Temporal Patterns of Structural Sagebrush Connectivity from 1985 to 2020

Abstract

The sagebrush biome within the western United States has been reshaped by disturbances, management, and changing environmental conditions. As a result, sagebrush cover and configuration have varied over space and time, influencing processes and species that rely on contiguous, connected sagebrush. Previous studies have documented changes in sagebrush cover, but we know little about how the connectivity of sagebrush has changed over time and across the sagebrush biome. We investigated temporal connectivity patterns for sagebrush using a time series (1985–2020) of fractional sagebrush cover and used an omnidirectional circuit algorithm to assess the density of connections among areas with abundant sagebrush. By comparing connectivity patterns over time, we found that most of the biome experienced moderate change; the amount and type of change varied spatially, indicating that areas differ in the trend direction and magnitude of change. Two different types of designated areas of conservation and management interest had relatively high proportions of stable, high-connectivity patterns over time and stable connectivity trends on average. These results provide ecological information on sagebrush connectivity persistence across spatial and temporal scales that can support targeted actions to address changing structural connectivity and to maintain functioning, connected ecosystems.

1. Introduction

Many ecosystems are undergoing unprecedented change due to climate [1,2] and human land-use changes [3,4]. The resulting magnitude and direction of changes in landscape composition and configuration can vary through space and time and affect various species and ecosystem processes. Landscape change affects the physical contiguity of and linkages among landscape features, that is, the structural connectivity [5]. As such, structural connectivity is dynamic and can affect how ecological processes function across the landscapes at local and immediate scales as well as across broad spatiotemporal extents. However, connectivity is often assessed for a single period in time based on available input data, potentially failing to capture lost connections, relevant temporal variation, and trends of ecological patterns and processes [6]. Approaches that characterize connectivity over time are therefore needed to more fully measure the changes in landscape patterns and understand the effects of those changes [6], especially where connectivity may be lost.

Connectivity is a characteristic of many well-functioning and resilient ecosystems. The linkages across a landscape vary in time and space, supporting a diversity of ecosystem processes, including organic matter and nutrient exchanges [7], seed dispersal [8], and wildlife dispersals and migrations [9]. When considering whole ecosystems, characterizing the structural connectivity of key landscape features can emphasize species-agnostic connections [10] and increase our understanding of how the landscape can facilitate the movement of processes across space and time. This structural perspective can complement and contrast with functional connectivity, which emphasizes how the landscape facilitates or impedes the specific spatial functions of ecological processes such as species dispersal [5]. Moreover, connectivity studies can help reveal how landscape configuration may contribute to an ecosystem’s resilience to disturbances (e.g., in temperate forest ecosystems, [11]). Connectivity can facilitate population and metapopulation dynamics, such as the demographic rescue of isolated populations and the recolonization or establishment of new populations, which can mediate the negative effects of habitat loss and fragmentation [12,13,14]. Enduring structural connections can also support resilience through the possibility of adaptation to changing landscapes, such as species’ range shifts [15].

Measuring temporal connectivity patterns can be challenging due to the limited availability of historical spatial data relevant to the condition and process being measured and the varying time frames of different kinds of landscape change [16]. However, access to historical time series of land cover and vegetation data are making temporal connectivity analyses increasingly possible, and provide new insights into the patterns of change affecting ecosystems and wildlife. Connectivity patterns at one point in time represent only a snapshot of the dynamic trends occurring [17]. By contrast, the use of timeseries data to identify the location, direction, and magnitude of variation in connectivity provides information about where and how the landscape is changing. It can be particularly valuable to characterize trends in structural connectivity to identify locations that are declining or have declined over time, as physical connections become weaker or impeded relative to a prior condition. As landscapes become more fragmented, with constrained or channelized linkages, there may be risk of increased species extinction [18], biodiversity loss [19], and disruption of ecosystem services [20].

The sagebrush biome of the western United States supports a rich ecosystem across broad geographic and ecological gradients. The biome has experienced significant changes in vegetation composition and configuration related to wildfires, annual grass invasion, energy development, conifer expansion, agriculture conversion, and urbanization, as well as conservation and restoration actions ([21], chapters J-P). We expect that these changing patterns of sagebrush cover [22] directly affect structural connectivity within the biome. Although past work on connectivity in the biome has investigated species-specific functional connectivity for wildlife, such as Greater Sage-grouse (Centrocercus urophasianus) gene flow [23] and pronghorn (Antilocapra americana) migration [24], no studies have taken a structural approach to assess sagebrush connectivity patterns over time. Vegetation timeseries maps are increasingly available across the sagebrush ecosystem and allow for such temporal assessments. For example, historical sagebrush fractional cover from the Rangeland Condition Monitoring Assessment and Projection (RCMAP, [25]) maps several kinds of co-occurring vegetation, and these mapping products are increasing opportunities for spatial analyses at broad spatial scales. Structural connectivity analysis can complement existing vegetation cover maps and species-specific functional connectivity models by focusing on the physical linkages resulting from changing sagebrush cover. Structural connectivity of sagebrush could potentially facilitate diverse ecosystem processes and benefit species that have common landscape needs in the sagebrush ecosystem.

Measuring how sagebrush connectivity and linkages have changed over time could help guide future conservation and restoration management efforts within the sagebrush biome. Within the biome, strategies to protect and extend core sagebrush areas are being developed [26]. Connectivity approaches that provide outputs at the broad, biome-wide scale can support regional comparisons and prioritization support for land managers. Structural connectivity analyses can also identify local areas in which protection or restoration actions may be needed to increase landscape resilience (e.g., via stepping stones and networks; see [27]). Studying the connectivity patterns for sagebrush over time will be useful in helping to better understand how and for what processes connectivity is important in the biome.

