Optimizing landscape-scale monitoring programmes to detect the effects of megafires

Abstract

Aim: By virtue of their spatial extent, landscape-scale monitoring programmes may be well-suited to support before-after/control-impact (BACI) studies of major disturbances such as megafires, but they may only have a few affected sites, reducing statistical power. I tested whether a variety of hypothetical but logistically plausible survey designs could provide statistically powerful assessments of hypothetical but empirically based population responses to megafire. Location: Simulation focused on fire-prone forests. Methods: I varied the number of affected sites, detection probability, baseline occupancy and species’ responses to fire to create 72 monitoring scenarios. I then simulated 10 years of site occupancy data in which a megafire or other disturbance occurred between years 5 and 6 (n = 500 iterations). Results: Statistical power to correctly reject the null hypothesis of no population response to megafire was high (mean across all scenarios = 0.78), but power to identify the correct population response (e.g. post-fire occupancy declined and then recovered) was low (mean = 0.29). Statistical power to not underestimate the effect of a megafire on site occupancy was fairly high (mean = 0.66), but power to accurately estimate site occupancy was low (mean = 0.25). Statistical power increased with the number of affected sites (i.e., sample size) and with the intensity of the focal species’ response to megafire. Case study simulations based on an existing acoustic monitoring programme in the Sierra Nevada, USA, indicate that it is likely to identify Spotted Owl and Black-backed Woodpecker population responses to megafires. Researchers can use the included simulation tutorial to implement their own power analyses. Main conclusions: Landscape-scale monitoring programmes can identify population changes following megafires, but cannot reliably produce nuanced results, especially with only five years of post-fire data. Smaller-bodied species, which could be studied at more sites, are likely to be better focal species for megafire BACI studies. Aggregating landscape-scale studies’ sampling coverage across many fire footprints to study the overall effects of megafires—rather than the effects of individual fires—may be a more informative approach.