Biological soil crusts (biocrusts) are topsoil communities of organisms that contribute to soil fertility and erosion resistance in drylands. Anthropogenic disturbances can quickly damage these communities and their natural recovery can take decades. With the development of accelerated restoration strategies in mind, I studied physiological mechanisms controlling the establishment of cyanobacteria in biocrusts, since these photoautotrophs are not just the biocrust pioneer organisms, but also largely responsible for improving key soil attributes such as physical stability, nutrient content, water retention and albedo. I started by determining the cyanobacterial community composition of a variety of biocrust types from deserts in the Southwestern US. I then isolated a large number of cyanobacterial strains from these locations, pedigreed them based on their 16SrRNA gene sequences, and selective representatives that matched the most abundant cyanobacterial field populations. I then developed methodologies for large-scale growth of the selected isolates to produce location-specific and genetically autochthonous inoculum for restoration. I also developed and tested viable methodologies to physiologically harden this inoculum and improve its survival under harsh field conditions. My tests proved that in most cases good viability of the inoculum could be attained under field-like conditions. In parallel, I used molecular ecology approaches to show that the biocrust pioneer, Microcoleus vaginatus, shapes its surrounding heterotrophic microbiome, enriching for a compositionally-differentiated "cyanosphere" that concentrates the nitrogen-fixing function. I proposed that a mutualism based on carbon for nitrogen exchange between M. vaginatus and its cyanosphere creates a consortium that constitutes the true pioneer community enabling the colonization of nitrogen-poor, bare soils. Using the right mixture of photosynthetic and diazotrophic cultures will thus likely help in soil restoration. Additionally, using physiological assays and molecular meta-analyses, I demonstrated that the largest contributors to N2-fixation in late successional biocrusts (three genera of heterocystous cyanobacteria) partition their niche along temperature gradients, and that this can explain their geographic patterns of dominance within biocrusts worldwide. This finding can improve restoration strategies by incorporating climate-matched physiological types in inoculum formulations. In all, this dissertation resulted in the establishment of a comprehensive "cyanobacterial biocrust nursery", that includes a culture collection containing 101 strains, isolation and cultivation methods, inoculum design strategies as well as field conditioning protocols. It constitutes a new interdisciplinary application of microbiology in restoration ecology.