To characterize connectivity that is comparable across spatial and temporal scales, it is necessary to use scalable and consistent methods among time periods and locations. Connectivity analyses based on circuit theory [28] provide such an approach. Concepts of electrical flow share commonalities with gene flow and other ecological processes that follow random-walk models [28,29], and approximate the mathematical probability of paths or circuits across systems whose components variably resist or conduct current flow. Circuit-based analyses can be scaled up from local to landscape extents to assess the broad spatial context for connectivity, and classified outputs can be compared across temporal extents [30]. The circuit connectivity paradigm evaluates connectivity as the contributions of multiple random-walk dispersal pathways across heterogeneous landscapes, modeled as the density of electrical current flow over a network of resistors [28,29]. Similar to the flow of electrical current, the flow of processes, wildlife, or their genes among landscape patches can be less concentrated (e.g., diffuse) when there are many potential pathways or more concentrated (e.g., channelized) when those pathways are constrained. The temporal grain, or frequency, at which structural connectivity is assessed should align with the major drivers of landscape change. As the landscape composition and configuration change among maps of different time periods, current density patterns will shift. The resulting direction and magnitude of those changes are used to interpret where and how connectivity has changed.

We used omnidirectional circuit theory [31] to assess broad patterns of sagebrush structural connectivity across the sagebrush biome from 1985 to 2020. The species-agnostic approach assumes the structural connections among high sagebrush cover are relevant for supporting spatial ecological processes. Our objectives were to quantify the temporal changes in structural connectivity by identifying patterns of stability and variability, trends of gain or loss, and changes in relative flow (i.e., impedance or channelization) over time. We also considered these temporal patterns of structural connectivity within and among priority conservation areas relevant for sagebrush management.

2. Materials and Methods

2.1. Study System

The sagebrush biome is an extensive system that spans 14 states in the western United States. Federal agencies manage over half of the land within the biome and approximately 39% of the land is under private ownership. For our study, we consider the sagebrush ecosystems, where the shrub composition primarily includes big sagebrush (Artemisia tridentata ssp.) with subspecies of mountain (A. t. spp. vaseyana), Wyoming big sagebrush (A. t. subsp. wyomingensis), and basin (A. t. ssp. tridentata), as well as other, less-abundant, non-big sagebrush and non-sagebrush shrub species (chapter A, [21] ). To characterize this vegetation community, we used time series data of sagebrush fractional components at five-year intervals from 1985 through 2020 (eight timesteps, [32]). Sagebrush cover can be lost as a result of disturbances due to wildfires, cultivation, grazing, and development, as well as climatic and other changes. Recovery of sagebrush through natural or assisted revegetation can require several decades [33], and we conducted temporal sampling every five years over a period of 35 years (eight time steps). These eight continuous rasters estimated the proportion of sagebrush (A. spp.) canopy in each 30-m pixel. For each timestep, we calculated the amount of area within the biome where sagebrush was present (cover > 1%) and abundant (cover > 15%). For context, 15% cover was ~90th percentile value of biome-wide values for sagebrush in 2020.

There are various designations for conservation and management interest areas within the sagebrush biome. We considered two: Priority Areas for Conservation (PACs) for Greater Sage-grouse (C. urophasianus) [34] and Sagebrush Ecological Integrity classes (SEI; [35]. Although primarily species-specific, the PACs are areas that states have designated as important and are incorporated into various state- and biome-scale management considerations. The PACs were established by states using sage-grouse population densities and movements between breeding and nesting habitats, reflecting the most intact populations (at a single time point, 2015). Each state manages these designations with consideration of protecting sage-grouse from further habitat losses and population declines. The SEI classes were developed as part of the broader Sagebrush Conservation Design report [26] and are not wildlife-based. The SEI classes reflect a non-rangeland class and three rangeland classes based on ground conditions, averaged across years (2001, 2006, 2011, 2016, and 2020). The SEI rangeland classes were categorized based on landscape characteristics, with high SEI scores defined by abundant sagebrush, native understories, and minimal threats. In order of descending quality, the SEI rangeland classes are core sagebrush areas, growth opportunity areas, and other rangeland areas. These two designations (PACs and SEI classes) focus on landscape-level planning and strategy across the biome, which is directly relevant to structural connectivity.

2.2. Circuit-Based Connectivity Model

We calculated structural connectivity using the omnidirectional circuit-based Omniscape algorithm (v0.5.7, [31,36]; see Supplementary Materials and/or [37] for initialization file) in Julia (v1.6.7, [38,39]). We used the sagebrush fractional component for each selected year to represent the conductance surface. Areas with higher sagebrush cover were considered proportionally higher conductance for ecological processes; inversely, we considered areas with lower cover (and therefore lower conductance) to indicate where connectivity may be impeded without making assumptions about which specific landscape features might serve as barriers for different ecological flows.

At each time step, we used that year’s conductance surface to determine the current flow sources and strengths [36] and 15% sagebrush cover for the source threshold. We used a 75-km moving window size, as we were not targeting a specific ecosystem process or wildlife function; beyond this distance we saw diminishing changes in connectivity outputs. We reduced memory requirements by resampling the input rasters to 270 m; the resulting pixels represented the mean fractional sagebrush cover of nine 30-m pixels and reduced some detail originally captured in the RCMAP products. To reduce computation time, we set the Omniscape block size equivalent to 1/10 of the window radius [40], tiled the conductance raster inputs, and used parallel processing to run models on the U.S. Geological Survey Denali Supercomputer [41].

We used Omniscape to calculate the cumulative current density, which represents the sum of potential linkages under random-walk movement among source pixels. We reclassified the cumulative current density outputs into quantiles (20 classes) to facilitate trend calculations and comparisons across years. We additionally calculated the normalized cumulative current density, which represents the relative flow of current by dividing cumulative current density by flow potential under null resistance conditions [36]. We categorized pixels into three relative flow classes using standard deviation (SD). Normalized cumulative current density values equal to or near one indicated relatively diffuse flow patterns (1 ± 1 × SD). Low normalized cumulative current density values indicated impeded flow (<1 – 1 × SD), whereas high values indicated channelized flow (>1 + 1 × SD).

