Reconstructing continental‐scale variation in soil δ 15 N: a machine learning approach in South America

Soil δ15N spatial distribution is primarily shaped by climatic patterns in the semiarid Caatinga, Northeast Brazil

Soil nitrogen isotopic composition (δ15Nsoil) is an invaluable tool as it integrates nitrogen (N) transformations in soils. In addition to serving as a baseline to understand the N cycle, spatial representations of δ15Nsoil across landscapes (or isoscapes) is a multi-purpose tool useful to investigate, for example, plant-microbe interactions, animal migration and forensics. We investigate the climatic and edaphic controls of δ15Nsoil utilising data from 29 geographical locations sampled across the semiarid Brazilian Caatinga biome. The sampling covered a mean annual precipitation (PA) gradient ranging from 0.51 to 1.36 m a-1 and eight soil types originating from three different geological origins. Our data show that the combination of higher aridity and lower seasonality (ψ) leads to higher values of δ15Nsoil. Moreover, soil total carbon had a positive relationship with δ15Nsoil, appearing within the best-supported models according to the information-theoretic approach undertaken here. The contribution to the plant communities by the Fabaceae trees expressed as their basal area was not related to δ15Nsoil values, suggesting that the magnitude of biological N fixation in the Caatinga is not large enough to be reflected in the soil. In addition, considering PA in a categorical fashion, i.e., 'high' (> 0.8 m a-1) and 'low' PA (< 0.8 m a-1), we found that, within the wetter category, δ15Nsoil was positively related to several soil properties (i.e., clay content, effective cation exchange capacity, exchangeable calcium, silt content, pHH2O, total phosphorus and sum of bases) and negatively related to sand content. Our study provides new insights into the functioning of semiarid ecosystems from a pedo-isotopic perspective and contributes to the overall understanding of the N cycle in the Caatinga region, with the potential to support the development of new conceptualisation of biogeochemical process and testing of global models that simulate N and C cycles.

Overestimated nitrogen loss from denitrification for natural terrestrial ecosystems in CMIP6 Earth System Models

Denitrification and leaching nitrogen (N) losses are poorly constrained in Earth System Models (ESMs). Here, we produce a global map of natural soil ¹⁵N abundance and quantify soil denitrification N loss for global natural ecosystems using an isotope-benchmarking method. We show an overestimation of denitrification by almost two times in the 13 ESMs of the Sixth Phase Coupled Model Intercomparison Project (CMIP6, 73 ± 31 Tg N yr^-1), compared with our estimate of 38 ± 11 Tg N yr^-1, which is rooted in isotope mass balance. Moreover, we find a negative correlation between the sensitivity of plant production to rising carbon dioxide (CO₂) concentration and denitrification in boreal regions, revealing that overestimated denitrification in ESMs would translate to an exaggeration of N limitation on the responses of plant growth to elevated CO₂. Our study highlights the need of improving the representation of the denitrification in ESMs and better assessing the effects of terrestrial ecosystems on CO₂ mitigation.

Global distribution and climate sensitivity of the tropical montane forest nitrogen cycle

Tropical forests are pivotal to global climate and biogeochemical cycles, yet the geographic distribution of nutrient limitation to plants and microbes across the biome is unresolved. One long-standing generalization is that tropical montane forests are nitrogen (N)-limited whereas lowland forests tend to be N-rich. However, empirical tests of this hypothesis have yielded equivocal results. Here we evaluate the topographic signature of the ecosystem-level tropical N cycle by examining climatic and geophysical controls of surface soil N content and stable isotopes (δ¹⁵N) from elevational gradients distributed across tropical mountains globally. We document steep increases in soil N concentration and declining δ¹⁵N with increasing elevation, consistent with decreased microbial N processing and lower gaseous N losses. Temperature explained much of the change in N, with an apparent temperature sensitivity (Q₁₀) of ~1.9. Although montane forests make up 11% of forested tropical land area, we estimate they account for >17% of the global tropical forest soil N pool. Our findings support the existence of widespread microbial N limitation across tropical montane forest ecosystems and high sensitivity to climate warming.