Hierarchically nested and biologically relevant range-wide monitoring frameworks for greater sage-grouse, western United States
Dates
Publication Date
2022-12-01
Time Period
2019
Citation
O’Donnell, M.S., Edmunds, D.R., Aldridge, C.L., Heinrichs, J.A., Monroe, A.P., Coates, P.S., Prochazka, B.G., Hanser, S.E., and Wiechman, L.A., 2022, Hierarchically nested and biologically relevant range-wide monitoring frameworks for greater sage-grouse, western United States: U.S. Geological Survey data release, https://doi.org/10.5066/P9D1K0LX.
Summary
We produced 13 hierarchically nested cluster levels that reflect the results from developing a hierarchical monitoring framework for greater sage-grouse across the western United States. Polygons (clusters) within each cluster level group a population of sage-grouse leks (sage-grouse breeding grounds) and each level increasingly groups lek clusters from previous levels. We developed the hierarchical clustering approach by identifying biologically relevant population units aimed to use a statistical and repeatable approach and include biologically relevant landscape and habitat characteristics. We desired a framework that was spatially hierarchical, discretized the landscape while capturing connectivity (habitat and movements), and [...]
Summary
We produced 13 hierarchically nested cluster levels that reflect the results from developing a hierarchical monitoring framework for greater sage-grouse across the western United States. Polygons (clusters) within each cluster level group a population of sage-grouse leks (sage-grouse breeding grounds) and each level increasingly groups lek clusters from previous levels. We developed the hierarchical clustering approach by identifying biologically relevant population units aimed to use a statistical and repeatable approach and include biologically relevant landscape and habitat characteristics. We desired a framework that was spatially hierarchical, discretized the landscape while capturing connectivity (habitat and movements), and supported management questions at different spatial scales. The spatial variability in the amount and quality of habitat resources can affect local population success and result in different population growth rates among smaller clusters. Equally so, the spatial structure and ecological organization driving scale-dependent systems in a fragmented landscape affects dispersal behavior, suggesting inclusion in population monitoring frameworks. Studies that compare conditions among spatially explicit hierarchical clusters may elucidate the cause of differing growth rates at local scales affected by changes in habitat quality compared to larger scaled processes affecting growth rates, such as regional climate/vegetation communities. Therefore, the use of multiple scales (hierarchical cluster levels) that group demographic data can provide information driving population changes at different spatial scales, thereby providing a tool for population monitoring and adaptive management.
O’Donnell, M. S., D. R. Edmunds, C. L. Aldridge, J. A. Heinrichs, A. P. Monroe, P. S. Coates, B. G. Prochazka, S. E. Hanser, and L. A. Wiechman. 2022. popcluster: hierarchical population monitoring frameworks (version 2.0.0): U.S. Geological Survey software release, https://doi.org/10.5066/P9X68ADU.
When considering range-wide declines in sage-grouse populations and regional variability of population sizes, temporal and spatial scale are essential considerations. Further, targeted management actions are needed at spatial scales that align with factors causing population change. There is a need to understand mechanisms driving population changes, allowing targeted management actions to conserve populations. Yet, to our knowledge, repeatable multi-scaled and biologically-informed methods to support population monitoring of sage-grouse have yet to be developed. We developed a biologically-informed approach to clustering habitat and population units to improve opportunities for multi-scale monitoring and evaluation of broadly distributed populations using a spatially balanced framework. In doing so, this hierarchical monitoring framework can be used 1) as a long-term population monitoring framework for greater sage-grouse (Centrocercus urophasianus), 2) to track the outcomes of local and regional populations by comparing population changes across scales (hierarchical levels), and 3) to inform where to best spatially target studies that identify the processes and mechanisms causing population trends to change among spatial and temporal scales.
When using these data to evaluate population changes or identify a spatially balanced sampling protocol, all cluster levels are designed to work together, as cluster levels are spatially related and nested. Therefore, we recommend evaluating multiple cluster levels, as different factors affect populations at different spatial scales. If a single scale is desired when analyzing population growth rates or other analyses, as these data are intended for multi-scale efforts, allow your data to decide which scale(s) are appropriate. The finest scale level (level 1) represents a conditioning period for the clustering process and was not intended for use in population assessments (sage-grouse movements beyond population units largest). These cluster levels are specific to greater sage-grouse, but they may be appropriate for other sagebrush obligate species, and the user will need to make this determination. The methods used for developing this hierarchical population monitoring framework can also be applied to other species with high site fidelity to geographic locations where demographic data are collected (for example, breeding/calving sites and nesting sites).
These data reflect Thiessen polygons around greater sage-grouse breeding sites (leks), which are incrementally grouped from fine-scaled (fewest number of grouped leks) to coarsest-scaled (greatest number of grouped leks) cluster levels. The lek data have an accuracy of approximately +/- 30 m (1:240,000) and therefore the Thiessen polygons will have a similar accuracy. However, these boundaries are not defined by landscape characteristics (e.g., ridges), but rather equal distance between grouped leks. The products from this study aim to support multiple research and management needs. These data replace the interim products developed in Nevada and Wyoming (https://doi.org/10.5066/P9J0B7JR).
Several intended uses of these sampling units include the following:
1. The sampling units can inform a spatially balanced monitoring framework (for example, Generalized Random Tessellation Stratified sampling framework). The framework can inform rotating sampling designs of known leks to monitor as well as sampling designs to systematically search for unknown leks on the landscape.
2. The sampling units can inform groupings of population counts, which in turn can be used to evaluate population growth rates across cluster levels and time. For example, we have used these population units to assess greater sage-grouse population trends across the range and history of monitoring (1950s to present) and to inform a targeted annual warning system for adaptive management of habitat and populations.
3. The sampling units can inform the examination of relationships between the groupings of quantified landscape changes and population changes.
4. The sampling units can inform the partitioning of the landscape for developing seasonal habitat model development.
