Fire is a global disturbance that is predicted to increase in frequency and severity in many parts of the world due to climate change. As such, there is an increasing need to understand how ecosystem components respond to this disturbance. As fires are heterogenous across at local and global scales, it is often difficult to describe the overarching effects of fire on ecological communities. Furthermore, many communities are often overlooked in fire studies despite having a substantial effect on ecological function and the adjacent communities. One such community is biological soil crust (biocrust). Biocrusts are composed of micro and macro-organisms bound to the soil surface and are thus directly exposed to the heat of the fire. These communities also interact directly and indirectly with the surrounding vegetation by modulating seed survival, nutrient cycling, and soil stability. The goal of this study is to synthesize and analyze existing data elucidating the recovery of biocrust and vegetation cover following fire at the global-scale and suggest avenues for future research. We performed a meta-analysis of studies from 1900-2019 to specifically address the response of biocrust and their associated plant communities after fire and determine the moderating factors governing their response.
Results/Conclusions
We obtained 289 biocrust experiments from 17 unique studies. Five of those studies also measured plant cover resulting in 326 vegetation experiments. Biocrust and plant communities responded differently to fire. Although biocrust cover was reduced by 74% (P<0.001), this result is confounded by high between study heterogeneity indicating biocrust response to fire may be highly site specific. The plant communities in this study were not affected by fire, with notable exceptions. Exotic, annual grasses responded negatively (P<0.001) and perennial grasses, regardless of native status, responded positively to fire (P<0.001). Time since fire, fire type, and climate were not significant moderators for either community. Finally, although this was a “global analysis”, 81% of the studies occurred in the United States and all the studies were in either grasslands or shrublands. This is not characteristic of the global distribution of biocrusts and is thus particularly biased towards North American shrublands. Given this, we suggest future research should: 1) increase the number of studies in underrepresented ecosystems; 2) determine the relationship between vegetation fire characteristics and the biocrust community; 3) determine the appropriate scale to test these relationships and 4) consistently report fire severity, frequency, and history in ecological fire studies.