2.3. Analyses

We conducted three types of analyses for temporal patterns of sagebrush structural connectivity. We calculated (1) connectivity stability and variability patterns, (2) temporal connectivity trends, and (3) temporal trends in relative flow. We assessed these patterns across the spatial extent of the sagebrush biome and subset the three results for the PACs and SEI classes.

To investigate connectivity stability and variability patterns, we identified where connectivity had been consistently high or low over the 35 years. For each pixel in the landscape, we calculated the number of times it had been in either the top (75th percentile, high connectivity) or bottom (25th percentile, low connectivity) range of the connectivity distribution based on cumulative current density values. To understand the temporal connectivity trends between 1985 and 2020, we fit a linear regression for each pixel to calculate sagebrush connectivity class as a function of time. We used the mapped trends and the distribution of slope values to understand the spatial patterns and magnitude of connectivity change.

We also calculated the temporal trends in relative flow between 1985 and 2020 by fitting a linear regression for each pixel based on the normalized cumulative current density values. We then identified areas where connectivity was changing by becoming increasingly impeded or channelized over time. To calculate area of increasing impedance, we identified pixels that had a starting condition of diffuse or impeded flow (1985 normalized cumulative current density values ≤ 1) that became more impeded (negative trend slope). We repeated this to identify increasingly channelized areas by identifying pixels that had a starting condition of diffuse or channelized flow (1985 normalized cumulative current density values ≥ 1) and increasingly channelized flow (positive trend slope). We also compared the total area within each of the three relative flow classes for 1985 and 2020.

3. Results

The presence and abundance of sagebrush changed over time (RCMAP, [32]). Areas where sagebrush was present (≥1% cover) comprised approximately 75% of the biome extent in 1985 and 2020, with variation among years (Figure 1). Areas with abundant sagebrush (>15% cover) increased overall between 1985 and 2020 (Figure 1) and covered less than one tenth of the biome by area. These variations likely reflect the combination of change in true on-the-ground sagebrush cover and the effects of the remote sensing and sampling processes.

3.1. Biome-Wide Analyses of Temporal Connectivity

Changes in sagebrush cover at different time steps resulted in varying structural connectivity patterns. Calculating the connectivity at each time step (Figure 2a for 2020; see [37] for full time series) and analyzing the time series allowed us to quantify the stability and variability of connections (Figure 2b) and the temporal trends in connectivity (Figure 2c). Areas with consistently high connectivity comprised 19.3% of the biome, with 339,923 km² of the landscape falling in the top connectivity classes (75th percentile or above) at all eight timesteps. An additional 11% of the biome had high connectivity in at least one of the time steps. Areas with consistently low connectivity across all time steps (25th percentile or below) comprised 20.4% of the biome (359,708 km²), with an additional 9.7% of the biome having low connectivity in at least one of the time steps. Overall, approximately 60.3% of the biome (1,060,502 km²) had moderate or non-persistent connectivity patterns among time steps (Figure 2b; gray). Temporal trends in connectivity varied spatially (Figure 2c), with some areas increasing in connectivity (teal) and other regions decreasing (brown). On average, areas had limited to moderate change in connectivity class over time with a unimodal normal distribution of slope values (gray; mean = 0.00, SD = 1.15).

Relative flow patterns, which categorize impeded, diffuse, or channelized flow, provided additional insight into changes in local connectivity patterns over time. Calculating the relative flow at each time step allowed us to interpret how the presence and location of sagebrush cover for that year could be limiting or facilitating potential connections among areas with abundant sagebrush (Figure 3a for 2020; see [37] for full time series). Over time, these patterns changed. The most common relative flow pattern in the biome was diffuse flow, and represented areas where many potential, unconstrained connections among sagebrush occurred (Figure 3a for 2020). Diffuse relative flow comprised over half the biome over time; 57.9% of the biome in 1985 (10,191,611 km²) and 56.9% in 2020 (10,018,043 km²). Impeded areas, where the landscape had relatively higher than expected resistance, comprised approximately a third of the landscape. Some areas became more impeded over time (Figure 3b), with 34.4% of the landscape classified as impeded in 1985 (6,049,437 km²) and 35.3% in 2020 (6,217,998 km²). Although channelized areas were limited, some areas became more channelized over time (Figure 3c), and the channelized flow class increased by about 5000 km² between 1985 and 2020 (7.73% in 1985 = 1,360,282 km², 7.76% in 2020 = 1,365,288 km²).

3.2. Temporal Connectivity Patterns within Conservation Areas

Connectivity patterns and trends differed within PACs compared with areas outside of them. PACs, which cover approximately 17.8% of the sagebrush biome (313,505 km²), generally contained areas with relatively high connectivity. For example, the mean connectivity class for areas within PACs in 2020 was 15.17 (SD = 4.84), higher than the mean connectivity class for those outside PACs that year (mean = 9.49 SD = 5.44). Areas with consistent, high connectivity at all eight time-steps comprised approximately half of the available area within PACs. Most of the consistent low-connectivity area was found outside PACs (Figure 4b). On average, the trend slope within PACs showed a slight increase in connectivity, as measured by connectivity class, between 1985 and 2020, while the area outside of PACs slightly decreased over time (Figure 4c). Within one standard deviation, however, these trend slopes overlapped with zero.

Connectivity patterns and trends also differed across the Sagebrush Ecological Integrity (SEI) classes. Over time, SEI core sagebrush and growth opportunity areas had greater proportions of consistent, high connectivity than the other rangeland and non-rangeland areas (Figure 4e). On average, the areas within SEI non-rangeland and growth areas had minimal increases in connectivity, as measured by connectivity class trend between 1985 and 2020, while area in core and other rangelands slightly decreased over time (Figure 4f). However, as with the PACs, the standard deviation of values within each SEI class was much larger than the average trend and overlapped with zero.

