality nutrition from their food, especially when growing, often are in better condition and obtain larger body size than those individuals that consume lower quality forage (Cook et al., 2004; Monteith et al., 2009; Parker et al., 2009; Tollefson et al., 2010; Jackson et al., 2021). Individuals on a higher nutritional plane are more likely to survive, produce more offspring during their lifetime and, thus, exhibit high reproductive fitness (Stewart et al., 2005; Parker et al., 2009; Stephenson et al., 2020). Furthermore, individuals that have access to better nutrition during portions of the year and have higher body condition are better able to survive prolonged periods of low-quality forage or harsh climatic conditions than those with lesser nutritional resources (Mautz, 1978; Stewart et al., 2005; Monteith et al., 2013; Monteith K. L. et al., 2014).

Selection of high-quality forage is dependent upon environmental constraints in the region, but selection of forages also varies among individuals in the population. Over long temporal scales, seasonal variation in availability and quality of forage become linked with life-history characteristics of animals inhabiting the region. For instance, timing of parturition among large mammals has evolved to coincide with the highest available forage quality to meet the metabolic needs of lactation (Bowyer, 1991; Long et al., 2016; Neumann et al., 2020). Furthermore, ungulates often select high-quality forages, usually early growth of forbs and grasses during spring and early summer but switch to browsing on shrubs during autumn and winter months, when vegetation quality is often lower and forage is limited (Boeker et al., 1972; Collins and Urness, 1983; Miller and Gaud, 1989; Beck and Peek, 2005a). Seasonal shifts in forage selection allow individuals to maintain better body condition, produce larger offspring, and higher recruitment rates, resulting in increased reproductive fitness (Parker et al., 2009).

Selection of forage often varies with sex (Bleich et al., 1997 2019; Barboza and Bowyer, 2000, 2001), reproductive status (Clutton-Brock et al., 1982; Cook et al., 2013), and population density (Stewart et al., 2011; Bowyer et al., 2014, 2020). Intraspecific competition leads to much of the variation we observe among individuals in selection of nutritional resources across the landscape. Polygynous ungulates exhibit strong sexual dimorphism and often exhibit a dramatic difference in body size between males and females. As a result, dietary requirements often differ between sexes and are reflected in differences in selection of forages and often linked to reproductive success (Barboza and Bowyer, 2000, 2001; Schroeder et al., 2010). Males are usually larger bodied than females, have larger rumens, fewer metabolic demands associated with reproduction, may thrive on lower quality forage than that required by reproductive females, and have no contribution to survival of offspring beyond mating (Barboza and Bowyer, 2000). Furthermore, males can use habitats with greater predation risk than females with dependent young (Bleich et al., 1997; Barboza and Bowyer, 2000). In addition to meeting nutritional requirements associated with gestation, following parturition females must protect and provision their offspring until weaning (Oftedal, 1985; Robbins, 1993; Barboza et al., 2009). High-quality diets acquired by parturient females, or those provisioning dependent young, result in larger offspring with higher survival and recruitment (Keech et al., 2000; Monteith et al., 2009; Heffelfinger et al., 2018). Lactating females often lose body condition, as a result of increased protein and energy demands, while non-reproductive females are able to regain somatic reserves during similar time periods (Holt et al., 1992; Barboza and Parker, 2008), likely resulting in selection of different forages by females of varying reproductive status.

While provisioning young, some species may trade nutrition of the mother for greater safety and higher survival of offspring, thereby exhibiting risk-averse behavior (Festa-Bianchet, 1988; Berger, 1991; Bleich et al., 1997; Long et al., 2009). Females making this tradeoff can only do so for a limited period because metabolic needs during lactation are higher than at other times of the year and must be met to ensure offspring obtain the appropriate amount of nutrients during growth (Barboza et al., 2009). Therefore, we expect the degree of those nutritional tradeoffs to decline as juveniles grow and females exhaust somatic reserves obtained prior to parturition. This tradeoff does not necessarily indicate that females select the highest quality diets available, but rather they select vegetation that is sufficient to overcome the constraints associated with gestation and lactation, while also enhancing safety of offspring. Additionally, if females can access high quality and abundant vegetation while simultaneously selecting habitat types that increase the safety and survival of offspring (Pierce et al., 2004; Heffelfinger et al., 2020), then a tradeoff would not be expected to occur.

