Resource selection of a montane endemic: Sex-specific differences in white-bellied voles (Microtus longicaudus leucophaeus).

Resources that an individual selects contrasted against what is available can provide valuable information regarding species-specific behavior and ecological relationships. Small mammals represent excellent study organisms to assess such relationships. Isolated populations that exist on the edge of a species' distribution often exhibit behavioral adaptations to the extremes experienced by a species and can provide meaningful insight into the resource requirements of the species. We deployed radio transmitters in a peripheral population of the long-tailed vole (Microtus longicaudus) during the mating season. We developed models of resource selection at multiple scales (within home range and patch). We found voles generally selected areas close to water and roads and consisting of high understory vegetation primarily composed of grasses. Resource selection varied between sexes suggesting different resource needs during the breeding season. The differential resource needs of voles might be a result of the energetic requirements for reproduction and are representative of a promiscuous or polygynous mating system.

Introduction

Patterns of animal spatial distribution can have profound implications for conservation and management of species and their habitat [1]. Understanding patterns of resource selection can provide fundamental information about species ecology and their resource requirements that can inform current ecological knowledge and management strategies [1-4]. Differential habitat selection by individuals for the highest quality areas available can be observed in individual home ranges [5,6]. An individual’s home range is often defined by multiple abiotic and biotic features of the environment and can shift based on the dynamic characteristics of these features and an individual’s physiological requirements [7-9]. Which resources an individual selects within its home range can therefore provide valuable information about species-specific behavior and ecological relationships [9,10].

Resource selection functions are the most common way that resource use is evaluated from individual location data [2,11]. Resource selection functions are fit in a use–availability framework, whereby environmental covariates (e.g., elevation, distance to water) at the site the animal was located (the used locations) are contrasted with covariates at random locations taken from a realistically available area for selection (the available sample; [2,11]). Evaluating space use at multiple spatial scales provides detailed characterizations of species habitat profiles and informs management practices [2,12-14]. Resource selection can vary based on sex due to different strategies employed by either sex depending on mating system [15,16]. During mating seasons, marked differences exist in the movement and resource selection between sexes, particularly in promiscuous and polygynous species [17,18]; in these systems, individuals allocate limited resources to reproduction. In promiscuous and polygynous mating systems, males optimize their reproductive efforts by copulating with as many females as possible, whereas females maximize their reproductive efforts by obtaining and converting food into offspring [16]. Differential resource selection based on sex is driven by differing resource requirements determined by mating systems or strategies [15,19]. Identifying the resource needs at the individual level will better inform management decisions to preserve vulnerable populations.

Small mammals are an essential component of many ecosystems by providing vital ecosystem services [20,21]. Small mammals consume invertebrates, vegetation, and seeds, potentially aiding local plant communities by controlling damaging insect populations and influencing seed distribution [20,22]. Many predators specialize on small mammals in their diet [20,21]. Some fossorial and semi-fossorial small mammals, such as prairie dogs (Cynomys spp.) and kangaroo rats (Dipodomys spp.), are keystone species that fulfill a crucial role in an ecosystem’s maintenance and health through bioturbation and habitat alterations [20,23,24]. Understanding the biology and ecology of small mammal populations is vital to inform conservation and management of ecosystems.

The long-tailed vole (Microtus longicaudus) is a small semi-fossorial rodent and habitat generalist [25]. The long-tailed vole boasts one of the largest latitudinal distribution ranges of all vole species in North America, stretching from Alaska to Arizona [25,26]. The importance of rear-edge populations, that is the populations that exist in the lower latitudes of widely distributed species, are often undervalued [27]. These edge populations can be a model that presents a focused view of how core populations may react and adapt to further species range contraction and expansion [28,29]. Like all microtines, long-tailed voles are herbivores with diets primarily consisting of leaves, stems, and roots of herbaceous plants [30]. The mating season for this widely distributed vole varies throughout the species range; populations at lower latitudes have an extended mating season from May–October with most reproductive activity in June–July [25]. Microtus presents a spectrum of mating systems from promiscuous (M. pennsylvanicus), polygynous (M. xanthognathus), and sometimes monogamous (M. ochrogaster); some species display one, two, or all three systems depending on the social environment and resource availability [17]. The mating system of long-tailed voles is not well understood, but likely exists within the spectrum displayed by congeners. The white-bellied long-tailed vole (M. l. leucophaeus; hereafter referred to as “white-bellied vole”), a sub-species of the long-tailed vole, is endemic to the Pinaleño Mountains in southeastern Arizona [31]. This is the southernmost population of M. longicaudus and lacks informative research. Unlike many other populations of long-tailed voles, M. l. leucophaeus does not co-occur with other vole species, making this population particularly unique [31]. Whereas most other species in the genus inhabit areas dominated by grassy cover, the long-tailed vole is found in a variety of habitats ranging from typical grassy areas to sparsely vegetated or woody shrub areas [25,32]. Long-tailed voles are often less aggressive than other vole species and, as a result, are relegated to less favorable habitat [25,32]. The Pinaleño Mountains population presents a unique opportunity to study resource selection in the absence of other vole species that may relegate long-tailed voles to sub-optimal habitat [31]. Furthermore, white-bellied voles are a Species of Greatest Conservation Concern in the State of Arizona [33].

In this study, we examined how white-bellied voles use the landscape available in the absence of congeneric competition at multiple spatial scales. We modeled resource selection within the home range (3rd order; [34]) and for specific fine scale resources (4th order; [34]). Our objectives were to model fine scale patterns of resource selection for a peripheral population of long-tailed voles and examine differences between the sexes. We predicted that white-bellied voles will select areas with herbaceous grassy cover over areas dominated by woody vegetation. We predicted that males and females would show different patterns of resource selection across spatial scales due to differing energetic needs (i.e. proximity to water or forage) and reproductive strategies (i.e. nest sites).

Materials and methods

Study area

The Pinaleño Mountains (32.7016°N, -109.8718°E) are a portion of the northern extent of the Madrean Sky Island Complex of the southwestern United States [35]. At 3,269 m the Pinaleño Mountains have the highest peak in the complex and encompass an area of approximately 780 km2. The Pinaleño Mountains have diverse vegetation communities as a result of the 2,367 m elevational gradient [31]. Forest landscapes in the Pinaleño Mountains have been fragmented due to roads, insect outbreaks, and large-scale fires. These high levels of disturbance resulted in habitat that is extremely patchy and poorly connected [36]. Our study area was located in the upper elevations of the range, consisting of high mountain meadows and mixed conifer forests (2,870–3,050 m) of Douglas-fir (Pseudotsuga menziesii), ponderosa pine (Pinus ponderosa), southwestern white pine (Pinus strobiformis), Engelmann spruce, and corkbark fir (Abies lasiocarpa var. arizonica) [37]. We conducted our study in 9.1 ha of meadow and mixed conifer forest.

Sampling design

We trapped small mammals during the summers of 2018 (May–August) and 2019 (May–August) over multiple sessions. We trapped areas that were > 100 m apart and between 50 and 500 m from roads to avoid any negative road impacts, such as avoidance or mortality [38]. We used single door folding Sherman traps (7.62 x 8.89 x 22.86 cm; H.B. Sherman Traps, Inc. Tallahassee FL) baited with a mixture of peanut butter and oats.

