Benefits of thinning and burning for understory diversity vary with spatial scale and time since treatment

Fuel-reduction treatments have been used effectively in dry, fire-adapted forests to reduce risk of high-severity crown fire, but it is less certain whether they achieve ecological objectives such as promoting understory diversity. Using long-term data from a fuel-reduction experiment, we tested how conclusions about treatment effectiveness are related to the spatial and temporal scales over which ecological responses are evaluated. We modeled the richness of herbs and shrubs at smaller (1-50 m2) and larger (1000 m2) spatial scales as functions of pre-treatment richness, thinning intensity, burning, and time since treatment (2-3 vs. 10-13 years). Pre-treatment richness was the strongest predictor of post-treatment richness, reflecting high rates of species survival, irrespective of treatment. Richness was enhanced by burning and, to a lesser extent, by thinning, although the timing and strengths of these effects varied with spatial scale. Among herbs, annuals showed an early and persistent increase after burning at the larger scale, but a lagged response at the smaller scale-the latter enabled by limited competition from perennial herbs and shrubs. In contrast, perennials showed lagged responses to thinning at smaller scales and to burning at larger scales, suggesting slow recruitment via seed or vegetative spread. Non-natives were unresponsive to treatments but were also uncommon before treatment, suggesting limited propagule pressure from the surrounding landscape. Rates of colonization were unrelated to pre-treatment richness; thus, the potential for understory enrichment was not constrained by initial diversity. The low cover and lagged responses of woody and herbaceous perennials suggest that further enrichment is possible, particularly at smaller spatial scales. Our results illustrate that uncertainty about the ecological effectiveness of fuels treatments can relate to the spatial and temporal scales over which responses are measured. They underscore the value of long-term, multi-scale assessments in defining the spatial and temporal contexts of ecological outcomes, and in evaluating the necessity for, or timing of, further intervention.