Quantifying diet quality and composition, at multiple states of reproduction, is useful to understanding habitat selection and life-history strategies of wild ungulates. Furthermore, this information can positively influence the conservation of target species through habitat management. For instance, identifying how diet content and quality adjust across different life-history strategies can be used to improve the reproductive fitness of individuals by focusing habitat management on those forage species that best improve an individuals’ nutritional condition. Additionally, this information can be used to ensure that translocation sites have similar vegetative characteristics to support individuals throughout different reproductive periods, therefore, increasing the likelihood of success of the translocation through high survival and the recruitment of offspring.

We used a population of desert bighorn sheep (Ovis canadensis nelsoni) to identify changes in diet composition and quality throughout the year, especially around reproduction (prior to parturition, following parturition and while lactating) and for females following death of their offspring. Our objectives were to (1) identify diet composition and quality across seasons and between adult females and their offspring and (2) identify diet composition and quality across the parturition season based on timing of parturition and offspring provisioning status. We hypothesized that if females with offspring traded nutrition for safety of offspring, then those females without offspring would have higher quality diets than those with dependent young. Additionally, we hypothesized that females that were not provisioning offspring would have higher diet diversity compared to those with dependent young, because they likely are able to exploit a greater diversity of habitat types than mothers with dependent young.

We found minimal overlap in season*year interactions in our seasonal fecal nitrogen model, indicating strong differences in fecal nitrogen among seasons and years (Table 1). In spring 2017, diet quality of adult females overlapped that of juveniles, but did not overlap the following year. In 2018, we observed strong overlap in diet quality among spring, summer, and winter diet quality, whereas only summer and winter diets overlapped in 2017 (Table 1). Juveniles (i.e., young of the year) had the highest fecal nitrogen regardless of year, likely because they were gaining extra nutrients from nursing (Table 1; Figure 2). Adult females tended to have their highest quality diets during spring months across all years; spring diet quality notably declined from 2016 to 2018 (Table 1; Figure 2). Diet quality during summer months was consistently higher than winter months, yet summer and winter months overlapped 19 and 28% in 2017 and 2018 (Table 1; Figure 2). Diet quality during winter was consistently lowest (Table 1; Figure 2).

We found overlap in all parameters for parturition period and a positive relationship of ordinal date on percent fecal nitrogen (Table 2). Predictions from estimates for parturition period at ordinal date 115 (April 25) indicated strong overlap in nitrogen content in fecal samples between post-parturient females and those that had not given birth (Figure 3). Pre-parturient females had more variation in percent fecal nitrogen than those that were provisioning offspring (Table 2; Figure 3), but pre-parturient females had lower fecal nitrogen. Predictions at ordinal date 115 indicated females that had lost a neonate and those that were not pregnant had the highest percent of nitrogen present in feces (Figure 3).

We found strong overlap across season and year in diversity of diet in fecal samples. Diversity of diet varied across seasons as well as between juveniles and adults (Table 3; Figure 4). Furthermore, diet diversity decreased as the year progressed (Table 3; Figure 4). Juveniles tended to have a higher diversity of diet, however, there was strong overlap with adult samples across seasons (Table 3; Figure 4). Diet composition was distinct between seasons and age classes, although minimal variance was explained by the estimated parameters (Season: MS = 1.376, r² = 0.071, value of p = 0.001; Year: MS = 0.622, r² = 0.011, value of p = 0.02; Interaction: MS = 0.306, r² = 0.016, value of p = 0.166).

