Evaluating post-fire recovery of biological soil crusts and desert ecosystem services

Background/Question/Methods

Biological soil crusts are complex mosaics of cyanobacteria, green algae, lichens, mosses, microfungi, and other bacteria and may constitute up to 70% of living biomass in desert ecosystems. Within these crusts, cyanobacteria: glue soil particles together and create a matrix that protects soil surfaces from wind and water erosion; improves ecohydrology; and provides N to plants through N fixation processes. Unfortunately, crusts and the ecosystem services they provide are becoming more susceptible to fire as exotic annual grass invasions are facilitating the spread of catastrophic wildfires. Despite this, our understanding of crust recovery following fire is limited. To evaluate how fast crusts recover from fire, we created burn manipulations and tracked crust form and function through time in a cold desert ecosystem (UT, USA). Specifically, in burned and unburned treatments, we: monitored the cover of crust species; quantified the biomass of active bacteria using reverse transcriptase quantitative PCR of 16S rRNA; measured nitrogen fixation potential with acetylene reduction assays; and assessed soil infiltration rates. Crust dynamics were evaluated before and after the fire (i.e., 1 week and 2 months post-fire) and will continue to be measured seasonally for the next three years. 

Results/Conclusions

We have found no evidence that crust form or function has recovered two months after the fire. Based on cover estimates, none of the 15 species of lichens or mosses survived the fire but burnt crusts remained attached to soil surfaces and total crust cover did not decline. The activity of total soil bacteria in crusts (i.e., soil depth from 0-2 mm) decreased at least an order of magnitude in burnt compared to control soils from barren interspaces and beneath sagebrush. Nitrogen fixation potentials decreased by at least ten-fold in burnt than undisturbed crusts, suggesting a reduction in soil inorganic N availability for recovering or establishing plant species. Additionally, soil infiltration rates drastically declined in burnt soils one week after the fire and continue to be depressed after two months. For example, infiltration rates decreased from 1.0 to -3.0 mL H₂O per second in burnt compared to undisturbed crusts, thus providing evidence for the development of a new fire-induced hydrophobic layer. Our results demonstrate that fire strongly destabilizes crusts and crust-mediated processes, and we expect our future findings to produce a timeline indicating when crust ecosystem services will recover and, once again, benefit overall ecosystem health.