Changquan Wang

Spatiotemporal variations and factors affecting soil nitrogen in the purple hilly area of Southwest China during the 1980s and the 2010s

Determination of soil nitrogen distributions and the factors affecting them is critical for nitrogen fertilizer management and prevention of nitrogen pollution. In this paper, the spatiotemporal variations of soil nitrogen and the relative importance of their affecting factors were analysed at a county scale in the purple hilly area of the mid-Sichuan Basin in Southwest China based on soil data collected in 1981 and 2012. Statistical results showed that soil total nitrogen (TN) increased from 0.88 g kg(-1) in 1981 to 1.12 g kg(-1) in 2012, whereas available nitrogen (AN) decreased from 84.22 mg kg(-1) to 74.35 mg kg(-1). In particular, AN showed a significant decrease in agricultural ecosystems but remained stable in woodland and grassland. Correspondingly, most of the study area exhibited increased TN content and decreased AN content in space. The nugget/sill ratios of TN and AN increased from 0.419 to 0.608 and from 0.733 to 0.790, whereas spatial correlation distances decreased from 12.00 km to 9.50 km and from 9.50 km to 9.00 km, respectively, suggesting that the spatial dependence of soil nitrogen became weaker and that the extrinsic factors played increasingly important roles in affecting the soil nitrogen distribution. Soil group and land use type were the two dominant factors in 1981, followed by topographic factors, vegetation coverage and parent material, whereas land use type became the most important factor in 2012, and the relative contribution of topographic factors declined markedly. The results suggested that land use related to cultivation management and fertilizer application was the decisive factor for soil nitrogen change. The increase in TN content and the decrease in AN content over the study period also suggested improper use of nitrogen fertilizer, which can result in nitrogen loss through increasing nitrification rates. Thus, effective measures should be taken to increase the uptake rate of nitrogen and prevent nitrogen pollution.

Spatial prediction of soil nutrient in a hilly area using artificial neural network model combined with kriging

ABSTRACT It is widely recognized that using correlated environmental factors as auxiliary variables can improve the prediction accuracy of soil properties. In this study, a radial basis function neural network (RBFNN) model combined with ordinary kriging (OK) was proposed to predict spatial distribution of four soil nutrients based on the same framework used by regression kriging (RK). In RBFNN_OK, RBFNN model was used to explain the spatial variability caused by the selected auxiliary factors, while OK was used to express the spatial autocorrelation in RBFNN prediction residuals. The results showed that both RBFNN_OK and RK presented prediction maps with more details. However, RK does not always obtain mean errors (MEs) which were closer to 0 and lower root mean square errors (RMSEs) and mean relative errors (MREs) than OK. Conversely, MREs of RBFNN_OK were much closer to 0 and its RMSEs and MREs were relatively lower than OK and RK. The results suggest that RBFNN_OK is a more unbiased method with more stable prediction performance as well as improvement of prediction accuracy, which also indicates that artificial neural network model is more appropriate than regression model to capture relationships between soil variables and environmental factors. Therefore, RBFNN_OK may provide a useful framework for predicting soil properties.

Cadmium adsorption, chelation and compartmentalization limit root-to-shoot translocation of cadmium in rice ( Oryza sativa L.)

Strategies to reduce cadmium (Cd) in rice grain, below concentrations that represent serious human health concerns, require that the mechanisms of Cd distribution and accumulation within rice plants be established. Here, a comprehensive hydroponic experiment was performed to investigate the differences in the Cd uptake, chelation and compartmentalization between high (D83B) and low (D62B) Cd-accumulation cultivars contrasting in Cd accumulation in order to establish the roles of these processes in limiting Cd translocation from root to shoot. D83B showed 3-fold higher Cd accumulation in the shoots than the cultivar D62B. However, a short-term Cd uptake experiment showed more Cd uptake by D62B than by D83B. The distribution of Cd in roots and shoots differed significantly. D83B translocated 38% of total Cd taken up to the shoots, whereas D62B retained most of the Cd in the roots. D62B had higher amounts of non-protein thiols (NPTs) and glutathione (GSH) than D83B. The NPT and Cd distribution ratio (CDR) in the anionic form in the roots of D62B increased gradually as Cd concentration increased. In D83B, in contrast, levels of CDR in the cationic form increased significantly from 22.10 to 43.37%, while NPT only increased slightly. Furthermore, the percentage of Cd ions retained in thiol-rich peptides, especially in the HMW complexes, was significantly higher in D62B compared with D83B. However, D83B possessed a greater proportion of potentially mobile (cationic) Cd in the roots and showed superior Cd translocation from root to shoot. Taken as a whole, the results presented in this study revealed that Cd chelation, compartmentalization and adsorption contribute to the Cd retention in roots.

Prediction of soil cadmium distribution across a typical area of Chengdu Plain, China

A suitable method and appropriate environmental variables are important for accurately predicting heavy metal distribution in soils. However, the classical methods (e.g., ordinary kriging (OK)) have a smoothing effect that results in a tendency to neglect local variability, and the commonly used environmental variables (e.g., terrain factors) are ineffective for improving predictions across plains. Here, variables were derived from the obvious factors affecting soil cadmium (Cd), such as road traffic, and were used as auxiliary variables for a combined method (HASM_RBFNN) that was developed using high accuracy surface modelling (HASM) and radial basis function neural network (RBFNN) model. This combined method was then used to predict soil Cd distribution in a typical area of Chengdu Plain in China, considering the spatial non-stationarity of the relationships between soil Cd and the derived variables based on 339 surface soil samples. The results showed that HASM_RBFNN had lower prediction errors than OK, regression kriging (RK) and HASM_RBFNNs, which didn’t consider the spatial non-stationarity of the soil Cd-derived variables relationships. Furthermore, HASM_RBFNN provided improved detail on local variations. The better performance suggested that the derived environmental variables were effective and HASM_RBFNN was appropriate for improving the prediction of soil Cd distribution across plains.

Spatially distributed modeling of soil organic carbon across China with improved accuracy

There is a need for more detailed spatial information on soil organic carbon (SOC) for the accurate estimation of SOC stock and earth system models. As it is effective to use environmental factors as auxiliary variables to improve the prediction accuracy of spatially distributed modeling, a combined method (HASM_EF) was developed to predict the spatial pattern of SOC across China using high accuracy surface modeling (HASM), artificial neural network (ANN), and principal component analysis (PCA) to introduce land uses, soil types, climatic factors, topographic attributes, and vegetation cover as predictors. The performance of HASM_EF was compared with ordinary kriging (OK), OK, and HASM combined, respectively, with land uses and soil types (OK_LS and HASM_LS), and regression kriging combined with land uses and soil types (RK_LS). Results showed that HASM_EF obtained the lowest prediction errors and the ratio of performance to deviation (RPD) presented the relative improvements of 89.91%, 63.77%, 55.86%, and 42.14%, respectively, compared to the other four methods. Furthermore, HASM_EF generated more details and more realistic spatial information on SOC. The improved performance of HASM_EF can be attributed to the introduction of more environmental factors, to explicit consideration of the multicollinearity of selected factors and the spatial nonstationarity and nonlinearity of relationships between SOC and selected factors, and to the performance of HASM and ANN. This method may play a useful tool in providing more precise spatial information on soil parameters for global modeling across large areas.