Diet diversity of females was also dependent on the parturition categories (Table 4; Figure 5). Females that lost offspring had the greatest diet diversity, although there was 36% overlap with females that were provisioning offspring (Table 4; Figure 5). Pre-parturient females had the lowest diet diversity, but overlapped diet diversity with females provisioning offspring by 14% (i.e., post parturition; Table 4; Figure 5). Diet composition was also distinct between parturition categories, although the variance explained by the estimated parameters was low (Parturition categories: MS = 0.968, r² = 0.056, value of p = 0.001; Ordinal date: MS = 0.979, r² = 0.028, value of p = 0.001).

Diet composition was largely similar across seasons, with some variation in both frequency of occurrence and relative read abundance (Table 5). Shadscale saltbush, horsebrush (Tetradymia spp.), and granite prickly phlox (L. pungens) were the only plant taxa that showed significant differences across seasons in both frequency of occurrence (shadscale saltbush: value of p = 0.02, horsebrush: value of p = 0.04, granite prickly phlox: value of p = 0.02) and relative read abundance (shadscale saltbush: value of p = 0.02, horsebrush: value of p = 0.02, granite prickly phlox: value of p = 0.02) datasets (Table 5). Shadscale saltbush was only present in the spring and winter diets of adult females, whereas granite prickly phlox and horsebrush were only present in winter diets. Saxifragaceae (value of p = 0.04) was only present in the spring and summer diets of adults (Table 5). Relative read abundance indicated that Stansbury cliffrose (value of p = 0.02), increased in individual diets throughout the year (Table 5).

Frequency of occurrence and relative read abundance results indicate that females in different parturition groups had very similar diets (Table 6). Only pre-parturient females had higher presence and concentrations of Stansbury cliffrose in their diet compared to post-parturient and females that lost a juvenile to mortality (Table 6). Additionally, females that lost a juvenile to mortality had lower presence and concentrations of Stansbury cliffrose in their diet than females that were provisioning young (Table 6). Saxifragaceae only appeared in diets of females that were provisioning young or had lost a juvenile (Table 6). The frequency of occurrence of buckwheat was higher in females prior to parturition and greatly reduced following parturition (Table 6).

We found that juveniles had higher diet quality and diversity than adult females during the spring months. Those results were not surprising, given that juveniles were consuming milk in addition to vegetation, which increases nitrogen content in feces (Jenkins et al., 2001; Taillon et al., 2013). During spring months, juveniles had a wider credible interval for diversity of diets than adult females, which is likely related to lower sample sizes of juveniles. Juveniles selected plant species that had a high prevalence in the spring diets of adult females, which is consistent with habitat selection being a learned behavior (Nielsen et al., 2013; Larue et al., 2018). Nevertheless, juveniles did not sample plant species to the degree that adult females did, likely because vegetation is not the primary source of nutrition during their first 2 months of life, allowing juveniles to be more selective. That relationship may possibly be a result of how quickly juveniles learn to sample plants rather than an innate understanding of the highest quality or more palatable species.