To determine areas where we could reliably trap white-bellied voles we conducted preliminary trapping in 2018 based on historical locations and information from previous studies [31,39,40]. For each trapping session, we placed two transects that were open both day and night due to the lack of clarity on activity periods of long-tailed voles (diurnal [31]; nocturnal [25]). Our transects were 240 m long and located parallel to Soldier Creek and an un-named creek near Coronado National Forest access road 4567. Along each transect, we placed a pair of traps every 10 m, one on either side of the watercourse, for a total of 50 traps. We placed each pair of traps in dry locations 1–5 m from the watercourse depending on saturation of the soil, as some areas were diffused and bog-like. In 2018, we opened traps at dusk and checked them approximately 30 min after dawn. After the morning check, we left the traps open, and checked and closed them at 1000 h to avoid heat related mortalities. We reopened traps in the afternoon and checked the traps again at dusk. We discontinued nocturnal trapping after 2018 as diurnal trapping proved to be more successful. In 2019, we placed transects where voles were caught the previous year. Additionally, we opportunistically trapped in areas where we observed voles in 2019.

Animal handling

We processed and released animals immediately at the location of capture. We characterized age class (juvenile or adult), sex, and reproductive condition by visual inspection of testes (scrotal, abdominal) and teats (lactation) or vaginal condition (perforate, nonperforate). We recorded mass in g, and following standard marking methods, we used ear tags (1005–1, National Band and Tag Company, Newport, KY; Sikes et al. 2016 41) to mark individuals. We affixed very high frequency radio collars (SOM-2070, Wildlife Materials, Murphysboro, IL) to 31 adult white-bellied voles (body mass ≥ 30 g, mean ± SD = 56.2 ± 7.2 g) in 2019. Radio collars were < 10% of each individual’s body mass (mean = 3.09 ± 0.09 g) to minimize effects on daily activity and behavior [41,42].

Radio telemetry

We located animals five of the seven days per week, with an animal receiving several points throughout the day in May–September 2019. We obtained locations >1 h apart, distributed across daylight hours to ensure temporal independence of locations [43]. Animals were located by homing on individuals until one of the following occurred: 25 telemetry locations were achieved [44], the collar fell off, the vole was predated, or the collar’s signal could not be located. We used a handheld global positioning system device and used the point averaging function for ≥ 5 minutes to record the spatial location of all animal points.

All field work was conducted in accordance with the American Society of Mammalogists guidelines [41] and approved by the University of Arizona’s Institutional Animal Care and Use Committee (IACUC protocol #16–169) under permit from Arizona Game and Fish Department (Permit # SP651773).

Vegetation sampling

Upon locating a vole via telemetry, we marked the location with a pin-flag. To minimize the effect of our presence on vole movement we waited one hour to collect vegetation data at the location, given the vole had moved away from that location. We collected vegetation data at two points: the vole’s known location and at a paired, randomly generated location [14,45]. Random locations were placed 9.8 m away from the vole’s known location, which we based on the average hourly movement rate of other vole species [42,46-48]. We used a random number generator to represent eight intermediate and cardinal directions moving clockwise. At all known and random locations, we recorded understory cover, canopy cover, vegetation composition (categories: bare ground, coarse woody debris, grass, forb, fern, log, sedge, stump, rock, rush, shrub, tree, water). We used a 2.5 cm x 100 cm cover pole marked in 2 cm increments to measure understory cover at the center of each location [49]. We recorded the height of any obscuring vegetation from the four cardinal directions at a distance of 4 m and a height of 1 m, with any vegetation taller than 80 cm classified as 100% understory cover. We calculated percent understory cover by taking the average from all four measurements at each location, divided by the total height of the cover pole [49]. To calculate canopy cover we followed the standard equation for a convex spherical densiometer and applied the correction factor for the 17-dot variation for each point [50]. We used a 1 m2 quadrat centered at each location to characterize vegetation composition and percent cover through visual inspection [51]. For any woody species within the 1 m2 quadrat, we used diameter at breast height (DBH) to categorize live and dead woody species as either shrubs (woody plants < 10 cm DBH) or trees (≥ 10 cm DBH [52]).

Data analysis

We calculated 95% minimum convex polygons using the “adehabitatHR” package in R [53,54], and visualized estimates with ArcGIS Pro v2.4.1 [55] for all individuals that had at least five telemetry locations, the minimum number of locations needed by the “adehabitatHR” package to create an individual minimum convex polygon home range. The number of locations per individual ranged 5–25 (mean = 18).We pooled all animals into three groups: all voles, males, and females; pooling data across individuals while still accounting for individuals variation is ideal for low sample sizes [56]. In total, we used 493 known locations, 325 from 19 females and 168 from nine males. There was heterogeneity in the number of locations per individual, so to define availability at the within home range scale (3rd order), we generated random points at a 2:1 ratio, within each individual’s home range, to ensure availability was unique to each individual [56]. For all animal locations and random points, we extracted normalized difference vegetation index (NDVI) values, distance to roads, and distance to water in ArcGIS Pro. The remotely sensed imagery we used was taken in late summer 2017 and is the most recent and highest resolution (30 cm) available [57]; we overlaid this imagery with road and waterway layers and hand digitized where needed.

We used the paired random locations (9.8 m from known locations) taken in the field to define availability for patch scale (4th order) selection. We modeled selection based on habitat features at known vole locations compared to the corresponding random location, where we measured habitat characteristics simultaneously to remove effects of differing availability by weather and time of day [58].

We quantified resource selection of voles at multiple scales to assess habitat selection and identify key environmental characteristics. We standardized all covariates prior to running models. We fit generalized linear mixed-effects logistic regression models with individual as a random effect and a binomial use vs. availability design, with the lme4 package in R for within home range scale selection. To reduce bias based on unequal known locations, we used a random intercept term assigned to each individual [56,59]. At the patch scale, to compare each vole location with its random location, we used conditional mixed-effects logistic regression models using the mclogit function in R with a binomial error structure and logit link function [2,11,56]. We tested sets of a priori models at both scales based on previous research of habitat selection of long-tailed voles [25,31]. We tested models at both scales for three groups: all voles, males, and females (within home range: 2 models; patch scale: 5 models). We evaluated model support using Akaike’s Information Criterion adjusted for small sample sizes (AICc), and all models ≤ 2 AICc units of the top model were considered to be competing models [60,61].

Results

Captures

We logged 4,715 trap nights (3,300 in 2018 and 1,415 in 2019) and captured 194 individual voles in total: 45 unique voles in 2018 (17 males, 20 females, 8 juveniles) and 149 unique voles in 2019 (31 males, 55 females, 63 juveniles). In 2019, we collared and tracked 31 adult voles (≥30 g; 12 males, 19 females). To collect animal locations via radio telemetry, we logged approximately 558 person hours. More than half (58%; 7 males, 11 females) of these individuals were either lost from the study due to predation (4 males, 1 female), unknown cause but confirmed mortality (2 females), collar removal (1 male, 8 females), or no signal/collar malfunction (2 males); on average, loss of a vole (i.e. no longer able to have data collected) would occur 29.83 ± 20.69 (SD) days after receipt of collar.

Within home range selection

At the within home range scale, our top model for each group indicated negative selection (i.e. avoidance) by white-bellied voles for high NDVI values and areas farther from roads (Fig 1). For the all voles (sexes combined) group, our top model included all covariates: NDVI, distance to water, and distance to roads (Table 1). The top model for females included NDVI, distance to roads, and distance to water and the top model for males included NDVI and distance to roads (Table 1). For all voles and females, top models indicated selection for areas farther from water. The top model for males excluded distance to water and indicate avoidance of high NDVI values and areas farther from roads. However, our three groups each had two competing models that were within 2 ΔAIC units (Table 1).

