The stress-gradient hypothesis (SGH) predicts a shift from predominant competition to facilitation as abiotic stress increases. Most empirical tests of the SGH have evaluated the interactions between a single or a few pairs of species, have not considered the effects of multiple stress factors, and have not explored these interactions at nested spatial scales. We sampled 63 0.25-m2 plots, each subdivided into 100 5×5 cm and 25 10×10 cm sampling squares, in a semi-arid Mediterranean environment to evaluate how co-occurrence patterns among biological soil crusts (BSC)-forming lichens changed along natural stress gradients driven by water and nutrient (N, P, K) availability. According to the SGH, we tested the hypothesis that the fine-scale spatial arrangement of BSC-forming lichens should shift from prevailing interspecific segregation to aggregation as abiotic stress increases. Co-occurrence patterns ranged from significant species segregation to aggregation at the two spatial scales studied. When using the 5×5 cm grid, more plots showed significant species segregation than aggregation. At this sampling scale, co-occurrence increased as water and nutrient availability decreased and increased, respectively. Small increases in soil pH promoted species co-occurrence. Interspecific segregation was promoted as the cover of highly competitive species, such as Diploschistes diacapsis, increased. No significant relationships between co-occurrence and the surrogates of abiotic stress were observed when data was arranged in a 10×10 cm grid. Our co-occurrence analyses partially supported predictions from the SGH, albeit the results obtained were dependent on the type of abiotic stress and the spatial scale considered. They show the difficulties of predicting how co-occurrence patterns change along complex stress gradients, and highlight the need of incorporating the effects of abiotic stress promoted by different resources, such as water and nutrients, into the conceptual framework of the SGH.