U.S. Joint Fire Science Program


Date of this Version


Document Type



Final Report for Project #06-1-2-02


US government work.


Area burned has increased during the past few decades in the Mojave Desert due in part to increased dominance of highly flammable invasive non-native annual grasses. Management responses such as post-fire seedings have been implemented during the first 3 post-fire years to suppress the growth of the invasive annual grasses, promote recovery of native species, and facilitate the restoration of plant species diversity and abundance. Although there is a fair amount of information available on the effects of fire on plant diversity, density, and cover, there is very little information available regarding effects on soil seed banks to help guide the development of management prescriptions. This project was designed to evaluate the short-term effects of fire on soil seed bank diversity and density, and vegetation diversity and cover, following the Hackberry Fire Complex of summer 2005 in the eastern Mojave Desert. A secondary objective was to evaluate the correlations between measures of burn severity and seed bank and vegetation abundance to evaluate the utility of burn severity metrics in evaluating fire effects. The study region encompasses upper elevation blackbrush and lower elevation sagebrush ecotones of the Mojave Desert. Fire reduced soil seed bank diversity during the first two post-fire fall seasons, although evenness was slightly higher in burned areas during all three post-fire years, possibly due to loss of annual plant microhabitats previously created by shrub canopies. Fire also reduced seed bank density by 81%, but only during the first post-fire spring. Seed bank reductions were greater for non-natives, Bromus rubens in particular, than for natives. Aboveground vegetation diversity was reduced in burned areas during all three post-fire years due to declines in species richness of perennials, as native species richness was not affected. Fire reduced cover of perennials and increased cover of annuals during all three years, but fire did not affect cover of non-native annual grasses (Bromus rubens, Bromus tectorum, and Schismus spp.). Virtually all of the seed bank and annual plant vegetation metrics evaluated in this study returned to unburned condition by the second or third post-fire years, and varied more among years than due to burning. In addition, the effects of fire on seed bank density during the first year were over an order of magnitude higher than what typically seeding prescriptions would have replaced if they had been implemented. These results call into question the need to seed annual plant species after fires in the Mojave Desert. In contrast, persistent reductions in cover of perennials means that their seed sources were limited and post-fire seedings may have help to overcome this establishment limitation for those species, although further studies are needed to evaluate this dynamic. Both dNBR and CBI burn severity metrics were negatively correlated with total vegetation cover, annual cover in particular, during the first post-fire spring, which appeared to carry over to the seed bank during following fall which was also negatively correlated with dNBR. It therefore appears that dNBR may be a potentially useful tool in estimating reduced cover of annuals during the first post-fire spring, and reduced seed bank density during the following fall.