Fig 1

Home range scale beta coefficients.

Habitat selection patterns of white-bellied voles (M. l. leucophaeus) from the Pinaleño Mountains in southeastern Arizona, USA, summer 2019 represented by beta coefficients of variables explaining variation in habitat selection patterns of our top 3rd order generalized linear mixed-effects logistic regression model for a) all voles, b) Female, and c) Male voles. The x-axis depicts standardized regression coefficients, which provide an index of the strength of the linear relationship for explaining habitat selection patterns. The y-axis contains all the covariates included. The dotted line at zero represents the division between selection (right of line) and avoidance (left side of line). The coefficient estimates are represented as dots and their 95% confidence intervals as whiskers.

Table 1

Within home range scale (3rd order) a priori generalized linear mixed-effects logistic regression models.

Group	Covariates	AIC	ΔAIC
All voles	NDVI+Distance to roads+Distance to water	1840.3	0.0
	NDVI+Distance to roads	1840.8	0.5
Females	NDVI+Distance to roads+Distance to water	1239.5	0.0
	NDVI+Distance to roads	1239.7	0.2
Males	NDVI+Distance to roads	606.0	0.0
	NDVI+Distance to roads+Distance to water	607.1	1.1

NDVI, normalized difference vegetation index; AIC, Akaike information criterion.

ΔAIC is the difference in AIC values between each model and the lowest AIC model.

Patch level selection

For patch scale, the all voles, female, and male groups had the same top model which included: understory cover, canopy cover, bare ground, grass, forb, logs, coarse woody debris, distance to water, and distance to roads (S1 Table). The beta coefficients, for the all voles top model, indicated positive selection for high understory cover, canopy cover, grass, forb, logs, and coarse woody debris and avoidance of bare ground, distance to water, and distance to roads. White-bellied voles selected for areas with high understory cover and coarse woody debris (Fig 2). The white-bellied voles selected areas close to roads, however, we never documented voles crossing roads and only 2% of recorded locations were ≤ 10 m of roads. Of the three individuals that we documented ≤ 10 m from roads, none moved closer than 4 m to a road. Our model for female voles had positive beta coefficients for understory cover, canopy cover, logs, and coarse woody debris but negative beta coefficients for bare ground, grass, forb, distance to roads, and distance to water. Females strongly selected for areas with high understory cover and avoided areas of bare ground (Fig 2). We had a competing top model for females that included Grassy Cover as a covariate (ΔAIC = 0.6). Our male model indicates positive selection for all covariates except bare ground. Males strongly selected for areas with high log and coarse woody debris cover (Fig 2).

Fig 2

Patch scale beta coefficients.

Habitat selection patterns of white-bellied voles (M. l. leucophaeus) from the Pinaleño Mountains in southeastern Arizona, USA, summer 2019 represented by beta coefficients of variables explaining variation in habitat selection patterns of our top 4th order conditional mixed-effects logistic regression model for a) All voles, b) Female, and c) Male voles. The x-axis depicts standardized regression coefficients, which provide an index of the strength of the linear relationship for explaining habitat selection patterns. The y-axis contains all the covariates included; coarse woody debris is shortened to CWD. The dotted line at zero represents the division between selection (right of line) and avoidance (left side of line). The coefficient estimates are represented as dots and their 95% confidence intervals as whiskers.

Discussion

Resource selection is similar between the sexes of white-bellied voles. At the within home range scale, both sexes avoided areas with higher NDVI values corresponding to more heavily wooded areas. This indicates selection for more open areas with low tree cover. Our study area is a mosaic of mostly closed canopy forest and open grassy meadows as well as areas of patchy tree canopy cover with an understory consisting of bare ground, forbs, ferns, and grasses. Our data contrast with some previous studies of long-tailed vole habitat selection for wooded or shrubby areas and may be attributed to the lack of congeneric competition [25,32,62].

Both sexes selected areas relatively close to roads, however, we did not document any individuals that crossed a road. Several other small mammals avoid road surfaces due to the perceived threat of predation from the lack of cover [38]. Other highly mobile mammals such as squirrels exhibit clear avoidance of roads and rarely cross them [63]. Road type, size, and traffic volume are all factors that impact animal behavior and perceived risk [64]. The roads in our study area are hard compact, two-lane roads and are likely a barrier to vole movement. Selection for areas close to roads may be an artifact of roads occurring near flat open meadows where they are easier to construct [65] and voles selecting for roadside habitats due to an abundance of vegetation [66].

We found differences between the sexes in selection, at the within home range scale, of distance to water; the top model for females included distance to water and the top model for males did not. In mammals, food and water are common limitations for female fitness due to the increased energetic costs of pregnancy and lactation, whereas male fitness is limited by access to mates [15,67,68]. Vegetation community composition and predator avoidance are likely more important to males and resource selection may not be limited by water availability. Dependence on surface water by long-tailed voles has not been verified; some studies found surface water to be non-essential to long-tailed vole diets [25,69] whereas Findley et al. [70] found surface water is required for daily survival. It is possible that dependence of long-tailed voles on water is dictated by the type and quality of forage an individual consumes.

Despite selecting for similar areas overall, males and females displayed variability in selection patterns at the patch scale. Our All voles model indicates white-bellied voles avoid areas of bare ground and areas far from roads and water. The All voles model displays varying degrees of positive selection for all other covariates; logs, coarse woody debris, and understory cover were most heavily selected. White-bellied voles select areas that have high amounts of cover with the cover types downed logs and coarse woody debris being the most highly selected for. This aligns with previous research that has found a positive correlation between number of logs and high vole densities [32]. Where our findings differ from previous research is the association with areas of sparse herbaceous growth. Long-tailed voles are known to use areas consisting of primarily woody vegetation in the presence of other vole species [32,62]. White-bellied voles strongly avoided bare ground and selected areas with high herbaceous understory cover in our study, which is consistent with findings from vole removal and exclusion experiments [62]. Our empirical evidence further strengthens the conclusion that long-tailed voles inhabit and flourish outside of the ‘typical’ habitat dominated by woody vegetation [62]. In the absence of competing vole species, white-bellied voles can select highly productive herbaceous areas for forage, without impediment, while still staying close to areas with logs and coarse woody debris for nest sites and predator evasion.

In promiscuous and polygynous systems, females tend to spend more time in creation and maintenance of burrows whereas males travel from female to female and attempt to mate and defend their home range from other males [17]. When we evaluated resource selection by sex, the different priorities during the mating season (May–October: [25]) become apparent. Females avoid bare ground, areas far from roads, and areas far from water as well as areas with high percentage of grass and forbs. Females select areas with high understory cover consisting primarily of grass for forage and cover but strongly select areas suitable for nests with microhabitats consisting of logs and coarse woody debris near water. Females spend most of their time in suitable nest sites and only make short, infrequent trips into highly productive herbaceous areas to forage and gather nesting materials. Female voles require supplemental surface water sources outside of water obtained through herbaceous forage during times of high energetic requirements such as lactation [67].

