Despite the important role that biocrust communities play in maintaining ecosystem structure and functioning in drylands world-wide, few studies have evaluated how climate change will affect them. Using data from an 8-yr-old manipulative field experiment located in central Spain, we evaluated how warming, rainfall exclusion and their combination affected the dynamics of biocrust communities in areas that initially had low (< 20%, LIBC plots) and high (> 50%, HIBC plots) biocrust cover. Warming reduced the richness (35 ± 6%), diversity (25 ± 8%) and cover (82 ± 5%) of biocrusts in HIBC plots. The presence and abundance of mosses increased with warming through time in these plots, although their growth rate was much lower than the rate of lichen death, resulting in a net loss of biocrust cover. On average, warming caused a decrease in the abundance (64 ± 7%) and presence (38 ± 24%) of species in the HIBC plots. Over time, lichens and mosses colonized the LIBC plots, but this process was hampered by warming in the case of lichens. The observed reductions in the cover and diversity of lichen-dominated biocrusts with warming will lessen the capacity of drylands such as that studied here to sequester atmospheric CO and to provide other key ecosystem services associated to these communities.

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as

Soil surface communities dominated by mosses, lichens and cyanobacteria (biocrusts) are common between vegetation patches in drylands worldwide, and are known to affect soil wetting and drying after rainfall events. While ongoing climate change is already warming and changing rainfall patterns of drylands in many regions, little is known on how these changes may affect the hydrological behaviour of biocrust-covered soils. We used eight years of continuous soil moisture and rainfall data from a climate change experiment in central Spain to explore how biocrusts modify soil water gains and losses after rainfall events under simulated changes in temperature (2.5°C warming) and rainfall (33% reduction). Both rainfall amount and biocrust cover increased soil water gains after rainfall events, whereas experimental warming, rainfall intensity and initial soil moisture decreased them. Initial moisture, maximum temperature and biocrust cover, by means of enhancing potential evapotranspiration or by soil darkening, increased the drying rates and enhanced the exponential behaviour of the drying events. Meanwhile, warming reduced their exponential behaviour. The effects of climate change treatments on soil water gains and losses changed through time, with important differences between the first two years of the experiment and five years after its setup. These effects were mainly driven by the important reductions in biocrust cover and diversity observed under warming. Our results highlight the importance of long-term studies to understand soil moisture responses to ongoing climate change in drylands.

Little is known about the role of biocrusts in regulating the responses of N 2 O and CH 4 fluxes to climate change in drylands. Here, we aim to help filling this knowledge gap by using an eight-year field experiment in central Spain where temperature and rainfall are being manipulated (~1.9 °C warming, 33% rainfall reduction, and their combination) in areas with and without well-developed biocrust communities. Areas with initial high cover of well-developed biocrusts showed lower N 2 O emissions, enhanced CH 4 uptake and higher abundances of functional genes linked to N 2 O and CH 4 fluxes compared with areas with poorly-developed biocrusts.Moreover, biocrusts modulated the responses of gases emissions and related functional genes to warming and rainfall reductions. Specifically, we found under rainfall exclusion and its combination with warming a sharp reduction in N 2 O fluxes (~96% and ~197%, respectively) only under well-developed biocrust cover. Warming and its combination with rainfall exclusion reduced CH 4 consumption in areas with initial low cover of well-developed biocrust, whereas rainfall exclusion enhanced CH 4 uptake only in areas with high initial cover of well-developed biocrusts. Similarly, the combination of warming and rainfall exclusion increased the 3 abundance of the nosZ gene compared to the rainfall exclusion treatment and increased the abundance of the pmoA gene compared to the control, but only in areas with low biocrust cover. Taken together, our results indicate that well-developed biocrust communities could counteract the impact of warming and altered rainfall patterns on soil N 2 O and CH 4 fluxes, highlighting their importance and the need to preserve them to minimize climate change impacts on drylands.

Aim Microorganisms carrying pmoA and nosZ genes are major drivers of methane and nitrous oxide fluxes from soils. However, most studies on these organisms have been conducted in mesic ecosystems; therefore, little is known about the factors driving their distribution in drylands, the largest biome on Earth. We conducted a global survey to evaluate the role of climate‐ and soil‐related variables as predictors of the richness, abundance and community structure of bacteria carrying pmoA and nosZ genes. Location Eighty dryland ecosystems distributed worldwide. Time period From February 2006 to December 2011. Major taxa studied Methanotrophic (carrying the pmoA gene) and denitrifiying (carrying the nosZ gene) bacteria. Methods We used data from a field survey and structural equation modelling to evaluate the direct and indirect effects of climatic (aridity, rainfall seasonality and mean annual temperature) and soil (organic carbon, pH and texture) variables on the total abundance, richness and community structure of microorganisms carrying pmoA and nosZ genes. Results Taxa related to Methylococcus capsulatus or Methylocapsa sp., often associated with mesic environments, were common in global drylands. The abundance and richness of methanotrophs were not associated with climate or soil properties. However, mean annual temperature, rainfall seasonality, organic C, pH and sand content were highly correlated with their community structure. Aridity and soil variables, such as sand content and pH, were correlated with the abundance, community structure and richness of the nosZ bacterial community. Main conclusions Our study provides new insights into the drivers of the abundance, richness and community structure of soil microorganisms carrying pmoA and nosZ genes in drylands worldwide. We highlight how ongoing climate change will alter the structure of soil microorganisms, which might affect the net CH4 exchange and will probably reduce the capacity of dryland soils to carry out the final step of denitrification, favouring net N2O emissions.