Diet quality, diversity, and composition were dependent upon season and year for adult females. Our results indicated that fecal nitrogen was higher during spring and summer than winter months, which consistently had the lowest fecal nitrogen. Diet quality also regressed each year of the study, which may be related to drought conditions, although 2017 was the wettest year (PDSI: x¯ = 3.5, SD = 0.7) of our study, while 2018 was the driest (PDSI: x¯ = −1.66, SD = 0.42). Furthermore, spring and summer months had higher diet diversity than winter months, although the diet diversity was similar between summer and winter in 2018. Spring and early summer months coincide with vegetation green-up (Snyder et al., 2016); therefore, it is not surprising that females had higher fecal nitrogen and diet diversity during this period. This result is further supported by our diet composition data, because forbs became more prominent in diets during the spring, and females consumed more species of forbs during this period. In the Great Basin, many plant species senesce as precipitation decreases and temperatures rise, resulting in a reduction in protein and digestible energy available for herbivores during summer and winter months (Beck and Peek, 2005b). Many annual and perennial forbs are not available to herbivores year-round, resulting in individuals adjusting selection to species that are available during the summer and winter months (Krausman et al., 1989; Miller and Gaud, 1989; Bleich et al., 1997). Parturition of bighorn sheep at Lone Mountain occurred during spring months, March through May (Blum, 2021), when forage availability and quality were highest. The third trimester of pregnancy and lactation are the most expensive periods for female ungulates in terms of requirements for protein and digestible energy (Barboza et al., 2009). Therefore, our results are consistent with the notion that timing of parturition is likely the result of selection for females to give birth during the most nutritionally beneficial time period in our study system. Tradeoffs associated with behaviors that are risk-averse or risk-prone relative to risk of predation are associated with montane ungulates around parturition periods (Festa-Bianchet, 1988; Berger, 1991; Bleich et al., 1997; Rachlow and Bowyer, 1998; Barten et al., 2001; Poole et al., 2007; Long et al., 2009). Previous research has shown females use habitat types that are linked to survival of juveniles, but potentially have reduced quality or availability of forage (Festa-Bianchet, 1988; Berger, 1991; Bleich et al., 1997; Rachlow and Bowyer, 1998; Blum, 2021). Nevertheless, females may not have to make those tradeoffs in some ecosystems, and it is more evolutionarily advantageous to select areas that provide high quality and abundant vegetation, in addition to safety for offspring (Pierce et al., 2004; Heffelfinger et al., 2020).

We found mixed support for our hypothesis that individuals without dependent young would have more diverse diets than those provisioning offspring. Females that lost offspring had a 64% likelihood of having a more diverse diet than those females that were provisioning offspring, lending some support to our hypothesis. Contrary to our hypothesis, however, pre-parturient females had an 86% likelihood of having a less diverse diet than those females that were provisioning offspring. Those results may be a function of selection for the highest quality plants, or because of fewer species of plants on the landscape when most females were not provisioning young, given that forbs and annual grasses became more prevalent as the spring months progressed (Nicholson et al., 2006). Our results may indicate that females selected some forages at a higher rate, therefore reducing diet diversity, which may also be linked to availability. Females may also be selecting forage to maximize overall intake by selecting for species that offer larger bite masses, such as Stansbury cliffrose, which may further reduce diet diversity. The differences in diet diversity between females that had lost offspring and those provisioning offspring likely indicated a tradeoff because vegetation availability was similar in those time periods, but females that lost offspring had more diverse diets than females provisioning young. An increase in diet diversity may allow females to maximize quality of their diet and more efficiently replace somatic reserves lost during gestation and lactation. However, this relationship may not occur if individuals selected strongly for vegetation with the highest nutrient content, which could result in a decrease in diversity. Further research has shown that females adjust habitat selection when provisioning offspring to areas that are more beneficial to survival of young rather than nutritional availability (Bleich et al., 1997; Blum, 2021). We suspect that females that were not constrained to steeper, rugged habitats had more diverse diets because they were not limited by the presence of dependent young. While the loss of a neonate has implications for fitness, females can increase the quality of future offspring by increasing their nutritional condition rapidly after the loss of their offspring (Monteith et al., 2013; Morano et al., 2013; Monteith K. L. et al., 2014). This behavior may lead to higher survival rates of offspring the following year and increase the likelihood of recruiting young.

Collecting fecal samples from specific individuals proved challenging in our study area. Given that we rarely were able to match a fecal pile with a specific radio-collared individual, we collected as many piles in proximity to morning GPS locations of specific females as possible. We analyzed multiple pellets from each group of samples to get a better idea of diet quality and composition within the group, as well as to increase the probability of analyzing the sample from the target individual. If our samples were not representative of specific parturition periods, we would expect no clear trends in diet quality or composition across females prior to parturition, while provisioning offspring, and following a loss of offspring, which was not the case. Additionally, female bighorn sheep typically form groups comprised of other individuals exhibiting similar provisioning status and that, in turn, influenced habitat selection (Blum, 2021). Individuals within a group should also be exposed to the same vegetation types when foraging. Thus, our results likely reflect the diet selection of females provisioning and not provisioning offspring in our study area.

