Manuel Delgado-Baquerizo

Biological soil crusts (BSCs) greatly influence the N cycle of semi-arid ecosystems, as some organisms forming them are able to fix atmospheric N. However, BSCs are not always taken into account when studying biotic controls on N cycling and transformations. Our main objective was to understand how BSCs modulate the availability of N in a semi-arid Mediterranean ecosystem dominated by the tussock grass Stipa tenacissima. We selected the six most frequent soil cover types in the study area: S. tenacissima tussocks (ST), Retama sphaerocarpa shrubs (RS), and open areas with very low (BS), low (LC) medium (MC) and high (HC) cover of well developed and lichen-dominated BSCs. The temporal dynamics of available N dynamics followed changes in soil moisture. Available NH 4+-N did not differ between microsites, while available NO 3--N was substantially higher in the RS than in any other microsite. No significant differences in the amount of available NO 3--N were found between ST and BS microsites, but these microsites had more NO 3--N than those dominated by BSCs (LC, MC and HC). Our results suggest that BSCs may be inhibiting nitrification, and highlight the importance of this biotic community as a modulator of the availability of N in semi-arid ecosystems.

It has been suggested that the dominance of N forms should shift from dissolved organic nitrogen (DON) to nitrate along a gradient of increasing N availability. We aimed to apply this model at a local scale within a semi-arid ecosystem showing a high spatial heterogeneity in the distribution of vegetation and soil resources. By doing this, we seek a better understanding of the N cycling in spatially heterogeneous ecosystems. We took soil samples from the three major sources of spatial heterogeneity: the grass Stipa tenacíssima, the N-fixing shrub Retama sphaerocarpa, and open areas. We also sampled the biological soil crust (BSC) located in the latter areas as another source of spatial heterogeneity. BSC microsites were classified by four levels of soil coverage, ranging from high coverage (66%) to bare soil. The proportion of nitrate, ammonium and DON was determined in all microsites. DON was the dominant N form for open areas, while nitrate was dominant under the canopy of Retama; these microsites contained the lowest and highest N availability, respectively. Under BSC, DON was the dominant N form. We found high temporal variability in the dominance of N forms for all microsites. Our results suggest that the biome-derived model of Schimel and Bennett (2004) explaining N form dominance across N availability gradients may be extended to local gradients.

It has been suggested that the relative abundance of soil nitrogen forms should change along an N availability gradient. This model was originally described at a biome scale, and few studies have tested it at other scales. Moreover, none of them has examined whether changes in the relative rates of ammonification, nitrification and depolymerization rates also occurs. Our goal was to test whether these N transformation rates change along an N availability gradient which is likely to exist between forest, shrubs and grasses. We used three N availability indexes (total K2SO4-extractable N, ion exchange membrane N and the sum of N mineralization and depolymerization rates). Depolymerization dominated over mineralization in the two poorest plant communities, while ammonification and nitrification rates dominated in intermediate and nutrient rich plant communities respectively. These results confirm that the Schimel and Bennett model can be applied at a regional scale, and that N availability may be modulating not only the dominant N form, but also the relative abundance of a particular N transformation rate.► Depolymerization dominated over mineralization in the two poorest plant communities. ► The Schimel and Bennett model can be applied at a regional scale. ► N availability modulates the relative dominance of a particular N transformation rate.