English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

. Country-specific soil organic carbon (SOC) estimates are the baseline forthe Global SOC Map of the Global Soil Partnership (GSOCmap-GSP). Thisendeavor is key to explaining the uncertainty of global SOC estimates butrequires harmonizing heterogeneous datasets and building country-specificcapacities for digital soil mapping (DSM). We identified country-specificpredictors for SOC and tested the performance of five predictive algorithmsfor mapping SOC across Latin America. The algorithms included support vectormachines (SVMs), random forest (RF), kernel-weighted nearest neighbors (KK),partial least squares regression (PL), and regression kriging based onstepwise multiple linear models (RK). Country-specific training data and SOCpredictors (5  ×  5 km pixel resolution) were obtained fromISRIC – World Soil Information. Temperature, soil type, vegetationindices, and topographic constraints were the best predictors for SOC, butcountry-specific predictors and their respective weights varied across LatinAmerica. We compared a large diversity of country-specific datasets andmodels, and were able to explain SOC variability in a range between ∼  1 and ∼  60 %, with no universal predictive algorithm amongcountries. A regional ( n =  11 268 SOC estimates) ensemble of thesefive algorithms was able to explain ∼  39 % of SOC variability fromrepeated 5-fold cross-validation. We report a combined SOC stock of77.8  ±  43.6 Pg (uncertainty represented by the full conditionalresponse of independent model residuals) across Latin America. SOC stockswere higher in tropical forests (30  ±  16.5 Pg) and croplands(13  ±  8.1 Pg). Country-specific and regional ensembles revealedspatial discrepancies across geopolitical borders, higher elevations, andcoastal plains, but provided similar regional stocks (77.8  ±  42.2 and76.8  ±  45.1 Pg, respectively). These results are conservativecompared to global estimates (e.g., SoilGrids250m 185.8 Pg, the HarmonizedWorld Soil Database 138.4 Pg, or the GSOCmap-GSP 99.7 Pg). Countries withlarge area (i.e., Brazil, Bolivia, Mexico, Peru) and large spatial SOCheterogeneity had lower SOC stocks per unit area and larger uncertainty intheir predictions. We highlight that expert opinion is needed to set boundaryprediction limits to avoid unrealistically high modeling estimates. Formaximizing explained variance while minimizing prediction bias, the selectionof predictive algorithms for SOC mapping should consider density of availabledata and variability of country-specific environmental gradients. This studyhighlights the large degree of spatial uncertainty in SOC estimates acrossLatin America. We provide a framework for improving country-specific mappingefforts and reducing current discrepancy of global, regional, andcountry-specific SOC estimates.

No silver bullet for digital soil mapping: country-specific soil organic carbon estimates across Latin America