Greater Sage-Grouse Population Monitoring Framework: Frequently Asked Questions
The Greater Sage-grouse Population Monitoring Framework helps aid current assessments of sage-grouse population trends across spatial and temporal scales. It centers on four objectives: 1) create a standardized database of lek counts; 2) cluster leks to develop spatial population structures; 3) estimate spatial trends across temporal extents; and 4) develop a system to estimate where and when management action is likely to benefit declining populations of sage-grouse at the appropriate spatial scale on an annual basis.
The framework comprises three lines of evidence to help managers estimate past, present, and future population performance:
1. Trends Analysis – Understanding the Past
The trends analysis provides evidence of past population performance.
2. Targeted Annual Warning System (TAWS) – Tracking the Present
The TAWS provides evidence of present population performance.
3. Extirpation Probabilities – Anticipating the Future
The extirpation probabilities provide evidence of future population performance.
Questions related to the range-wide standardized lek database
- Who manages sage-grouse lek data?
- What type of data is in this database?
- How does standardized lek data affect management decisions?
- Why do leks appear in some years and not other years?
- Why are lek locations different between state databases and the range-wide database?
- How often is the range-wide database updated?
- Do aggregation of leks and associated male counts occur, and how will this affect the use of the data?
- Where can we find the sage-grouse standardized lek database?
Questions related to spatial clusters
- Why do some neighborhood clusters include a single or few leks while others may include 20 to 30 leks?
- How do the neighborhood and climate clusters differ from other sage-grouse population units?
- Why do clusters cross state boundaries?
- How should managers use the clusters when they include leks and associated male counts in neighboring states? What are the ramifications of not including population data in neighboring states?
- Where can we find the sage-grouse clusters?
Questions related to population trends and extirpation
- What time period of lek data was used in the analysis and why
- How do leks with no observations for a given year inform the trends?
- What does nadir mean and why is this relevant to understanding sage-grouse population trends?
- Why do sage-grouse populations seemingly cycle every 6–12 years? Is this important to managers and if so, how?
- What is lambda and how should managers interpret lambda values? What range of lambda values warrant concern based on the model’s confidence in those values?
- Should managers treat trends in different periods differently? How are trends from historic cycles relevant to managers today?
- Are there trend estimates at other cluster levels between 2 and 13?
- Are your trend estimates different than those published by state agencies?
- Where and when can we find these data/results?
Questions related to the Targeted Annual Warning System (TAWS)
- Why did you use fewer years of data in the TAWS than the trends analysis?
- Why do updated versions of TAWS result in different sequences of watches/warnings for the same lek/neighborhood?
- How do watches and warnings stop?
- Can a lek or neighborhood cluster have a declining population and not have a signal in the TAWS?
- Can a population signal when it is stable or increasing?
- Almost every lek in my area of interest has signaled at some point since the start of the TAWS (1990). How do I consider the timing of watches/warnings for present day management decisions?
- How do warnings and watches at the lek and neighborhood scale relate to scales of management action?
- Should watches/warnings at the lek and neighborhood be treated the same?
- Do neighborhood clusters with few leks affect TAWS?
- Where and when can we find these data/results?
Range-wide standardized lek database
Q. Who manages sage-grouse lek data?
A. State and federal wildlife agencies collect sage-grouse lek count data. The range-wide lek database is managed by the U.S. Geological Survey and Colorado State University in partnership with the state wildlife agencies.
Q. What type of data is in this database?
A. The U.S. Geological Survey and Colorado State University retain state data required for developing and maintaining the population monitoring framework that informs trends and targeted annual warning system. Please refer to the lek database cheat sheet for a list of columns, definitions, and domain-values used for the standardized database.
Q. How does standardized lek data affect management decisions?
A. The standardized lek database was developed to coalesce individual state data based on standard definitions and domain values, which, for most states, does not affect management decisions. First, our definition of a lek that might affect a state’s use of this database (≥2 displaying males observed for ≥2 years), particularly for states with populations on the periphery of the sage-grouse population range. Second, there were several instances where two states monitored the same lek (typically, these observations were historic), and we combined the observations to reflect a single lek (the lek was renamed to include both originating lek names). Third, we aggregate leks and counts of males when leks occur within 500 meters, as determined at the beginning of the project and agreed upon by all states. These modifications were intended to ensure a common monitoring and reporting framework implemented by most states and to remove data (for example, leks with too few observations) that would present problems during the development of clusters, trends, and the targeted annual warning system. Please refer to the lek database cheat sheet for a list of columns, definitions, lek aggregation, and domain values used for the standardized database.