4. Discussion

Viewing connectivity through the lens of dynamic change enables scientists and practitioners to assess the long-term implications of broad-scale landscape variation. We calculated temporal patterns of structural sagebrush connectivity in the western United States from 1985 to 2020 using a circuit-based approach. Most of the biome experienced changes in connectivity, but the location, magnitude, and type of change varied. The conservation and management areas we examined showed stable, high connectivity and positive connectivity trends over time. The sagebrush biome faces many threats related to habitat loss [42,43]; our research identified patterns of where and when structural changes occurred, laying the necessary groundwork to further study how structural connectivity may support a functioning, resilient sagebrush ecosystem.

Our connectivity models and time series analyses can provide key data to investigate the ecological importance of structural connectivity that persists over time. Currently, we do not fully know the ecological processes that benefit from sagebrush connectivity or the benefits these consistently high connectivity areas can provide for the sagebrush biome. The areas of stable, high connectivity we found within the sagebrush biome may provide spatial insurance [44] by preserving connections for spatial exchange. Maintaining structural connectivity in these regions will likely support multifunctional landscapes and ecosystem services, providing reliably connected areas for spatial ecological processes in the near term. However, we found declining connectivity trends also occurred within regions with consistently high connectivity over the past 35 years, indicating areas of potential concern. For example, we observed a loss of structural connectivity in the Dakotas, the western range of the Rocky Mountains, northwestern Nevada, and eastern Oregon (Figure 2c). Areas with decreasing connectivity may be at risk, or have already undergone ecological consequences such as biodiversity loss [45,46,47]. Connectivity may also be important in the face of climate change and disturbance to preserve metapopulations and biodiversity [11,18]. Identifying connected areas that could provide climate refugia may be long-term conservation priorities if the landscape includes or facilitates movement or processes. Connectivity analyses coupled with climate projections may further support the identification of priority locations for securing and augmenting connectivity in the sagebrush biome for multiple species, purposes, and timeframes [48]. Although our trend analysis does not quantify the ecological importance of the connectivity changes for these regions, documenting these temporal connectivity patterns contributes to our understanding of ecosystem change, and provides new insight into areas of conservation concern.

Temporal connectivity patterns can provide additional context beyond a single period, which can be important both for ecological understanding and management decision making. Gregory and Beier [16] found that the lag times of organismal responses to connectivity changes vary; therefore, the connectivity history of a location may foreshadow ecological consequences that are yet to come. In our analysis, areas that had low connectivity in 2020, but a history of high connectivity, could represent areas at risk for fragmentation-related ecological consequences. By contrast, areas with both current and historical low connectivity may have limited contributions to spatial exchange as they have a consistent and lesser ability to facilitate the spatial flow of ecosystem processes. The spatial variation in connectivity trends can help differentiate and strategize appropriate actions. Areas with steeply declining connectivity may need more action (or acceptance if other connectivity options exist) than those that have shown minimal declines. In some cases where disturbances such as wildfire can rapidly fragment the habitat, managers will need to consider the tradeoffs of mitigating disturbances and preserving contiguous habitat. Future work may investigate how the speed of temporal changes could have ecological implications, for example, when connections are lost rapidly (e.g., due to fire, land cover change) or lost more slowly (e.g., due to climate change). Areas with increasing connectivity could be proactively protected and investigated to understand what conditions support sagebrush growth and linkages. Relative flow can identify patterns such as increasing channelization that could identify high return on investments in protecting linkages, since they are often spatially limited but have disproportionate effects on connectivity [49].

The changing landscape led to constrained (i.e., channelized) connectivity patterns in some areas, that changed through time. Connectivity analysis for a single point and time can identify pinch points, but assessing trends can indicate at-risk connections that have become increasingly ‘pinched’ over time. We found a slight increase in channelized relative flow patterns across the biome between 1985 and 2020, indicating reduced connections among sagebrush. These at-risk areas could be used to target management actions to prevent further loss and avoid the loss of connections that could have a disproportionate effect on regional or biome-wide connectivity [49]. For example, impeded flow and loss of connections over time have the potential to limit the movement and migration of sagebrush-associated wildlife species such as pronghorn [24]. When conservation goals are species-specific, however, functional connectivity models that specifically account for focal species’ movement should be a more accurate reflection of connectivity limitations for movement and an important complement to species-agnostic structural models [10]. The location and magnitude of connectivity changes may be more consequential for some species or processes than for others. Structural connectivity and functional connectivity approaches represent different perspectives on the landscape and correspond with different modeling goals [10].

The evaluation of structural connectivity within PACs and SEI classes provided insight into where and how conservationists could focus management in the future. We observed that these areas contained consistently high connectivity; this was not surprising because they were delineated to protect sagebrush obligate species (PACs) and represent intact sagebrush areas (SEI classes). Our analyses also indicated that connectivity in these areas remained stable on average, suggesting past conditions and conservation actions were effective for sustaining sagebrush’s structural connectivity. In PACs, connectivity outside priority areas was lower and some locations were trending towards disconnection; such changes could become important if managers determine that these structural connectivity declines also negatively affect species’ dispersals among priority areas. Mapped outputs of the connectivity trends could be used to identify places with decreasing connectivity that could be the focus of connectivity improvements, or where persistent high-connectivity linkages among conservation areas should be protected. SEI growth opportunity areas with past high connectivity could be targeted for habitat restoration to help expand the core sagebrush areas and restore lost connections. For locations where non-rangeland and other rangeland SEI classes decreased in connectivity, restoration of these areas might have little return on investment if they are isolated with limited ecological connectivity, or if restoration objectives target species that require larger, connected sagebrush habitat. Future research could further investigate how the structural connectivity patterns we found correlate with existing functional connectivity models for sagebrush-obligate species (e.g., [23,50,51]).

