U.S. flag

An official website of the United States government

Dot gov

Official websites use .gov
A .gov website belongs to an official government organization in the United States.


Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.


Main content area

An improved similarity-based approach to predicting and mapping soil organic carbon and soil total nitrogen in a coastal region of northeastern China

Shuai Wang, Kabindra Adhikari, Qianlai Zhuang, Zijiao Yang, Xinxin Jin, Qiubing Wang, Zhenxing Bian
PeerJ 2020 v.8 no. pp. 1-26
altitude, climate, coasts, ecosystems, kriging, land use, landscapes, model validation, models, normalized difference vegetation index, prediction, regression analysis, soil depth, soil management, soil map, soil nutrient balance, soil organic carbon, soil properties, soil quality, soil sampling, soil surveys, temperature, topographic slope, topsoil, total nitrogen, vegetation cover, China
Soil organic carbon (SOC), and soil total nitrogen (STN) are major soil properties and indicators for soil quality and fertility. Accurate mapping SOC, and STN in soils shall help both managed and natural ecosystem and soil-landuse management. We proposed and tested an improved similarity-based approach (ISA) to predict and map topsoil (0-20 cm soil depth) SOC, and STN content in a coastal region of northeastern China. Soil survey data in 2012 with 126 point observations and six environmental variables (elevation, slope gradient, topographic wetness index, the mean annual temperature, the mean annual temperature, and normalized difference vegetation index) were used as predictors of SOC and STN. Soil sampling profiles at depth of 0-20 cm were used for model prediction and validation. The ISA model performance was compared with the geographically weighted regression (GWR), regression kriging (RK), boosted regression trees (BRT) considering mean absolute prediction error (MAE), root mean square error (RMSE), coefficient of determination (R2), and maximum relative difference (RD) indices calculated for the test data.. The most influential environmental predictors for both SOC, and STN were also identified. We found that the ISA method performed well with the highest R2 and lowest MAE, RMSE compared to GWR, RK, and BRT models. The ISA model could explain 76%, and 83% of the total SOC, and STN variability in the study area, respectively. Elevation had the largest influence on SOC, and STN distribution. We conclude that the proposed ISA method is a robust, and an effective method in mapping soil properties including SOC, and STN, particularly in areas with complex vegetation-landscape with limited samples. However, the methodology needs to be tested from different part of the world for its wider application.