In contrast, males avoid bare ground but selected areas farther from roads and water. Males selected a greater diversity of habitat characteristics, which suggests more movement within their home range. That males expend much of their time and energy to secure areas of abundant resources and therefore mating opportunities while excluding other males is indicative of resource defense polygyny [15]. M. ochrogaster and Myodes gapperi under non-drought conditions do not require supplemental water outside of water obtained from forage [67,71]. During the mating season, the highly mobile male white-bellied voles may avoid areas close to water because they consume forage with high moisture content that circumvents dependence on surface water as displayed by congeners [12,46]. An alternative explanation to males selecting areas far from water during the mating season is that males reduce water consumption as a tradeoff for more reproductive interactions similar to many promiscuous and polygynous ungulates during the breeding season [72,73]. Females are more sedentary and select areas near surface water to supplement their limited forage opportunities and increased energetic needs. Our findings of sex-based differential resource selection are consistent with previous research on other mammalian promiscuous and polygynous species [16,74].

Conclusions

White-bellied voles’ selection within home range for herbaceous vegetation as opposed to the woody habitat described for this species can help to inform future species management decisions. However, further research that incorporates additional variables at the within home range scale in conjunction with our results may be necessary. Given the potential importance of white-bellied voles in the Pinaleño Mountains, the ecosystem services they provide [23,25] and their imperiled conservation status in Arizona, it is crucial to understand the resource requirements and assess response to a changing climate to maintain this endemic population. By understanding the different patterns of resource selection for this subspecies in contrast to other populations will lead to better informed and more successful management decisions and illuminate key drivers of the species’ biology. Because environments inherently change through time, it is important to understand the foundation of what individuals require and how they behaviorally respond to such change. Resource selection functions allow us to quantify resource selection patterns by wildlife [59]. Identifying these patterns and how they change through time can provide crucial insights into underlying resource needs of wildlife populations to inform management and conservation.

Patch scale a priori 4th order conditional mixed-effects logistic regression models.

(DOCX)

Click here for additional data file.

25 Sep 2020

PONE-D-20-27373

Resource selection of a montane endemic: sex-specific differences in white-bellied voles (Microtus longicaudus leucophaeus)

PLOS ONE

Dear Dr. Dutt,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Nov 09 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Bi-Song Yue, Ph.D

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2.Thank you for stating the following in the Financial Disclosure section:

[NRD- Funding was provided by University of Arizona (http://www.arizona.edu/) and T&E Incorporated’s Grants for Conservation Research (www.tandeinc.com). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.].

We note that you received funding from a commercial source: [T&E Incorporated]

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

3. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Review PONE-D-20-27373

This manuscript presents results of an interesting study on habitat selection by an endemic subspecies of long-tailed vole. The authors used radio telemetry to determine used locations, which were compared to available locations. At the 3rd order scale they compared GIS variable, while at the 4th order scale they compared field collected vegetation data. Research on habitat selection of poorly understood species such as this are much needed. However, I found a number of problems with the presentation of the information that made it very difficult to understand the methods and results. For instance, the authors ostensibly investigated habitat selection at the “within home range” and “patch” scale (3rd and 4th order). However, they consistently referred to the 3rd order as home range selection, which is actually 2nd order selection per Johnson (1980). The methods do not say how the available locations were selected for this scale. The microhabitat data should probably also be considered 3rd order selection (usage of various habitat components within the home range. Fourth order would normally be for specific items such as den sites or specific food items. So, I think both scales analyzed are actually 3rd order.

Another problem is that they concluded that their results reflected fundamentally different habitat selection for long-tailed voles in the Pinaleno Mountains, in comparison with other populations where long-tailed voles live in presence of other competitively dominant vole species. There is no evidence for this conclusion. Importantly, they conflate their results at the 3rd or 4th order of selection with generalized statements about habitat at the 1st or 2nd order. You cannot extrapolate from one scale to another. Second, the habitat conditions described by their results are consistent with other places that long-tailed voles are known to occur, including places with other vole species. To make such a comparison they would need to have similar habitat selection results for another population of voles where other species also exist.

The discussion was very difficult to follow as it seemed to jump all over the place often not identifying what scale was being discussed. I had a very difficult time following the arguments in that section due to the writing problems. I suggest a more streamlined discussion that more concisely describes 3rd order habitat selection in this population voles (without conclusions about completion) and describes the differences in habitat selection between the sexes. Other points don’t seem very well justified to me.

38-42 There is a bit of redundancy in these sentences.

76 If you are talking about the geographic range, then you are discussing a feature of the species, not the individual. A species should be referred to with the Latin name or “the long-tailed vole”, which denotes the species. If you are talking about multiple individuals that actually exist out in nature then you can refer to long-tailed voles. It is not correct to use long-tailed voles to refer to the species.

82 Missing a period. Also, is there a reason to say monocots and dicots? What other herbaceous plants are there?

92 Populations in New Mexico are just as far south.

96 In the Southwest they seem to be found in relatively moist locations and are mainly riparian (e.g., Findley et al 1975), although they also occur in openings within mixed coniferous forest that allows some herbaceous cover (e.g., Lehmkuhl et al 2008 Northwest Science 82, Wampler et al. 2008 SWNAT 53, Sullivan and Sullivan 2018 Crop Protection 112).

97 Something wrong with the wording

100 missing hyphen in white-bellied.

107-109 But you don’t have a comparison with a situation where they occur with a congeneric competitor. You already made it clear that this population lacks another species of vole. I think you should delete “in the absence of congeneric competition”. I also don’t think your study design allows you to address this point. The most you can say is that your results reflect habitat selection at the third order in your study area. You don’t know how that would differ in presence of another species.

108-109 Can you state what you predict the difference will be?

114 According to the USGS place names database, Mount Graham is a specific summit of the Pinaleno Mountains. I suggest deleting Mt. Graham here unless one of your study sites is on Mount Graham, or work the name into the study area description another way.

129 What is meant by “negative road impacts”?

139 What do you mean by “depending on the saturation of the soil?

150 how did you get an accurate total length measurement on a live animal? Why were these measurements taken? What about mass?

157 A location once per day per animal?

159 The “collar was slipped” is wildlife slang; please rephrase. Can you rephrase to avoid word predated? How did you know they had been killed by a predator? Did you find them dead and partially consumed? The “Signal was lost” is slang.

159-160 It is still not clear to me if you recorded more than one location per animal each day.

157-161 How did you determine the location of the animal? Triangulation? Homing? Did you evaluate telemetry system error?

167 I don’t know how you located the vole. Did you see it? IF not, now can you be sure its location?

174 and 179 Is canopy cover the same as overstory cover?

174 I do not understand how the vegetation sampling was collected. Were all these variables collected at the location or on transects or a plot?

178 At what distance did you consider obscuring vegetation? It is not clear how you measured this.

180 Here you mention averaging 4 measurements but you did not explain how you took them.

181 Did you take the densitometer reading in the 4 cardinal directions?

183-184 centered at the location? How did you measure composition and percent cover?

184 What is the plot? Do you mean the 1 m square plot?

190 Why did you include animals with so few locations?

194 I think you mean within home range scale, at least that is what I think you said you were going to evaluate in the introduction.

194 where were the random points drawn from?

196 What date did NDVI come from and why?

197 What were the sorts of water sources in the remote sensed imagery. Often voles are associated with moist to wet spots that would not show up on imagery or maps. You trapped along creeks. Were they perennial?

205-206 I did not see mention of quadrates in the methods. Are these cover classes for percent cover?