Previous research indicates that lactation status influences the absorption rates for nitrogen in ungulates (Monteith K. B. et al., 2014). Those results indicate that given similar diets, lactating females will absorb greater amounts of nitrogen than non-lactating females and skew the nitrogen percent in feces (Monteith K. B. et al., 2014). Therefore, it is possible that our post parturition levels of fecal nitrogen are negatively skewed and that post parturient females have similar diet quality as females without dependent offspring. However, non-lactating females, especially those that were not pregnant, were commonly found in areas where male bighorn sheep occurred within the study area, which were segregated from parturient females (Blum, personal observation). Bleich et al. (1997) also showed that males used areas with higher quality vegetation during these periods, therefore, it is likely that non-lactating females were accessing higher quality forage than those females with dependent offspring. This was also supported by some of our diet composition results, which indicated that females that lost a neonate had higher frequency of occurrence and relative read abundance of some forbs and grasses in their diet. Therefore, we are confident that our results that indicate an apparent diet quality tradeoff are representative of the diet quality differences we observed between provisioning status.

We were unable to measure metabolizable energy content of fecal samples in our study; although, previous research has indicated that metabolizable and digestible energy, not digestible protein, was the limiting factor for ungulates during lactation (Holt et al., 1992; Parker et al., 1999). Energy and protein requirements increase substantially during lactation and are essential to the growth of offspring (Barboza et al., 2009; Parker et al., 2009). Although we cannot speak to the variation of metabolizable energy in the diet of bighorn sheep across seasons and parturition stages, similar relationships exist because protein and energy measurements are often correlated in ungulate diets throughout the year (Holt et al., 1992; Parker et al., 1999, 2009; Rad et al., 2015). The apparent tradeoffs of nutritional quality for safety of offspring are likely linked to metabolizable energy because females select areas with lower forage availability while provisioning offspring (Blum, 2021).

Our results indicated a high presence of single-leaf pinyon in bighorn sheep diets on Lone Mountain. We suspect that result is a combination of high concentrations of pollen across the landscape during spring months and low digestibility of some vegetation species (Ando et al., 2018). Given that different functional groups and plant taxa differ in digestibility, some species likely show higher presence in feces (Scasta et al., 2019). This bias is not unique to diet analyses using DNA but is often reported in microhistological analyses of diet composition (Anthony and Smith, 1974; Dearden et al., 1975). We did observe bighorn sheep foraging on single-leaf pinyon occasionally, but those events were rare and leads us to suspect that the high concentrations in bighorn diets is likely a result of pollen falling onto consumed vegetation (Minckley et al., 2008) and combined with low digestibility of pinyon pine. While Blum (2021) showed that female bighorn sheep avoided areas with a lot of trees, pinyon pine was present throughout the study area, albeit in low densities, therefore it is likely that pollen was widely distributed throughout the study area. DNA metabarcoding methods are an evolving technology that need further research and development to understand how estimated proportional diet composition represents actual dry matter intake. Additionally, primer biases may result in different amplification rates across forage species and result in courser taxa identification (Browett et al., 2021; Stapleton et al., 2022). To reduce the potential biases of DNA metabarcoding analyses, we used a combination of frequency of occurrence and relative read abundance. While relative read abundance measurements may be subject to higher rates of recovery bias, they tend to provide more accurate portrayals of population-level diets (Deagle et al., 2019). Additionally, frequency of occurrence measurements are generally less affected by recovery biases and are highly consistent in comparison to measurements of relative read abundance (Deagle et al., 2019). Using those measurements in conjunction provides a more accurate depiction of diet composition at the population-level and is encouraged to consider all sources of bias (Deagle et al., 2019).

