Climate change will alter precipitation patterns with consequences for soil C cycling. An understanding of how fluctuating soil moisture affects microbial processes is therefore critical to predict responses to future global change. We investigated how long‐term experimental field drought influences microbial tolerance to lower moisture levels (“resistance”) and ability to recover when rewetted after drought (“resilience”), using soils from a heathland which had been subjected to experimental precipitation reduction during the summer for 18 years. We tested whether drought could induce increased resistance, resilience, and changes in the balance between respiration and bacterial growth during perturbation events, by following a two‐tiered approach. We first evaluated the effects of the long‐term summer drought on microbial community functioning to drought and drying–rewetting (D/RW), and second tested the ability to alter resistance and resilience through additional perturbation cycles. A history of summer drought in the field selected for increased resilience but not resistance, suggesting that rewetting after drought, rather than low moisture levels during drought, was the selective pressure shaping the microbial community functions. Laboratory D/RW cycles also selected for communities with a higher resilience rather than increased resistance. The ratio of respiration to bacterial growth during D/RW perturbation was lower for the field drought‐exposed communities and decreased for both field treatments during the D/RW cycles. This suggests that cycles of D/RW also structure microbial communities to respond quickly and efficiently to rewetting after drought. Our findings imply that microbial communities can adapt to changing climatic conditions and that this might slow the rate of soil C loss predicted to be induced by future cyclic drought.

Nutrients constrain the soil carbon cycle in tropical forests, but we lack knowledge on how these constraints vary within the soil microbial community. Here, we used in situ fertilization in a montane tropical forest and in two lowland tropical forests on contrasting soil types to test the principal hypothesis that there are different nutrient constraints to different groups of microorganisms during the decomposition of cellulose. We also tested the hypotheses that decomposers shift from nitrogen to phosphorus constraints from montane to lowland forests, respectively, and are further constrained by potassium and sodium deficiency in the western Amazon. Cellulose and nutrients (nitrogen, phosphorus, potassium, sodium, and combined) were added to soils in situ, and microbial growth on cellulose (phospholipid fatty acids and ergosterol) and respiration were measured. Microbial growth on cellulose after single nutrient additions was highest following nitrogen addition for fungi, suggesting nitrogen as the primary limiting nutrient for cellulose decomposition. This was observed at all sites, with no clear shift in nutrient constraints to decomposition between lowland and montane sites. We also observed positive respiration and fungal growth responses to sodium and potassium addition at one of the lowland sites. However, when phosphorus was added, and especially when added in combination with other nutrients, bacterial growth was highest, suggesting that bacteria out-compete fungi for nitrogen where phosphorus is abundant. In summary, nitrogen constrains fungal growth and cellulose decomposition in both lowland and montane tropical forest soils, but additional nutrients may also be of critical importance in determining the balance between fungal and bacterial decomposition of cellulose.

No abstract

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

No abstract

No abstract

Fungi and bacteria are the two principal microbial groups in soil, responsible for the breakdown of organic matter (OM). The relative contribution of fungi and bacteria to decomposition is thought to impact biogeochemical cycling at the ecosystem scale, whereby bacterially dominated decomposition supports the fast turnover of easily available substrates, whereas fungal-dominated decomposition leads to the slower turnover of more complex OM. However, empirical support for this is lacking. We used soils from a detritus-input-and-removal treatment experiment in an oldgrowth coniferous forest, where above-and belowground litter inputs have been manipulated for 20 years. These manipulations have generated variation in OM quality, as defined by energetic content and proxied as respiration per g soil organic matter (SOM) and the δ 13 C signature in respired CO 2 and microbial PLFAs. Respiration per g SOM reflects the availability and lability of C substrate to microorganisms, and the δ 13 C signature indicates whether the C used by microorganisms is plant-derived and higher quality (more δ 13 C depleted) or more microbially processed and lower quality (more δ 13 C enriched). Surprisingly, higher quality C did not disproportionately benefit bacterial decomposers. Both fungal and bacterial growth increased with C quality, with no systematic change in the fungal-to-bacterial growth ratio, reflecting the relative contribution of fungi and bacteria to decomposition. There was also no difference in the quality of C targeted by bacterial and fungal decomposers either for catabolism or anabolism. Interestingly, respired CO 2 was more δ 13 C enriched than soil C, suggesting preferential use of more microbially processed C, despite its lower quality. Gross N mineralization and consumption were also unaffected by differences in the ratio of fungal-to-bacterial growth. However, the C/N ratio of mineralization was lower than the average C/N of SOM, meaning that microorganisms specifically targeted N-rich components of OM, suggesting selective microbial N-mining. Consistent with the δ 13 C data, this reinforces evidence for the use of more microbially processed OM with a lower C/N ratio, rather than plant-derived OM. These results challenge the widely held assumption that microorganisms favor high-quality C sources and suggest that there is a trade-off in OM use which may be related to the growth-limiting factor for microorganisms in the ecosystem.

Soil microbial communities perform vital ecosystem functions, such as the decomposition of organic matter to provide plant nutrition. However, despite the functional importance of soil microorganisms, attribution of ecosystem function to particular constituents of the microbial community has been impeded by a lack of information linking microbial function to community composition and structure. Here, we propose a function‐first framework to predict how microbial communities influence ecosystem functions. We first view the microbial community associated with a specific function as a whole and describe the dependence of microbial functions on environmental factors (e.g., the intrinsic temperature dependence of bacterial growth rates). This step defines the aggregate functional response curve of the community. Second, the contribution of the whole community to ecosystem function can be predicted, by combining the functional response curve with current environmental conditions. Functional response curves can then be linked with taxonomic data in order to identify sets of “biomarker” taxa that signal how microbial communities regulate ecosystem functions. Ultimately, such indicator taxa may be used as a diagnostic tool, enabling predictions of ecosystem function from community composition. In this paper, we provide three examples to illustrate the proposed framework, whereby the dependence of bacterial growth on environmental factors, including temperature, pH, and salinity, is defined as the functional response curve used to interlink soil bacterial community structure and function. Applying this framework will make it possible to predict ecosystem functions directly from microbial community composition.