Q. Why do leks appear in some years and not other years?
A. States revise their lek databases each year, for reasons including changes in lek names, updated lek locations for accuracy or because a lek moved, and possibly redaction of information from private landowners. Because methods of aggregating leks and counts of males are dynamic (based on leks with greatest number of males in last 20 years for all leks within 500 meters), the location and lek name may change. Rules used for retaining data for modeling can result in the exclusion of leks due to a lack of sufficient data for the trends and targeted annual warning system. Please refer to the lek database cheat sheet for details on rules used for inclusion of data in the lek database, trends, and targeted annual warning system.
Q. Why are lek locations different between state databases and the range-wide database?
A. Some leks are excluded because they lack the appropriate data to estimate trends and calculate signals in the Targeted Annual Warning System. Additionally, lek locations may change naturally from year to year or vary based on how satellite leks are aggregated.
Q. How often is the range-wide database updated?
A. Data are collected from states and standardized on an annual basis.
Q. Do aggregation of leks and associated male counts occur, and how will this affect the use of the data?
A. Yes, aggregation of leks occurs when lek locations are within 500 meters of each other; the location and male counts are aggregated. Please refer to the lek database cheat sheet for an overview of these rules.
Q. Where can we find the sage-grouse standardized lek database?
A. These data are managed by states and protected to ensure sage-grouse breeding sites are not disturbed. Access to these data must be requested directly through each state wildlife management agency.
Spatial Clusters
Q. Why do some neighborhood clusters include a single or few leks while others may include 20 to 30 leks?
A. The connectivity of leks and the clustering rules affect the number of leks per neighborhood. These rules are intended to identify the frequency of movements between leks based on habitat conditions, isolation by distance, and features impeding movements. Please refer to the lek database cheat sheet for an overview of these rules.
Q. How do the neighborhood and climate clusters differ from other sage-grouse population units?
A. Many spatial datasets are currently being used to manage sage-grouse populations. These include population and management units. A discussion about some of these datasets and how they differ from the population clusters (O’Donnell and others, 2022) used for the targeted annual warning system is provided (O’Donnell and others, 2019). The neighborhood and climate clusters are biologically informed, nested, and account for all life stages, enabling robust methods for comparing local and regional population changes (see the greater sage-grouse population monitoring framework website for a general description or details from the publication (O’Donnell and others, 2022). More recent products not evaluated in O’Donnell and others (2019) include the Bureau of Land Management (BLM) habitat assessment framework (documentation), which most closely shares similarities with our clusters but may lack a less rigorous biological context for sage-grouse (in other words, geared toward habitat more than population structure) and does not consider nesting of populations that support the comparison of local and regional population changes.
Q. Why do clusters cross state boundaries?
A. The greater sage-grouse subpopulations (clusters) cross state boundaries because the connectivity of leks indicated movements were possible based on biologically relevant context. Habitat conditions summarized at different spatial scales surrounding these leks indicated similar habitat resources available to birds, regardless of jurisdictional boundaries.
Q. How should managers use the clusters when they include leks and associated male counts in neighboring states? What are the ramifications of not including population data in neighboring states?
A. State managers can use the targeted annual warning system (TAWS; signals assigned to leks and neighborhood clusters when populations decline) to manage sage-grouse populations by considering neighborhood clusters crossing state boundaries. Birds may move between leks across state boundaries, affecting observations of males counted on leks. These data, together, can help explain signals at the neighborhood versus lek level and what might be causing signals (for example, if the state of habitat conditions in neighboring state has worsened).
Q. Where can we find the sage-grouse clusters?
A.The clusters are available for download as shapefiles on the U.S. Geological Survey’s ScienceBase: https://doi.org/10.5066/P9D1K0LX
Population trends and extirpation
Q. What time period of lek data was used in the analysis and why?
A. We included observation data recorded since 1960 because states were more consistently monitoring sage-grouse by that time, and the longer history provides greater context for population changes, which is important for understanding and identifying historic nadirs linked to population cycles.
Q. How do leks with no observations for a given year inform the trends?
A. We included leks if there were more than 5 years of active counts (greater than or equal to two males) during the 60-year time series. The state-space model imputes abundance estimates for each lek in each year. Abundance estimates for years with missing counts are generated from samples drawn from the probability distributions that define annual rates of population change at the neighborhood and climate-cluster levels. These estimates represent the average performance across all leks within those bounds. Leks with missing count data do not contribute information for those years.
