Abstract Understanding drivers of habitat loss and fragmentation can inform conservation policy and action. However, drivers vary across space and time, and complete datasets of potential drivers that align with time series of species habitat are rare. We developed a 38‐year time‐series of nesting/roosting habitat for the California spotted owl in the Sierra Nevada, CA, USA, and used a novel disturbance dataset to attribute habitat changes to potential drivers. We found that the total area of high‐quality nesting/roosting habitat declined by 43% between 1985 and 2022. Moreover, the average high‐quality habitat patch became 68% smaller and 33% more isolated over the study period. During 2002–2022, drought was the most abundant disturbance in spotted owl habitat (49% of habitat affected), followed by high‐severity fire (32%) and fuels management (20%). However, the amount of disturbance did not always match impact. Drought, while most common, was responsible for only 20% of estimated habitat losses. High‐severity fire was responsible for nearly half (45%) of habitat loss, while fuels management was responsible for only 9% of habitat loss (and 14% of habitat gains). Dominant drivers of habitat loss varied spatially, suggesting the potential to use local information to calibrate conservation action. Previously, the absolute and relative impacts of fuels management, severe wildfires, drought, and other disturbance agents on spotted owl habitat were unknown. Here, we demonstrated that the most influential factor driving losses of high‐quality spotted owl habitat was high‐severity fire, and that fuels management, while associated with relatively smaller habitat losses, was also associated with habitat gains, suggesting net benefits of fuels management via its effect of reducing high‐severity fire. We demonstrate the power of using dynamic species distribution models to map and track habitat over time and understand the factors that may be contributing to observed habitat losses and gains.
Hart et al. (Tue,) studied this question.