211-220 To me it seems that both of your scales are really third order, selection of sites within the home range. Your patch scale seems like description of microhabitat. Your other scale seems like within home range (not home range), although I cannot determine where you drawing the available points from. If within home range, I think your available points should be drawn from the MCP of each individual.

214 within home range scale?

218-219 Where are the models? How many? How rationalized? The sentence implies the same models for each scale. Do you not expect different selection at different scales?

224-225 I assume the male and female numbers are adults. How did you assign age class? Did you collar juveniles? Did you record sex on juveniles?

228-229 clarify what you mean by unknown but confirmed mortality. I think you mean you know it was dead but don’t know why. Collar slipping is slang. I think you mean that the collar fell off the animal. What does missing mean? Do you mean that you were unable to find a signal?

230 What do you mean by loss of vole?

222 You used individual as random effect but I don’t know the sample size of male and females included in the analyses.

235 by group do you mean gender?

237-238 Avoid using “shows” and “showing”

249 I don’t think you mentioned standardizing these variables in the methods, although perhaps I missed it.

256 Ideally I’d like to see the PCA results perhaps in supplemental material.

255 So did you not use the PCA? I am confused by what is being presented in this section.

258 This sentence does not read correctly—too many “models”.

260-263 Why not say that the Beta coefficients indicated positive selection for … and avoidance for ….

Supplement Table 1. In looking at this table I am confused about your choice of variables and a priori hypotheses. Is the global model the top model? Are these all the models? I don’t think I understand why the PCI is used versus the other variables in the models? For instance, was log and grass also cover classes? I also see no rationale for these models.

287 what do you mean by they showed varying degrees of selection?

290-291 Long-tailed voles are commonly captured in wet meadows and other herbaceous systems even when other voles are present. I’m not sure you can make this conclusion. Your analysis was at the home range (or perhaps within home range) and patch scales, while the habitat data you are contrasting with pertains to landscape or macro scale. I do not think it is surprising that long-tailed voles select the herbaceous environments in their surroundings. The same is true elsewhere in their range. In reflecting on this and looking back at your study area and trapping transect locations I really cannot get a feel for the environment in which the study was done. Was a mosaic of forest and open areas?

297-298 There are a number of studies, mainly in the Midwest, that discuss the importance of roadside environments for voles and other species that select dense herbaceous vegetation. The idea is that these areas collect more moisture resulting in more lush vegetation.

286-308 long paragraph with lots of different topics

312-313 Selection for downed logs and coarse woody debris paints a different set of conditions than grassy areas.

315 long-tailed voles occur in relatively open areas. When they occur in forested ecosystems, their local distribution is in openings in the forest where there is higher herbaceous ground cover. Later, you also say that they select high herbaceous cover. You can only get high herbaceous cover if the sun is able to hit the ground, which does not happen in dense forests.

319-320 I don’t think you can say your results support the idea that they experience competitive exclusion from high quality vole habitat. You did not test that.

321-322. What is open habitat dominated by woody vegetation? I don’t understand what you mean by that.

323-324 This is a perfect description of the habitat of the long-tailed vole, but it holds true for almost any other mountain range where other species of voles also occur. I don’t think your description of habitat has to do with presence of the other species of vole. Otherwise, why would two species of vole so frequently co-occur together to the point that you catch them both on the same trap line?

You surveyed for voles along creeks in a generally forested area in the mountains. This is an ideal situation for long-tailed voles. It provides moist areas with herbaceous vegetation in openings and meadows within the forest. You can find long-tailed voles in any mountain range in this same conditions, regardless if there are other species.

338-339 It could also mean they select the location of home ranges based on different characteristics.

349 And higher water requirements during lactation

352-353 Again, I think you are confusing scales as I don’t see your results different from other locations where the species occurs.

354 There is a danger of extrapolating third and fourth order selection to higher order selection such as geographic distribution. You don’t know what first and second order selection are.

358 Are they imperiled? SGCN directs funding as is not

Reviewer #2: This is a very interesting paper that examines space use in a less-studied species of vole, The All Voles model showed that voles avoided areas of bare ground and and areas far from roads and water, Females followed the same pattern with the addition of areas with high understory cover. Females tend to remain in their terriroty near thier nest sites. Males avoided bare ground but chose areas furthre fron road s and water. The authros try to fir their data to the promiscusous/polygynous mating systems, I found this fit most tenuos as the authros really have no direct data of interatiions between and within the sexes and their statments about mating systmes comes mainly from a revew chapter by Jerry Wolff (1985).

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

20 Oct 2020

We appreciate the feed back from the Academic Editor and the reviewers on our manuscript. We have taken their comments into consideration and and responded to them in the document "Response to reviewers". We will also provide our responses here (below) to ensure that we are following procedure. Thank you for inviting us to resubmit this manuscript for potential publication.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2.Thank you for stating the following in the Financial Disclosure section:

[NRD- Funding was provided by University of Arizona (http://www.arizona.edu/) and T&E Incorporated’s Grants for Conservation Research (www.tandeinc.com). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.].

We note that you received funding from a commercial source: [T&E Incorporated]

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

3. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

________________________________________

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

• Reviewer #1: Review PONE-D-20-27373

This manuscript presents results of an interesting study on habitat selection by an endemic subspecies of long-tailed vole. The authors used radio telemetry to determine used locations, which were compared to available locations. At the 3rd order scale they compared GIS variable, while at the 4th order scale they compared field collected vegetation data. Research on habitat selection of poorly understood species such as this are much needed. However, I found a number of problems with the presentation of the information that made it very difficult to understand the methods and results. For instance, the authors ostensibly investigated habitat selection at the “within home range” and “patch” scale (3rd and 4th order). However, they consistently referred to the 3rd order as home range selection, which is actually 2nd order selection per Johnson (1980). The methods do not say how the available locations were selected for this scale. The microhabitat data should probably also be considered 3rd order selection (usage of various habitat components within the home range. Fourth order would normally be for specific items such as den sites or specific food items. So, I think both scales analyzed are actually 3rd order.

Another problem is that they concluded that their results reflected fundamentally different habitat selection for long-tailed voles in the Pinaleno Mountains, in comparison with other populations where long-tailed voles live in presence of other competitively dominant vole species. There is no evidence for this conclusion. Importantly, they conflate their results at the 3rd or 4th order of selection with generalized statements about habitat at the 1st or 2nd order. You cannot extrapolate from one scale to another. Second, the habitat conditions described by their results are consistent with other places that long-tailed voles are known to occur, including places with other vole species. To make such a comparison they would need to have similar habitat selection results for another population of voles where other species also exist.

The discussion was very difficult to follow as it seemed to jump all over the place often not identifying what scale was being discussed. I had a very difficult time following the arguments in that section due to the writing problems. I suggest a more streamlined discussion that more concisely describes 3rd order habitat selection in this population voles (without conclusions about completion) and describes the differences in habitat selection between the sexes. Other points don’t seem very well justified to me.

• We have made a number of changes in our discussion to hopefully address the concern of this reviewer and have added sections heading to provide further structure.

38-42 There is a bit of redundancy in these sentences.

Thank you for bringing attention to this, lines 39 through 42 were re-worded and combined to reduce redundancy.

76 If you are talking about the geographic range, then you are discussing a feature of the species, not the individual. A species should be referred to with the Latin name or “the long-tailed vole”, which denotes the species. If you are talking about multiple individuals that actually exist out in nature then you can refer to long-tailed voles. It is not correct to use long-tailed voles to refer to the species.