To our knowledge, there are no other studies that have investigated the variation in diet quality and composition in free-ranging bighorn sheep associated with times around parturition, although many studies have addressed variation in diet quality and content across seasons or between adults and juveniles (Shank, 1982; Festa-Bianchet, 1988; Rominger et al., 1988; Krausman et al., 1989; Miller and Gaud, 1989; Irwin et al., 1993; Bleich et al., 1997; Denryter et al., 2020). Availability of forage species across most of those studies was highly variable and did not match the vegetation community available on Lone Mountain. Bighorn sheep rely on grass and shrubs as forage throughout much of their range (Bleich et al., 1997), which was also demonstrated in our study throughout the year (Table 5). Furthermore, most studies showed an increase in forbs in diets during spring and summer months, when they are more available and usually higher quality than at other times of the year (Krausman et al., 1989; Miller and Gaud, 1989; Wagner and Peek, 2006). Nevertheless, vegetation abundance and quality in other desert regions may not fall within these time periods (McKee et al., 2015). We also observed higher concentrations of forbs in diets of bighorn sheep during the spring and summer months, while availability of those forages was substantially reduced during the winter. Shrub concentration in the diets of bighorn sheep on Lone Mountain increased in winter months, similar to results in other research (Krausman et al., 1989; Irwin et al., 1993; Bleich et al., 1997). Other studies have shown strong similarities to diets of adult females and juveniles (Miller and Gaud, 1989; McKinney and Smith, 2007). We also showed a strong similarity between adult female diets and those of their offspring; our results, however, indicated higher degrees of selection for specific species by nursing juveniles, while diet richness of adults was greater.

Understanding diet characteristics are essential to the conservation of wild ungulates. Managing for ecosystems containing vegetation that meets the dietary needs of target species increases the health of populations. Biologists can use this information to guide habitat management and increase the nutritional availability across the landscape. For instance, our findings indicate that biologists should focus reseeding efforts on Stansbury cliffrose, buckwheat, species in the Saxifragaceae family, and native grasses when attempting to increase vegetation for desert bighorn sheep in this region. These species were among the most common in the diets of bighorn sheep following parturition and increasing their availability on the landscape will benefit females provisioning young. Understanding diet characteristics are also useful for identifying translocation sites for bighorn sheep. To increase the likelihood of a translocated population succeeding, biologists should select mountain ranges with similar vegetation types and availability as the source population. Furthermore, ensuring that those target mountain ranges have vegetation types that have the necessary nutrients to meet the metabolic needs of females is essential to the success of a population.

Our study provides further insight into the dietary relationships between female and juvenile bighorn sheep and their environments. A primary focus of nutritional ecology is the link between diet and the environment and gaining an understanding of how diet selection influences life history strategies and demography. We demonstrated tradeoffs between forage quality for females of different provisioning status, which lends further support to tradeoffs associated with risk-averse and risk-prone use of habitats, and forages that are commonly associated with montane ungulates. Selection for high quality forage, which often coincides with areas of greater risk of predation, when females are not provisioning young allows females to increase their somatic reserves or increase the nutrients directed towards subsequent reproduction. This relationship provides further insight into the habitat selection strategies of many montane ungulates around parturition, and how those strategies influence reproductive fitness of individuals.

The datasets generated for this study are available on request to the corresponding author.

The animal study was reviewed and approved by University of Nevada, Reno IACUC Committee (Protocol #00651).

MB, KMS, and MC conceived and designed the study with input from BW and VB. MB analyzed the data with input from KTS and KMS. MC, JB, and BW organized captures and aided in data collection. MC, KMS, and MB obtained funding with help from BW and VB. BS provided equipment and guidance for analysis of fecal samples for C and N. All authors contributed to writing and editing of the manuscript.

This work would not be possible without funding from the Nevada Department of Wildlife, The Fraternity of Desert Bighorn, Dallas Safari Club, Nevada Bighorns Unlimited, Wild Sheep Foundation, Wildlife Heritage Trust, Nevada Wildlife Record Book, the Desert Bighorn Council, and Nevada Muleys.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.