Q. What does nadir mean and why is this relevant to understanding sage-grouse population trends?
A. The nadir of a population cycle is when the population reaches its minimum size before it increases again. Understanding when a population nadir occurs is important for correctly evaluating population trends from growth rates (nadir-to-nadir). Evaluating population growth rates between a population peak and a population nadir at some future time will incorrectly identify the steepness of change (slope) in the growth rate and, therefore, can increase challenges to wildlife managers for current and future policies and management actions.
Q. Why do sage-grouse populations seemingly cycle every 6–12 years? Is this important to managers and if so, how?
A. Sage-grouse populations cycle with natural patterns in precipitation. Because precipitation drives vegetation conditions that support foraging, nesting and brood-rearing, populations tend to increase in wetter periods and decline in drier ones. Accounting for these cycles is important, so managers do not inaccurately estimate population performance by comparing a natural peak to a natural nadir within a given cycle.
Q. What is lambda and how should managers interpret lambda values? What range of lambda values warrant concern based on the model’s confidence in those values?
A. Lambda (λ) values reflect a population's finite rate of increase or growth factor for a given period. It is based on the ratio of population size between Nt+# and Nt. When λ > 1, the population is increasing, when λ = 1, there is a stable population, and when λ < 1, the population is declining. The TAWS can help managers identify when population declines are unexpected and may warrant management intervention.
Q. Should managers treat trends in different periods differently? How are trends from historic cycles relevant to managers today?
A. Yes, trends should be interpreted according to the period that they reference. More recent trends will better reflect current conditions in the landscape. However, interpreting recent trends in the context of historical trends improves overall understanding of trajectories. Stable-to-increasing trends are a management target. However, declining trends that are not declining as sharply as prior, historical trends can be viewed as an improvement and may be evidence of effective management. Historic trends help situate recent trends within a broader context of management effectiveness.
Q. Are there trend estimates at other cluster levels between 2 and 13?
A. No. The neighborhood cluster (level 2) was identified as most appropriate for capturing inter-lek movements of greater sage-grouse. These represent ‘closed’ populations at scales amenable to management actions, meaning there is little or no movement in or out of these clusters. The climate cluster (level 13) was identified as the most appropriate scale where climatic patterns closely correlate with population dynamics and support high-level trend summaries.
Q. Are your trend estimates different than those published by state agencies?
A. The methods used to estimate sage-grouse populations vary across states. The USGS Open-File Report by Coates and others (2021) outlines a standard approach for estimating trends across geographic scales and time periods using data collected by each state. The report provides comparisons to states that had documented greater sage-grouse population trends, which were similar after adjusting the modeling periods. The report appendix provides trends for each state.
Q. Where and when can we find these data/results?
A. The latest trends results are available for download as a shapefile (neighborhood and climate clusters) on the U.S. Geological Survey’s ScienceBase: https://doi.org/10.5066/P9OQWGIV. Data are updated annually and released around February of each year. Check ScienceBase for the most recent results!
Targeted annual warning system (TAWS)
Q. Why did you use fewer years of data in the TAWS than the trends analysis?
A. Data collection efforts were less robust between 1960–1990 compared to 1990–Present. The primary difference between trends and TAWS analyses is the temporal scales of inference. The TAWS makes inferences at fine temporal scales (3–5-year window with annual calculations), which means it requires more data. The trend analysis by comparison makes inferences across multiple decades, which allows a greater amount of missing data.
Q. Why do updated versions of TAWS result in different sequences of watches/warnings for the same lek/neighborhood?
A. The methods used for modeling lek count data have remained relatively unchanged throughout, except for minor improvements that were implemented between the original OFR (data spanning 1960–2019) with only minor methodological adjustments, and that most differences in updated TAWS stems from state-provided data updates (e.g., revised historical counts, corrected lek locations, or added missing observations).
Q. How do watches and warnings stop?
A. Watches and warnings stop when the population stabilizes, increases, or re-couples with the climate cluster for one or multiple years in a row. If the population declines faster than the climate cluster for multiple consecutive years, it can trigger a watch/warning again.
Q. Can a lek or neighborhood cluster have a declining population and not have a signal in the TAWS?
A. Yes. A signal occurs when a population is declining, and the decline is decoupled from the climate cluster trend. If a lek or neighborhood declines at the same rate as or slower than the climate cluster, it will not signal.