We appreciate this clarification, we have edited ‘long-tailed voles’ to read ‘The long-tailed vole’.

82 Missing a period. Also, is there a reason to say monocots and dicots? What other herbaceous plants are there?

Thank you for pointing these out, we added a period and replaced ‘monocots and dicots’ with herbaceous plants.

92 Populations in New Mexico are just as far south.

The mammalian species account for Microtus longicaudus states that this species occurs at elevations of 8000 ft and above (Smollen and Keller 1987). The southernmost mountain range in New Mexico that contains Microtus longicaudus are the Sacramento Mountains (Findley and Jones 1962) in New Mexico. The southernmost point that is above 8000 ft for each mountain are as follows:

o Pinaleño mountains AZ latitude: 32.6136

o Sacramento mountains NM latitude: 32.6343

The difference in latitude shows the New Mexico population is over 6 km north latitudinally than the Arizona population.

96 In the Southwest they seem to be found in relatively moist locations and are mainly riparian (e.g., Findley et al 1975), although they also occur in openings within mixed coniferous forest that allows some herbaceous cover (e.g., Lehmkuhl et al 2008 Northwest Science 82, Wampler et al. 2008 SWNAT 53, Sullivan and Sullivan 2018 Crop Protection 112).

Thank you for your comment on this. We restructured this sentence to reflect the variety of habitats in which long tailed voles are found. Also, ‘white-bellied vole’ was changed to ‘long-tailed vole’.

97 Something wrong with the wording

Thank you for bringing this to our attention, the sentence was re-worded.

100 missing hyphen in white-bellied.

Thank you for pointing out this typographical error we have added the hyphen in white-bellied.

107-109 But you don’t have a comparison with a situation where they occur with a congeneric competitor. You already made it clear that this population lacks another species of vole. I think you should delete “in the absence of congeneric competition”. I also don’t think your study design allows you to address this point. The most you can say is that your results reflect habitat selection at the third order in your study area. You don’t know how that would differ in presence of another species.

Thank you for this comment, we have removed “in the absence of congeneric competition” and reworded some of the sentence to better describe the habitat we intended.

108-109 Can you state what you predict the difference will be?

We appreciate bringing this point of clarification up. We have added some specific predictions to line 109.

114 According to the USGS place names database, Mount Graham is a specific summit of the Pinaleno Mountains. I suggest deleting Mt. Graham here unless one of your study sites is on Mount Graham, or work the name into the study area description another way.

Thank you for your concern on this, the name ‘Mt. Graham’ was deleted to avoid confusion.

129 What is meant by “negative road impacts”?

Examples of negative road impacts were included to improve clarity.

139 What do you mean by “depending on the saturation of the soil?

Certain sections of the stream were diffused and bog-like. In these areas the traps were placed further away from the main channel of the stream to ensure the traps were on dry ground. This information has been added to line 141 for reader clarification.

150 how did you get an accurate total length measurement on a live animal? Why were these measurements taken? What about mass?

We added mass to this sentence and removed data that was not presented in the manuscript.

157 A location once per day per animal?

Animals were located at least 5 days per week. Every animal was located once per day that we were in the field, with one animal receiving several points throughout the day spaced 1 hour apart. These additional details have been included in line 157 for reader clarification.

159 The “collar was slipped” is wildlife slang; please rephrase. Can you rephrase to avoid word predated? How did you know they had been killed by a predator? Did you find them dead and partially consumed? The “Signal was lost” is slang.

Thank you for catching this, we have removed all jargon from this section and re-worded where needed. We feel that keeping the phrase “the vole was predated” is a more concise message than rephrasing to something of the effect “the vole was killed by a predator”.

159-160 It is still not clear to me if you recorded more than one location per animal each day.

Every animal was located once per day that we were in the field (Monday-Friday), with one animal receiving several points throughout the day spaced 1 hour apart. These additional details have been included in line 157 for reader clarification.

157-161 How did you determine the location of the animal? Triangulation? Homing? Did you evaluate telemetry system error?

Animal locations were determined via homing as stated in line 158. We then recorded the animal location on a handheld GPS unit. We added clarification about GPS point averaging to address the issue of location error in line 163.

167 I don’t know how you located the vole. Did you see it? IF not, now can you be sure its location?

Prior to collar deployment we tested our ability to locate collars in various situations by placing collars in known areas (bare ground, in grass, in holes). We did this to familiarize ourselves with the equipment and ensure confidence in our locations. Visuals on voles were common but did not always happen. If a visual was not obtained this was noted on our data sheet. If the animal was tracked to the same location 3 times in a row with no visuals it was assumed that the collar had fallen off and a thorough search was conducted to recover the collar, if possible, and those 3 previous points were removed from the dataset.

174 and 179 Is canopy cover the same as overstory cover?

As far as we are able to tell these terms appear to be synonymous. Some authors use overstory to describe parts of the canopy (Wojtkowski 2008) and vis versa (Gold 2020). Some authors also use the combined term “canopy overstory” (Dellasala 2019).

1. Paul A. Wojtkowski, 4 - Agrobiodiversity, Editor(s): Paul A. Wojtkowski, Agroecological Economics, Academic Press, 2008, Pages 45-72.

2. Michael A. Gold, Anthromes—Temperate and Tropical Agroforestry, Editor(s): Michael I. Goldstein, Dominick A. DellaSala, Encyclopedia of the World's Biomes, Elsevier, 2020, Pages 107-116.

3. Dominick A. DellaSala, “Real” vs. “Fake” Forests: Why Tree Plantations Are Not Forests, Editor(s): Michael I. Goldstein, Dominick A. DellaSala, Encyclopedia of the World's Biomes, Elsevier, 2020, Pages 47-55.

174 I do not understand how the vegetation sampling was collected. Were all these variables collected at the location or on transects or a plot?

All variables were collected at all known vole locations and random points as stated in line 176 ” At all known and random locations, we recorded understory cover, canopy cover, vegetation composition (categories: bare ground, coarse woody debris, grass, forb, fern, log, sedge, stump, rock, rush, shrub, tree, water)”.

Understory cover-lines 183-188: “We used a 2.5 cm x 100 cm cover pole marked in 2 cm increments to measure understory cover at the center of each location [49]. We recorded the height of any obscuring vegetation from the four cardinal directions at a distance of 4 m and a height of 1 m, with any vegetation taller than 80 cm classified as 100% understory cover. We calculated percent understory cover by taking the average from all four measurements at each location, divided by the total height of the cover pole”

Canopy cover-lines 189-190: “To calculate canopy cover we followed the standard equation for a convex spherical densiometer and applied the correction factor for the 17-dot variation for each point”

Vegetation composition-lines 191-194: “We used a 1 m2 quadrat centered at each location to characterize vegetation composition and percent cover through visual inspection [51]. For any woody species within the 1 m2 quadrat, we used diameter at breast height (DBH) to categorize live and dead woody species as either shrubs (woody plants < 10 cm DBH) or trees (≥ 10 cm DBH [52])”

178 At what distance did you consider obscuring vegetation? It is not clear how you measured this.

We recorded the height of any obscuring vegetation from the four cardinal directions at a distance of 4 m and a height of 1 m. This clarification has been added into lines 180 and 181.

