AUTHOR=Wang Shaohua , Qi Huxiao , Li Tianyu , Qin Yong , Fu Gang , Pan Xu , Zha Xinjie 

TITLE=Can normalized difference vegetation index and climate data be used to estimate soil carbon, nitrogen, and phosphorus and their ratios in the Xizang grasslands?

JOURNAL=Frontiers in Earth Science

VOLUME=Volume 11 - 2023

YEAR=2024

URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2023.1340020

DOI=10.3389/feart.2023.1340020

ISSN=2296-6463

ABSTRACT=Accurately quantifying the relative effects of climate change and human activities on soil carbon, nitrogen and phosphorus in alpine grasslands and their feedbacks is an important aspect of global change, and high-precision models are the key to solving this scientific problem with high quality. Therefore, nine models, including random forest model (RFM), generalized boosted regression model (GBRM), multiple linear regression model (MLRM), support vector machine model (SVMM), recursive regression tree model (RRTM), artificial neural network model (ANNM), generalized linear regression model (GLMR), conditional inference tree model (CITM), and eXtreme gradient boosting model (eXGBM), were used for modelling soil organic carbon (SOC), total nitrogen (TN), total phosphorus (TP), ratio of SOC to TN (C:N), ratio of SOC to TP (C:P), and ratio of TN to TP (N:P) at 0–10, 10–20 and 20–30 cm under non-grazing and free-grazing scenario in Xizang’s grasslands. Annual radiation (ARad), annual precipitation (AP) and annual temperature (AT) was used as independent variables under non-grazing scenario, whereas ARad, AP, AT and growing season maximum normalized difference vegetation index (NDVImax) was used as independent variables under free-grazing scenario. Overall, the RFM and GBRM had the greater accuracies than the other seven models. However, the tree numbers of the GBRM were much larger than those of the RFM, indicating that the GBRM may have greater model complexity and lower running speed. Therefore, the RFM had the best performance among the nine models in modelling SOC, TN, TP, C:N, C:P and N:P in Xizang’s grasslands. The RFM established in this study can not only help scientists save time and money on massive sampling and analysis, but also be used for the database construction of SOC, TN, TP and their ratios, and further scientific research related to ecological and environmental issues (e.g., examining whether soil systems intensified global warming over the past few decades can be revealed by exploring whether climate change and human activities altered soil organic carbon) in grasslands of the Xizang’s Plateau.