Q. Can a population signal when it is stable or increasing?
A. No. A population must be declining to signal. However, because the TAWS makes inferences over a temporal moving window, there can be lags in signals, which may result in a watch/warning during a year of population growth.
Q. Almost every lek in my area of interest has signaled at some point since the start of the TAWS (1990). How do I consider the timing of watches/warnings for present day management decisions?
A. TAWS watches/warnings signify aberrant declines (declining faster than the climatically induced decline) for a specific time and location. When those signals stop occurring, it signifies a period of population growth, stability, or coupling (with the climate cluster). If a population had watches/warnings historically, but does not exhibit them at the present time, it can be assumed that the perturbation has ceased to exist in that area, the population is no longer susceptible to the perturbation, or the population is on a natural upward swing in the regularly occurring oscillations. When the latter occurs, it may require several years of data collection to determine whether the population is still being negatively impacted (i.e., when it enters a natural declining period of oscillation and may decline faster than the climate cluster).
Q. How do warnings and watches at the lek and neighborhood scale relate to scales of management action?
A. Lek watches/warnings will have greater variability than the neighborhood scale due to the open population structure that they exhibit, but at the lek scale, managers can more easily pinpoint what is going on in the field (e.g., geothermal plant disturbance at local site) and identify local changes that might be causing populations to decline. Neighborhood watches/warnings are useful because, if leks are declining relative to populations in the climate region, managers may need to consider landscape-scale factors (e.g., fire, local drought, or large loss of sagebrush) that might be driving the decoupling. Furthermore, neighborhood-scale movement analyses indicate that these populations were minimally affected by migration. Thus, changes in population size can be more directly linked to births and deaths.
Q. Should watches/warnings at the lek and neighborhood be treated the same?
A. Watches/warnings at lek and neighborhood cluster levels represent population responses to stimuli operating at unique spatial scales. Watches/warnings at the lek scale will occur when small and large perturbations occur near that lek. Watches/warnings at the neighborhood scale require many leks to be affected, simultaneously (within a 3–5-yr window), to push the entire neighborhood in a downward trajectory that outpaces natural decline (e.g., climatically induced declines). Therefore, watches/warnings at the neighborhood scale will be in response to large perturbations only.
Q. Do neighborhood clusters with few leks affect TAWS?
A. The number of leks and the average size of leks within a neighborhood cluster have little-to-no impact on the rate at which a cluster experiences watches or warnings.
Q. Where and when can we find these data/results?
A. The latest trend results are available for download as a shapefile (neighborhood and climate clusters) on the U.S. Geological Survey’s ScienceBase: https://doi.org/10.5066/P9OQWGIV. Data are updated annually and released around February each year. Check ScienceBase for updated results!
Co-production
We continue working with all collaborators to improve sage-grouse management tools. Each year, we develop a new standardized database to include newly digitized historical data and improve data quality using rigorous quality control methods. These data are incorporated into the TAWS framework to produce results that are delivered in time for annual agency decision making.
Data restrictions
State wildlife agencies collect and manage lek databases. Because sage-grouse are a species of conservation concern and sensitive to activities during breeding, these data are available only after acquiring data-sharing agreements with individual states.
Funders
U.S. Geological Survey (Ecosystem Mission Area, land management research program and species management research program, Wyoming Landscape Conservation Initiative) and U.S. Bureau of Land Management.
Partners
State Wildlife Agencies (California Department of Fish and Wildlife; Colorado Parks and Wildlife; Idaho Department of Fish and Game; Montana Fish, Wildlife & Parks; Nevada Department of Wildlife; North Dakota Game and Fish Department; Oregon Department of Fish and Wildlife; South Dakota Department of Game, Fish and Parks; Utah Division of Wildlife Resources; Wyoming Game and Fish Department; Washington Department of Fish and Wildlife), Colorado State University, BLM, US Fish and Wildlife Service, US Forest Service, researchers who provided field data to evaluate results.
Data Harmonization for Greater Sage-Grouse Populations
Greater Sage-Grouse Population Monitoring Framework
A targeted annual warning system (TAWS) for identifying aberrant declines in greater sage-grouse populations
Estimating trends for greater sage-grouse populations within highly stochastic environments
Hierarchical Units of Greater Sage-Grouse Populations Informing Wildlife Management
Data Harmonization for Greater Sage-Grouse Populations
Greater Sage-Grouse Population Monitoring Framework
A targeted annual warning system (TAWS) for identifying aberrant declines in greater sage-grouse populations
Estimating trends for greater sage-grouse populations within highly stochastic environments