The circuit-based approach allowed us to assess the landscape in a scalable and repeatable way using open-source resources. The omnidirectional outputs are relevant for landscape-based processes that are not node- or patch-dependent and provide values that could be classified and compared across the time series. However, this simple connectivity analysis required assumptions that others may want to modify or explore further. A key assumption of our structural connectivity model was that sagebrush cover was the sole determinant of current sources and strengths in the circuit model. We decided to minimize other assumptions about which other (non-sage) variables would be relevant for connectivity and to keep it as species- and function-agnostic as possible. Connectivity of sagebrush-related vegetation was of interest in this study, but additional layers could be added to more broadly represent vegetation in the ecosystem, such as perennial grass components. Other connectivity models that consider ecological integrity or naturalness (e.g., [40]) may also be relevant, although these products tend to be limited to specific time frames. Similarly, species-specific connectivity assessments should consider additional spatial criteria related to habitat and matrix conditions and movement limitations. Such assessments could complement each other to provide a broader and multi-faceted view of connectivity throughout the sagebrush biome.

A lack of comparable time series maps may have limited past explorations of dynamic connectivity; however, new remote sensing data products, methodological advances, and computing resources are yielding opportunities to better understand dynamic landscapes. We show that dynamic connectivity assessments can provide actionable insights for near-term conservation and restoration of the sagebrush biome by quantifying past patterns and changes in sagebrush presence and cover. As disturbances alter the structure of the sagebrush biome and many landscapes around the globe, we need analytical methods that are capable of monitoring and describing simple changes in vegetation communities over large spatial extents. Patterns and trend data from temporally dynamic connectivity analyses can refine target locations for vegetation restoration and the protection of ecological areas of importance, and in doing so, can provide a foundation on which to build efficient management strategies. Our analyses provide data products that can be used to investigate both the mechanisms of changes in connectivity and the ecological implications of higher or lower structural connectivity for species and ecological processes across the sagebrush biome. Future research can extend our work to examine the implications of the loss (or gain) in structural connectivity for biodiversity, population persistence, movement, or other species and processes in the sagebrush ecosystem, while considering different scales, lag times, and species or process responses.

Author Contributions

Conceptualization, J.A.H., M.S.O. and E.K.B.; Methodology, J.A.H., M.S.O. and E.K.B.; Software, E.K.B.; Formal Analysis, E.K.B.; Investigation, E.K.B.; Resources, J.A.H., M.S.O., C.L.A. and E.K.B.; Data Curation, E.K.B.; Writing—Original Draft Preparation, E.K.B.; Writing—Review & Editing, J.A.H., M.S.O., C.L.A. and E.K.B.; Visualization, E.K.B.; Funding Acquisition, J.A.H., M.S.O. and C.L.A. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

The data presented in this study are openly available in Buchholtz et al. 2023 at https://doi.org/10.5066/P9ED3OHH (accessed on 27 January 2023).

Acknowledgments

We are grateful for the technical support of the USGS Advanced Research Computing team and access to the U.S. Geological Survey Denali supercomputer. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Conflicts of Interest

The authors declare no conflict of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures

Figure 1. (a) Sagebrush cover in 2020; (b) Sagebrush presence and abundance from 1985 to 2020 based on annual data depicting sagebrush fractional cover [32] in the sagebrush biome of the western United States. Values represent mean percent cover at 270 m. Large points indicate the years at 5-year intervals that were used for the analyses in this paper. Note the different scales of the y-axes for presence and abundance.

Figure 2. Overall structural connectivity patterns in the sagebrush biome of the western United States were based on omnidirectional cumulative current density outputs. (a) Structural sagebrush connectivity displaying a single time step (2020), classified by cumulative current quantiles; (b) Zones with consistently high and low connectivity patterns over eight timesteps from 1985 to 2020. Gray areas had moderate or inconsistent connectivity; (c) Linear trends for quantile-based connectivity classes over time, calculated per pixel, from 1985 to 2020. Increasing connectivity (teal), decreasing connectivity (brown).

Figure 3. Relative flow patterns for sagebrush structural connectivity in the sagebrush biome of the western United States. (a) Relative flow classes in 2020, for impeded (Imp.), diffuse (Dif.), and channelized (Chn.) patterns; (b) Areas with increasingly impeded flow from 1985 to 2020 (blue gradient); (c) Areas with increasingly channelized flow from 1985 to 2020 (red gradient). Slopes for B and C indicate change in relative flow patterns calculated as a linear regression per pixel from 1985 to 2020.

Figure 4. Temporal patterns of structural sagebrush connectivity in the western United States within designated areas of conservation and management importance (i.e., the distribution of Figure 2b,c within management designations). (a) The distribution of Priority Areas for Conservation (PACs; U.S. Fish and Wildlife Service 2013) within the sagebrush biome; (b) Proportions of the areas within PACs and outside of PACs that have stable high (yellow) and stable low (blue) connectivity patterns from 1985 to 2020; (c) Mean trend and standard deviation for linear change in connectivity class between 1985 and 2020 within and outside PACs; (d) The distribution of Sagebrush Ecological Integrity classes (SEI; Doherty et al. 2022b) within the sagebrush biome; (e) Proportions of areas that had stable high (yellow) or stable low (blue) connectivity within each SEI class from 1985 to 2020; (f) Mean trend and standard deviation for linear change in connectivity class between 1985 and 2020 within each SEI class.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/land12061176/s1, Figure S1: Initialization (.ini) file used for running Omniscape v0.5.7 circuit-based connectivity algorithm.

