Dataset of digital soil mapping products for the Qinghai-Tibet Plateau (2015-2024)

Based on the "second Qinghai Tibet Plateau comprehensive scientific investigation" and "China's soil series investigation and compilation of China's soil series" "The obtained soil survey profile data, using predictive Digital Soil Mapping paradigm, using geographic information and remote sensing technology for fine description and spatial analysis of the soil forming environment, developed adaptive depth function fitting methods, and integrated advanced ensemble machine learning methods to generate a series of soil attributes (soil organic carbon, pH value, total nitrogen, total phosphorus, total potassium, cation exchange capacity, gravel content (>2mm) in the Qinghai Tibet plateau region." , sand, silt, clay, soil texture type, unit weight, soil thickness, etc.) and quantify the spatial distribution of uncertainty. Compared with the existing soil maps, it better represents the spatial variation characteristics of soil properties in the Qinghai Tibet Plateau. The data set can provide soil information support for the study of soil, ecology, hydrology, environment, climate, biology, etc. in the Qinghai Tibet Plateau.

0 2022-06-03

Basic soil property dataset of high-resolution China Soil Information Grids (2010-2018)

Soil is the basis of human survival and development. Many United Nations Sustainable Development Goals (SDGs) are directly related to the utilization and management of soil resources. However, most of the existing soil information in the world and China comes from historical soil survey, which is coarse and out-of-date, and can not meet the needs of dealing with global and regional problems such as food security, water shortage, land degradation and climate change. China has a vast territory with complex and diverse soil landscape and strong human activities. The establishment of high-precision soil information grid is of great significance in scientific frontier breakthrough and has broad prospects in applications. Here, we adopted predictive soil mapping paradigm and developed adaptive depth function fitting method and integrated it with state-of-the-art ensemble machine learning in a high performance parallel computing environment to generate 90-m resolution national gridded maps of soil properties (soil organic carbon, pH value, total nitrogen, total phosphorus, total potassium, cation exchange capacity, coarse fragments (> 2mm), sand, silt, clay, soil texture classes, bulk density, soil thickness, etc.) at multiple depths across China. Their uncertainty in soil predictions is also estimated in a spatial way. This was based on more than 5000 representative soil profile samples obtained from the "project of National Soil Series Survey and Compilation of Soil Series of China" in recent years and a suite of detailed covariates to characterize soil-forming environments using geographical information and remote sensing techniques. Compared with previous soil maps, we achieved significantly more detailed and accurate predictions which could well represent soil variations across the territory. This work has constructed China's first version of high-resolution National Soil Information Grids, which is also a significant contribution to the GlobalSoilMap.net project. It is expected to have a wide application prospect in the fields of soil resources, agriculture, hydrology, ecology, climate, environment and so on, such as soil monitoring and management, soil function evaluation, land surface process modelling and forensic soil evidence provenance.

0 2021-11-30