180 Here you mention averaging 4 measurements but you did not explain how you took them.

Thank you for bringing this to our attention, the explanation of how the 4 measurements were obtained have been edited for clarity on lines 180 and 181. The new sentence reads “We recorded the height of any obscuring vegetation from the four cardinal directions at a distance of 4 m and a height of 1 m, with any vegetation taller than 80 cm classified as 100% understory cover”.

181 Did you take the densitometer reading in the 4 cardinal directions?

Yes, we took densitometer readings in the four cardinal directions. The standard equation for a convex spherical densiometer using the 17-dot variation requires readings from the four cardinal directions (Strickler 1959).

183-184 centered at the location? How did you measure composition and percent cover?

Yes, we took these measurements centered on the location. These were assessed through visual inspection using a quadrat. These clarifications have been added, thank you for bringing this to our attention.

184 What is the plot? Do you mean the 1 m square plot?

Thank you for bringing this to our attention. Yes, we meant the 1 m square quadrat. We have edited the sentence to read ‘within the one m2 quadrat” to improve clarity.

190 Why did you include animals with so few locations?

Thank you for your concern on this. We evaluated three groups: All voles, males, and females. The number of locations an individual had was not a limiting factor for our models because all points were pooled into a larger dataset per group. Pooling data across individuals and then including a random effect for individual ID is ideal in low sample size situations while still controlling for individual variation (Gillies et al. 2006). This information has been added to like 200-202. At the 3rd order, we needed to constrain the available locations within an individual’s home range. The reason we included animals with at least 5 locations in these models, was because that was the minimum number of locations required by the adehabitatHR package in R.

194 I think you mean within home range scale, at least that is what I think you said you were going to evaluate in the introduction.

Thank you for identifying this issue. All references to ‘home range scale’ have been changed to ‘within home range scale’ to assist with clarity.

194 where were the random points drawn from?

All the random locations were generated with ArcMap tools within an individual’s MCP home range. We have added this clarification to the sentence.

196 What date did NDVI come from and why?

The imagery the NDVI was generated from was from late summer 2017. This imagery was used because it was the most recent and highest resolution imagery available. Also, with our study being conducted during the same time of year we felt that this imagery was as representative as could be. This information has been added to line 208 and 209.

197 What were the sorts of water sources in the remote sensed imagery. Often voles are associated with moist to wet spots that would not show up on imagery or maps. You trapped along creeks. Were they perennial?

Both creeks that we trapped near were ephemeral and visible on a USFS waterways layer for the Coronado National Forest. Additionally, a small seep that was located nearby was marked via GPS and digitized in the distance to water portion of our analysis. Despite the two creeks being designated as ephemeral, throughout the course of our study, they retained water. This information and citation have been added to lines 208 and 209.

205-206 I did not see mention of quadrates in the methods. Are these cover classes for percent cover?

Thank you for your concern. This sentence was removed however, this issue has been corrected in the methods.

211-220 To me it seems that both of your scales are really third order, selection of sites within the home range. Your patch scale seems like description of microhabitat. Your other scale seems like within home range (not home range), although I cannot determine where you drawing the available points from. If within home range, I think your available points should be drawn from the MCP of each individual.

Johnson 1980 describes 3rd order as the usage of various habitat components within the home range and 4th order as the actual procurement of food times if the 3rd order determines a feeding site. Johnson goes on to say that these orders can be divided more finely. We argue that the 4th order does not exclusively pertain to food items and can contain selection for micro habitat as other researchers have done (e.g. Compton et al. 2002, Sprague and Bateman 2018). At the 3rd order our available points were randomly generated withing each individuals MCP home range; this information was added to line 203. At the 4th order the corresponding available point was located 9.8 m away in a randomly generated direction from the known point as stated in lines 176-180.

214 within home range scale?

All references to ‘home range scale’ have been changed to ‘within home range scale’ to match terminology defined by Johnson 1980.

218-219 Where are the models? How many? How rationalized? The sentence implies the same models for each scale. Do you not expect different selection at different scales

Thank you for bringing this to our attention. We have included the reasoning we based our a priori models on with this addition on line 223: ‘based on previous research of habitat selection of long-tailed voles’. Additionally, we have changed the wording to be clear that we tested different models for each scale and how many models we tested.

224-225 I assume the male and female numbers are adults. How did you assign age class? Did you collar juveniles? Did you record sex on juveniles?

Adults were determined by weight (anything above 27 g). Juveniles were not collared or sexed. On lines 230 and 231 we state, “we collared and tracked 31 adult voles” but we have added clarification as well.

228-229 clarify what you mean by unknown but confirmed mortality. I think you mean you know it was dead but don’t know why. Collar slipping is slang. I think you mean that the collar fell off the animal. What does missing mean? Do you mean that you were unable to find a signal?

Thank you for bringing this to our attention. This section has been changed to read ‘unknown cause but confirmed mortality’ and ‘missing’ has been changed to ‘no signal’ to assist with clarity for the reader.

230 What do you mean by loss of vole?

Thank you for your concern on this section. We have added the explanation “(i.e. no longer able to have data collected)” for clarity.

222 You used individual as random effect, but I don’t know the sample size of male and females included in the analyses.

Thank you for bringing this to our attention. There were 9 males and 19 females used in this analysis. This information has been added to lines 196 and 197.

235 by group do you mean gender?

Thank you for your concern on this, the groups we tested our models with were all voles, males, and females. This has been clarified for the readers on lines 228 and 229.

237-238 Avoid using “shows” and “showing”.

Thank you for your input on this, these terms have been edited throughout the manuscript.

249 I don’t think you mentioned standardizing these variables in the methods, although perhaps I missed it.

Thank you for pointing out this deficiency, we have added the sentence ‘We standardized all covariates prior to running models’ to line 216 and 217.

256 Ideally I’d like to see the PCA results perhaps in supplemental material.

We decided to remove all mention of the PCA from the manuscript since it was unused and only caused confusion. Our intention in running a PCA was to cut down our 12 covariates into a smaller number of principal components. However, the results of our PCA showed we needed 9 principal components to account for 87% of the variation in the data. Originally, we included PC1 in our models because it accounted for 17% of the variation in data. We replaced any models that included grass and bare ground with PC1 as these were the covariates that PC1 represented. However, none of our top models for any group included PC1 based on delta AICc. In an effort to cut down on confusion we decided to remove all mention of our PCA analysis from this manuscript.

255 So did you not use the PCA? I am confused by what is being presented in this section.

Please see above response for concerns regarding PCA analyses.

258 This sentence does not read correctly—too many “models”.

Thank you for bringing this to our attention, we reworded the sentence.

260-263 Why not say that the Beta coefficients indicated positive selection for … and avoidance for ….

Thank you for this suggestion, we changed the wording of this sentence to reflect this comment.

Supplement Table 1. In looking at this table I am confused about your choice of variables and a priori hypotheses. Is the global model the top model? Are these all the models? I don’t think I understand why the PCI is used versus the other variables in the models? For instance, was log and grass also cover classes? I also see no rationale for these models.

These are all the models we ran at the 4th order. We did not run a global model in our analysis, we ran a priori models based on important factors identified in previous studies of long-tailed voles. This new addition was addressed in a previous comment on line 223. Additionally, all reference to PC1 have been removed to avoid confusion.

287 what do you mean by they showed varying degrees of selection?

Thank you for bringing this to our attention, we re-worded this line to be clearer.