References

1. Grimm, N.B.; Chapin, F.S., III; Bierwagen, B.; Gonzalez, P.; Groffman, P.M.; Luo, Y.; Melton, F.; Nadelhoffer, K.; Pairis, A.; Raymond, P.A. et al. The Impacts of Climate Change on Ecosystem Structure and Function. Front. Ecol. Environ.; 2013; 11, pp. 474-482. [DOI: https://dx.doi.org/10.1890/120282]

2. Weiskopf, S.R.; Rubenstein, M.A.; Crozier, L.G.; Gaichas, S.; Griffis, R.; Halofsky, J.E.; Hyde, K.J.W.; Morelli, T.L.; Morisette, J.T.; Muñoz, R.C. et al. Climate Change Effects on Biodiversity, Ecosystems, Ecosystem Services, and Natural Resource Management in the United States. Sci. Total Environ.; 2020; 733, 137782. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2020.137782] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32209235]

3. Hasan, S.S.; Zhen, L.; Miah, M.G.; Ahamed, T.; Samie, A. Impact of Land Use Change on Ecosystem Services: A Review. Environ. Dev.; 2020; 34, 100527. [DOI: https://dx.doi.org/10.1016/j.envdev.2020.100527]

4. Yee, S.H.; Paulukonis, E.; Simmons, C.; Russell, M.; Fulford, R.; Harwell, L.; Smith, L.M. Projecting Effects of Land Use Change on Human Well-Being through Changes in Ecosystem Services. Ecol. Model.; 2021; 440, 109358. [DOI: https://dx.doi.org/10.1016/j.ecolmodel.2020.109358]

5. Calabrese, J.M.; Fagan, W.F. A Comparison-Shopper’s Guide to Connectivity Metrics. Front. Ecol. Environ.; 2004; 2, pp. 529-536. [DOI: https://dx.doi.org/10.1890/1540-9295(2004)002[0529:ACGTCM]2.0.CO;2]

6. Zeller, K.A.; Lewison, R.; Fletcher, R.J.; Tulbure, M.G.; Jennings, M.K. Understanding the Importance of Dynamic Landscape Connectivity. Land; 2020; 9, 303. [DOI: https://dx.doi.org/10.3390/land9090303]

7. Amoros, C.; Bornette, G. Connectivity and Biocomplexity in Waterbodies of Riverine Floodplains. Freshw. Biol.; 2002; 47, pp. 761-776. [DOI: https://dx.doi.org/10.1046/j.1365-2427.2002.00905.x]

8. Auffret, A.G.; Plue, J. Scale-Dependent Diversity Effects of Seed Dispersal by a Wild Herbivore in Fragmented Grasslands. Oecologia; 2014; 175, pp. 305-313. [DOI: https://dx.doi.org/10.1007/s00442-014-2897-7]

9. Webster, M.S.; Marra, P.P.; Haig, S.M.; Bensch, S.; Holmes, R.T. Links between Worlds: Unraveling Migratory Connectivity. Trends Ecol. Evol.; 2002; 17, pp. 76-83. [DOI: https://dx.doi.org/10.1016/S0169-5347(01)02380-1]

10. Marrec, R.; Abdel Moniem, H.E.; Iravani, M.; Hricko, B.; Kariyeva, J.; Wagner, H.H. Conceptual Framework and Uncertainty Analysis for Large-Scale, Species-Agnostic Modelling of Landscape Connectivity across Alberta, Canada. Sci. Rep.; 2020; 10, 6798. [DOI: https://dx.doi.org/10.1038/s41598-020-63545-z]

11. Craven, D.; Filotas, E.; Angers, V.A.; Messier, C. Evaluating Resilience of Tree Communities in Fragmented Landscapes: Linking Functional Response Diversity with Landscape Connectivity. Divers. Distrib.; 2016; 22, pp. 505-518. [DOI: https://dx.doi.org/10.1111/ddi.12423]

12. Edelsparre, A.H.; Shahid, A.; Fitzpatrick, M.J. Habitat Connectivity Is Determined by the Scale of Habitat Loss and Dispersal Strategy. Ecol. Evol.; 2018; 8, pp. 5508-5514. [DOI: https://dx.doi.org/10.1002/ece3.4072] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29938069]

13. Ovaskainen, O.; Saastamoinen, M. Frontiers in Metapopulation Biology: The Legacy of Ilkka Hanski. Annu. Rev. Ecol. Evol. Syst.; 2018; 49, pp. 231-252. [DOI: https://dx.doi.org/10.1146/annurev-ecolsys-110617-062519]

14. Dallas, T.A.; Saastamoinen, M.; Schulz, T.; Ovaskainen, O. The Relative Importance of Local and Regional Processes to Metapopulation Dynamics. J. Anim. Ecol.; 2020; 89, pp. 884-896. [DOI: https://dx.doi.org/10.1111/1365-2656.13141] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31705670]

15. Huang, J.-L.; Andrello, M.; Martensen, A.C.; Saura, S.; Liu, D.-F.; He, J.-H.; Fortin, M.-J. Importance of Spatio–Temporal Connectivity to Maintain Species Experiencing Range Shifts. Ecography; 2020; 43, pp. 591-603. [DOI: https://dx.doi.org/10.1111/ecog.04716]

16. Gregory, A.J.; Beier, P. Response Variables for Evaluation of the Effectiveness of Conservation Corridors. Conserv. Biol.; 2014; 28, pp. 689-695. [DOI: https://dx.doi.org/10.1111/cobi.12252]

17. Bishop-Taylor, R.; Tulbure, M.G.; Broich, M. Evaluating Static and Dynamic Landscape Connectivity Modelling Using a 25-Year Remote Sensing Time Series. Landsc. Ecol; 2018; 33, pp. 625-640. [DOI: https://dx.doi.org/10.1007/s10980-018-0624-1]

18. Morelli, T.L.; Maher, S.P.; Lim, M.C.W.; Kastely, C.; Eastman, L.M.; Flint, L.E.; Flint, A.L.; Beissinger, S.R.; Moritz, C. Climate Change Refugia and Habitat Connectivity Promote Species Persistence. Clim. Chang. Responses; 2017; 4, 8. [DOI: https://dx.doi.org/10.1186/s40665-017-0036-5]

