An integrated perspective of the most important factors driving the abundance of ammonia-oxidizing bacteria (AOB) and archaea (AOA) in natural ecosystems is lacking, especially in drylands. We evaluated how different climatic, abiotic, and nutrient-related factors determine AOA and AOB abundance in bare and vegetated microsites from grasslands throughout the Mediterranean Basin. We found a strong negative relationship between the abundance of AOA genes and soil fertility (availability of C, N, and P). Aridity and other abiotic factors (pH, sand content, and electrical conductivity) were more important than soil fertility in modulating the AOA/AOB ratio. AOB were more abundant under vegetated microsites, while AOA, highly resistant to stressful conditions, were more abundant in bare ground areas. These results suggest that AOA may carry out nitrification in less fertile microsites, while AOB predominate under more fertile conditions. Our results indicate that the influence of aridity and pH on the relative dominance of AOA and AOB genes is ultimately determined by local-scale environmental changes promoted by perennial vegetation. Thus, in spatially heterogeneous ecosystems such as drylands, there is a mutual exclusion and niche division between these microorganisms, suggesting that they may be functionally complementary.

Changes after Aug 12, 2013:  The qPCR (AOA and AOB) data in the original Dryad repository file were inaccurate. During the generation of the clean dataset published in Dryad, the data were back-transformed from the log-scaled metrics used in the paper (now included), rather than including the raw, untransformed data. Please note that the data presented in the published paper itself were always correct. We have now updated the repository to include both the log-transformed data used in the paper and the original qPCR (AOA and AOB) values.