In anthropic savannah ecosystems from the Iberian Peninsula (i.e., dehesa), complex interactions between climate change, pathogen outbreaks and human land use are presumed to be behind the observed increase in holm oak decline. These environmental disturbances alter the plant-soil microbial continuum, which can destabilize the ecological balance that sustains tree health. Yet, little is known about the underlying mechanisms, particularly the directions and nature of the causal-effect relations between plants and soil microbial communities. 

In this study, we aimed to determine the role of plant-soil feedbacks in climate-induced holm oak decline in the Iberian dehesa. Using a gradient of holm oak health, we reconstructed key soil biogeochemical cycles mediated by soil microbial communities. We used quantitative microbial element cycling (QMEC), a functional gene-array-based high-throughput technique to assess microbial functional potential in carbon (C), nitrogen (N), phosphorous (P), and sulfur (S) cycling. 

The onset of holm oak decline was positively related with the increase in relative abundance of soil microbial functional genes associated with denitrification and phosphorous mineralization (i.e., nirS3, ppx and pqqC; parameter value: 0.21, 0.23 and 0.4; p<0.05). The structural equation model (ꭓ² = 32.26, p-value = 0.73), moreover, showed a negative association between these functional genes and soil nutrient availability (i.e., mainly mineral nitrogen and phosphate). Particularly, the holm oak crown health was mainly determined by the abundance of phosphate (parameter value=0.27; p-value<0.05) and organic phosphorus (parameter value=-0.37; p-value<0.5). 

These data have been collected in holm oak dehesas. To account for the soil spatial heterogeneity, three different soil samples were collected below each of the 162 holm oak trees at a distance of 1 m from each trunk. The depth at which we collected the soil was determined by the depth where the shallowest holm oak fine root density peak was located. As we collected both the soil and the roots at the same depth, and this holm oak root depth was affected by the presence of the herbaceous root layer and historical land management practices, we excavated until we reached the soil layer predominantly containing holm oak roots. This typically occurred at an average depth of 15 cm and generally did not exceed 30 cm. Then, the three soil subsamples were pooled in one single composite sample that was maintained at 4°C (12h) until processing in the laboratory. These soil samples were then used to do soil chemical analyses and to quantify soil microbial functional genes. Specifically, for the soil chemical analyses, the 162 soil samples were dried at room temperature (~20ºC), sieved using a 2-mm mesh size and stored in darkness (cf. section 2.4.). Regarding the analysis of soil microbial functional genes, aliquots from the 162 soil samples were frozen at -20ºC just upon arrival at the laboratory every day after sampling and stored in darkness for approximately one month after sampling, until DNA extraction (cf. section 2.5.).