19. Horváth, Z.; Ptacnik, R.; Vad, C.F.; Chase, J.M. Habitat Loss over Six Decades Accelerates Regional and Local Biodiversity Loss via Changing Landscape Connectance. Ecol. Lett.; 2019; 22, pp. 1019-1027. [DOI: https://dx.doi.org/10.1111/ele.13260]

20. Mitchell, M.G.E.; Suarez-Castro, A.F.; Martinez-Harms, M.; Maron, M.; McAlpine, C.; Gaston, K.J.; Johansen, K.; Rhodes, J.R. Reframing Landscape Fragmentation’s Effects on Ecosystem Services. Trends Ecol. Evol.; 2015; 30, pp. 190-198. [DOI: https://dx.doi.org/10.1016/j.tree.2015.01.011] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25716547]

21. Remington, T.E.; Deibert, P.A.; Hanser, S.E.; Davis, D.M.; Robb, L.A.; Welty, J.L. Sagebrush Conservation Strategy—Challenges to Sagebrush Conservation; Open-File Report U.S. Geological Survey: Reston, VA, USA, 2021; Volume 2020-1125.

22. Shi, H.; Rigge, M.; Homer, C.G.; Xian, G.; Meyer, D.K.; Bunde, B. Historical Cover Trends in a Sagebrush Steppe Ecosystem from 1985 to 2013: Links with Climate, Disturbance, and Management. Ecosystems; 2018; 21, pp. 913-929. [DOI: https://dx.doi.org/10.1007/s10021-017-0191-3]

23. Row, J.R.; Doherty, K.E.; Cross, T.B.; Schwartz, M.K.; Oyler-McCance, S.J.; Naugle, D.E.; Knick, S.T.; Fedy, B.C. Quantifying Functional Connectivity: The Role of Breeding Habitat, Abundance, and Landscape Features on Range-Wide Gene Flow in Sage-Grouse. Evol. Appl.; 2018; 11, pp. 1305-1321. [DOI: https://dx.doi.org/10.1111/eva.12627]

24. Zeller, K.A.; Schroeder, C.A.; Wan, H.Y.; Collins, G.; Denryter, K.; Jakes, A.F.; Cushman, S.A. Forecasting Habitat and Connectivity for Pronghorn across the Great Basin Ecoregion. Divers. Distrib.; 2021; 27, pp. 2315-2329. [DOI: https://dx.doi.org/10.1111/ddi.13402]

25. Rigge, M.; Homer, C.; Cleeves, L.; Meyer, D.K.; Bunde, B.; Shi, H.; Xian, G.; Schell, S.; Bobo, M. Quantifying Western U.S. Rangelands as Fractional Components with Multi-Resolution Remote Sensing and In Situ Data. Remote Sens.; 2020; 12, 412. [DOI: https://dx.doi.org/10.3390/rs12030412]

26. Doherty, K.; Theobald, D.M.; Bradford, J.B.; Wiechman, L.A.; Bedrosian, G.; Boyd, C.S.; Cahill, M.; Coates, P.S.; Creutzburg, M.K.; Crist, M.R. et al. A Sagebrush Conservation Design to Proactively Restore America’s Sagebrush Biome; Open-File Report U.S. Geological Survey: Reston, VA, USA, 2022; Volume 2022-1081.

27. Crist, M.R.; Knick, S.T.; Hanser, S.E. Range-Wide Connectivity of Priority Areas for Greater Sage-Grouse: Implications for Long-Term Conservation from Graph Theory. Condor; 2017; 119, pp. 44-57. [DOI: https://dx.doi.org/10.1650/CONDOR-16-60.1]

28. McRae, B.H.; Dickson, B.G.; Keitt, T.H.; Shah, V.B. Using Circuit Theory to Model Connectivity in Ecology, Evolution, and Conservation. Ecology; 2008; 89, pp. 2712-2724. [DOI: https://dx.doi.org/10.1890/07-1861.1]

29. McRae, B.H.; Beier, P. Circuit Theory Predicts Gene Flow in Plant and Animal Populations. Proc. Natl. Acad. Sci. USA; 2007; 104, pp. 19885-19890. [DOI: https://dx.doi.org/10.1073/pnas.0706568104] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/18056641]

30. Dickson, B.G.; Albano, C.M.; Anantharaman, R.; Beier, P.; Fargione, J.; Graves, T.A.; Gray, M.E.; Hall, K.R.; Lawler, J.J.; Leonard, P.B. et al. Circuit-Theory Applications to Connectivity Science and Conservation. Conserv. Biol.; 2019; 33, pp. 239-249. [DOI: https://dx.doi.org/10.1111/cobi.13230]

31. McRae, B.H.; Popper, K.; Jones, A.; Schindel, M.; Buttrick, S.; Hall, K.R.; Unnasch, B.; Platt, J.R. Conserving Nature’s Stage: Mapping Omnidirectional Connectivity for Resilient Terrestrial Landscapes in the Pacific Northwest; The Nature Conservancy: Portland, ON, USA, 2016; 47.

32. Rigge, M.; Bunde, B.; Meyer, D.K.; Shi, H.; Postma, K. Rangeland Condition Monitoring Assessment and Projection (RCMAP) Fractional Component Time-Series Across the Western U.S. 1985-2020. U.S. Geol. Surv. Data Release; 2021; [DOI: https://dx.doi.org/10.5066/P95IQ4BT]

33. Monroe, A.P.; Aldridge, C.L.; O’Donnell, M.S.; Manier, D.J.; Homer, C.G.; Anderson, P.J. Using Remote Sensing Products to Predict Recovery of Vegetation across Space and Time Following Energy Development. Ecol. Indic.; 2020; 110, 105872. [DOI: https://dx.doi.org/10.1016/j.ecolind.2019.105872]

34. U.S. Fish and Wildlife Service. Greater Sage-Grouse (Centrocercus Urophasianus) Conservation Objectives: Final Report; U.S. Fish and Wildlife Service: Denver, CO, USA, 2013; 38.