290-291 Long-tailed voles are commonly captured in wet meadows and other herbaceous systems even when other voles are present. I’m not sure you can make this conclusion. Your analysis was at the home range (or perhaps within home range) and patch scales, while the habitat data you are contrasting with pertains to landscape or macro scale. I do not think it is surprising that long-tailed voles select the herbaceous environments in their surroundings. The same is true elsewhere in their range. In reflecting on this and looking back at your study area and trapping transect locations I really cannot get a feel for the environment in which the study was done. Was a mosaic of forest and open areas?

We have amended our statement to read “Our data contrast with some previous studies” and added the additional reference of Anich and Hadly 2013 to further support our statement. Randall 1978, Randall and Johnson 1979, and Anich and Hadly 20013 found that long-tailed voles compete with other vole species and experience competitive exclusion or inverse space occupancy. They found that when in the presence of other vole species long-tailed voles are rarely found in tall grass meadows and are relegated to the ecotonal tree-line areas.

Our study area is a mosaic of mostly closed canopy forest and open grassy meadows as well as areas of patchy tree canopy cover with an understory consisting of bare ground, forbs, ferns, and grasses.

297-298 There are a number of studies, mainly in the Midwest, that discuss the importance of roadside environments for voles and other species that select dense herbaceous vegetation. The idea is that these areas collect more moisture resulting in more lush vegetation.

This is interesting and we have incorporated a reference in the paragraph on line 341.

286-308 long paragraph with lots of different topics

Thank you for bringing this to our attention. We have broken this paragraph up.

312-313 Selection for downed logs and coarse woody debris paints a different set of conditions than grassy areas.

Thank you for this observation, this is something we wanted the readers to know. Because there may be confusion, we have re-worded the sentence to make this clearer. To clarify, voles selected for cover types of downed logs and coarse woody debris over other cover types at the patch scale.

315 long-tailed voles occur in relatively open areas. When they occur in forested ecosystems, their local distribution is in openings in the forest where there is higher herbaceous ground cover. Later, you also say that they select high herbaceous cover. You can only get high herbaceous cover if the sun is able to hit the ground, which does not happen in dense forests.

We apologize for our use of the term “open areas” by this we meant areas of bare ground and sparse herbaceous growth. We have corrected this in lines 330. Additionally, we ensured this issue was addressed throughout the manuscript.

319-320 I don’t think you can say your results support the idea that they experience competitive exclusion from high quality vole habitat. You did not test that.

We agree, this statement and the sentence was removed.

321-322. What is open habitat dominated by woody vegetation? I don’t understand what you mean by that.

Thank you for identifying this issue, we removed the word ‘open’.

323-324 This is a perfect description of the habitat of the long-tailed vole, but it holds true for almost any other mountain range where other species of voles also occur. I don’t think your description of habitat has to do with presence of the other species of vole. Otherwise, why would two species of vole so frequently co-occur together to the point that you catch them both on the same trap line?

You surveyed for voles along creeks in a generally forested area in the mountains. This is an ideal situation for long-tailed voles. It provides moist areas with herbaceous vegetation in openings and meadows within the forest. You can find long-tailed voles in any mountain range in this same conditions, regardless if there are other species.

Randall 1978, Randall and Johnson 1979, and Anich and Hadly 20013 found that long-tailed voles compete with other vole species and experience competitive exclusion or inverse space occupancy. They found when in the presence of other vole species long-tailed voles are rarely found in tall grass meadows and rarely caught in the same trap lines as other vole species. At the patch scale we found that white bellied voles were selecting for areas with high amounts of grassy cover. We have edited these lines to highlight the differences this population exhibits.

1. Randall JA. Behavioral mechanisms of habitat segregation between sympatric species of Microtus: Habitat preference and interspecific dominance. Behav Ecol Sociobiol. 1978;3(2):187–202.

2. Jan R, Johnson RE. Population densities and habitat occupancy by Microtus Longicaudus and M. Montanus. J Mammal. 1979;60(1):217–9.

3. Anich PS, Hadly EA. Asymmetrical competition between Microtus montanus and Microtus longicaudus in the Greater Yellowstone Ecosystem. Am Midl Nat. 2013;170:274–86.

338-339 It could also mean they select the location of home ranges based on different characteristics.

Thank you for this comment, we agree it could mean that. However, since we did not evaluate selection at the second order (home range scale), we would not feel comfortable incorporating this statement into this manuscript.

349 And higher water requirements during lactation.

In this sentence we believe the phrase “and increased energetic needs” includes these requirements and in an effort to keep this sentence as succinct as possible we did not incorporate this edit. Additionally, we discussed higher water requirements during lactation in line 365.

352-353 Again, I think you are confusing scales as I don’t see your results different from other locations where the species occurs.

Thank you for bringing this to our attention. To clarify our statement, we included “within home range” to line 368.

354 There is a danger of extrapolating third and fourth order selection to higher order selection such as geographic distribution. You don’t know what first and second order selection are.

We appreciate this insight; we have chosen to remove the mention of predictive use maps.

358 Are they imperiled? SGCN directs funding as is not.

In Arizona this subspecies is a Species of Greatest Conservation Concern as stated in line 100 and 101 (Arizona’s state wildlife action plan: 2012–2022). We have restated their Arizona status here for clarification.

Reviewer #2: This is a very interesting paper that examines space use in a less-studied species of vole, The All Voles model showed that voles avoided areas of bare ground and and areas far from roads and water, Females followed the same pattern with the addition of areas with high understory cover. Females tend to remain in their terriroty near thier nest sites. Males avoided bare ground but chose areas furthre fron road s and water. The authros try to fir their data to the promiscusous/polygynous mating systems, I found this fit most tenuos as the authros really have no direct data of interatiions between and within the sexes and their statments about mating systmes comes mainly from a revew chapter by Jerry Wolff (1985).

Thank you for your concern on this topic. There is very little information on this species in regard to intraspecific social interactions. Other studies have shown there can be pronounced sex differences in feeding behavior and habitat use in other species (Clutton-Brock and Iason 1987; Newsome 1980). We feel the data we have, shows a notable difference between male and female resource selection. This combined with information from Wolff 1985 we felt comfortable suggesting white-bellied voles have either promiscuous or polygynous mating systems. We exclude the possibility of a monogamous mating system because we would expect little to no differences in resource selection in a monogamous mating system . We have modified the wording to make it clear that we believe that our results ‘suggest’ that monogamy is less likely than the polygamous systems.

1. Clutton-Brock TH, Iason GR, Guinness FE. Sexual segregation and density-related changes in habitat use in male and female Red deer ( Cervus elaphus ). J Zool. 1987;211:275–89.

2. Newsome AE. Differences in the diets of male and female red kangaroos in central Australia. Afr J Ecol. 1980;18(1):27–31.

________________________________________

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Submitted filename: Responses to reviewers.docx

Click here for additional data file.

27 Oct 2020

Resource selection of a montane endemic: sex-specific differences in white-bellied voles (Microtus longicaudus leucophaeus)

PONE-D-20-27373R1

Dear Dr. Dutt,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Bi-Song Yue, Ph.D

Academic Editor

PLOS ONE

30 Oct 2020

PONE-D-20-27373R1

Resource selection of a montane endemic: sex-specific differences in white-bellied voles (Microtus longicaudus leucophaeus)

Dear Dr. Dutt:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Bi-Song Yue

Academic Editor

PLOS ONE