The importance of within-year repeated counts and the influence of scale on long-term monitoring of sage-grouse

Long‐term population monitoring is the cornerstone of animal conservation and management. The accuracy and precision of models developed using monitoring data can be influenced by the protocols guiding data collection. The greater sage‐grouse (Centrocercus urophasianus) is a species of concern that has been monitored over decades, primarily, by counting the number of males that attend lek (breeding) sites. These lek count data have been used to assess long‐term population trends and for multiple mechanistic studies. However, some studies have questioned the efficacy of lek counts to accurately identify population trends. In response, monitoring protocols were changed to have a goal of counting lek sites multiple times within a season. We assessed the influence of this change in monitoring protocols on model accuracy and precision applying generalized additive models to describe trends over time. We found that at large spatial scales including >50 leks, the absence of repeated counts within a year did not significantly alter population trend estimates or interpretation. Increasing sample size decreased the model confidence intervals. We developed a population trend model for Wyoming greater sage‐grouse from 1965 to 2008, identifying significant changes in the population indices and capturing the cyclic nature of this species. Most sage‐grouse declines in Wyoming occurred between 1965 and the 1990s and lek count numbers generally increased from the mid‐1990s to 2008. Our results validate the combination of monitoring data collected under different protocols in past and future studies—provided those studies are addressing large‐scale questions. We suggest that a larger sample of individual leks is preferable to multiple counts of a smaller sample of leks.