35. Doherty, K.; Theobald, D.M.; Bradford, J.B.; Wiechman, L.A. Biome-Wide Sagebrush Core Habitat and Growth Areas Estimated from a Threat-Based Conservation Design. A Sagebrush Conservation Design to Proactively Restore America’s Sagebrush Biome 2022–1081; U.S. Geological Survey: Reston, VA, USA, 2022.

36. Landau, V.; Shah, V.; Anantharaman, R.; Hall, K. Omniscape.Jl: Software to Compute Omnidirectional Landscape Connectivity. J. Open Source Softw.; 2021; 6, 2829. [DOI: https://dx.doi.org/10.21105/joss.02829]

37. Buchholtz, E.K.; O’Donnell, M.S.; Heinrichs, J.A.; Aldridge, C.L. Sagebrush Structural Connectivity Yearly and Temporal Trends Based on RCMAP Sagebrush Products, Biome-Wide from 1985 to 2020. U.S. Geol. Surv. Data Release; 2023; [DOI: https://dx.doi.org/10.5066/P9ED3OHH]

38. Bezanson, J.; Edelman, A.; Karpinski, S.; Shah, V.B. Julia: A Fresh Approach to Numerical Computing. SIAM Rev.; 2017; 59, pp. 65-98. [DOI: https://dx.doi.org/10.1137/141000671]

39. Hall, K.R.; Anantharaman, R.; Landau, V.A.; Clark, M.; Dickson, B.G.; Jones, A.; Platt, J.; Edelman, A.; Shah, V.B. Circuitscape in Julia: Empowering Dynamic Approaches to Connectivity Assessment. Land; 2021; 10, 301. [DOI: https://dx.doi.org/10.3390/land10030301]

40. Belote, R.T.; Barnett, K.; Zeller, K.; Brennan, A.; Gage, J. Examining Local and Regional Ecological Connectivity throughout North America. Landsc. Ecol.; 2022; 37, pp. 2977-2990. [DOI: https://dx.doi.org/10.1007/s10980-022-01530-9]

41. Falgout, J.T.; Gordon, J.; Williams, B.; Davis, M.J. USGS Advanced Research Computing, USGS Denali Supercomputer; U.S. Geological Survey: Reston, VA, USA, 2022.

42. Knick, S.T.; Connelly, J.W. Greater Sage-Grouse Ecology and Conservation of a Landscape Species and Its Habitats; University of California Press: Oakland, CA, USA, 2011; ISBN 978-0-520-26711-4

43. Ricca, M.A.; Coates, P.S. Integrating Ecosystem Resilience and Resistance Into Decision Support Tools for Multi-Scale Population Management of a Sagebrush Indicator Species. Front. Ecol. Evol.; 2020; 7, 493. [DOI: https://dx.doi.org/10.3389/fevo.2019.00493]

44. Loreau, M.; Mouquet, N.; Gonzalez, A. Biodiversity as Spatial Insurance in Heterogeneous Landscapes. Proc. Natl. Acad. Sci. USA; 2003; 100, pp. 12765-12770. [DOI: https://dx.doi.org/10.1073/pnas.2235465100]

45. Didham, R.K. Ecological Consequences of Habitat Fragmentation. eLS; John Wiley & Sons, Ltd.: Hoboken, NJ, USA, 2010; ISBN 978-0-470-01590-2

46. Haddad, N.M.; Brudvig, L.A.; Clobert, J.; Davies, K.F.; Gonzalez, A.; Holt, R.D.; Lovejoy, T.E.; Sexton, J.O.; Austin, M.P.; Collins, C.D. et al. Habitat Fragmentation and Its Lasting Impact on Earth’s Ecosystems. Sci. Adv.; 2015; 1, e1500052. [DOI: https://dx.doi.org/10.1126/sciadv.1500052]

47. Damschen, E.I.; Brudvig, L.A.; Burt, M.A.; Fletcher, R.J.; Haddad, N.M.; Levey, D.J.; Orrock, J.L.; Resasco, J.; Tewksbury, J.J. Ongoing Accumulation of Plant Diversity through Habitat Connectivity in an 18-Year Experiment. Science; 2019; 365, pp. 1478-1480. [DOI: https://dx.doi.org/10.1126/science.aax8992]

48. McGuire, J.L.; Lawler, J.J.; McRae, B.H.; Nuñez, T.A.; Theobald, D.M. Achieving Climate Connectivity in a Fragmented Landscape. Proc. Natl. Acad. Sci. USA; 2016; 113, pp. 7195-7200. [DOI: https://dx.doi.org/10.1073/pnas.1602817113]

49. Cameron, D.R.; Schloss, C.A.; Theobald, D.M.; Morrison, S.A. A Framework to Select Strategies for Conserving and Restoring Habitat Connectivity in Complex Landscapes. Conserv. Sci. Pract.; 2022; 4, e12698. [DOI: https://dx.doi.org/10.1111/csp2.12698]

50. Cross, T.B.; Schwartz, M.K.; Naugle, D.E.; Fedy, B.C.; Row, J.R.; Oyler-McCance, S.J. The Genetic Network of Greater Sage-Grouse: Range-Wide Identification of Keystone Hubs of Connectivity. Ecol. Evol.; 2018; 8, pp. 5394-5412. [DOI: https://dx.doi.org/10.1002/ece3.4056] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29938061]

51. Zimmerman, S.J.; Aldridge, C.L.; Hooten, M.B.; Oyler-McCance, S.J. Scale-Dependent Influence of the Sagebrush Community on Genetic Connectivity of the Sagebrush Obligate Gunnison Sage-Grouse. Mol. Ecol.; 2022; 31, pp. 3267-3285. [DOI: https://dx.doi.org/10.1111/mec.16470] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35501946]