Estimation Of Forest Soil Organic Carbon Storage Based On Digital Soil Mapping Technology

It is of great significance to master forest soil organic carbon storage for forest resource management and service.In order to improve the accuracy of estimation,it is necessary to obtain accurate carbon storage estimation materials,namely the spatial distribution information of soil attributes.In this study,Nanshan Forest Farm,a state-owned forest farm located in Jiyuan City,was selected as the study area.The conditional Latin hypercube sampling method was selected as the sampling method,and the sample soil was analyzed in the laboratory to obtain the data of sampling points.In the selection of landscape environmental factors,a total of six topographic factors such as elevation were selected as the predictors of the soil carbon storage estimation,and the fuzzy C-means(FCM)clustering was performed on these six factors.Twelve kinds of clustering categories can be obtained through fuzzy clustering,and values are assigned to each category,that is,the typical values of each category are selected,and then the ArcGIS 10.2 software is used to forecast and map the four properties of soil based on the clustering results.Based on the four soil attribute maps,the forecast map of organic carbon density can be obtained by calculating the organic carbon density according to the calculation formula.According to the density prediction map of each soil layer,the estimated value of soil organic carbon storage in the whole section of the study area can be obtained by superposition calculation.The results are as follows:1)The prediction chart of soil organic carbon density in the study area was calculated,and the average value of the organic carbon density was 4.24 kg·m-2.The accuracy of the prediction chart was evaluated,and each precision evaluation index was calculated.The mean error(ME),mean absolute error(MAE)and root mean square error(RMSE)were 0.08 kg·m-2,2.80 kg·m-2 and 5.03 kg·m-2 respectively.Through the results of precision evaluation,it can be found that the digital soil mapping method has higher accuracy and better stability in predicting results,which is more suitable for the estimation of carbon density.2)The final estimate of carbon storage in this study is 3.08×108 kg,which is a relatively accurate estimate.In the end,three methods are used to estimate carbon stocks.In addition to the digital mapping method,the traditional type method using only typical points and the traditional type method using all sample points are compared.By comparing the results of the three methods,it can be found that the two methods of the traditional type method are very different,and their predictions are particularly unstable.Although the digital mapping method uses fewer sample points,its accuracy is not low,and the prediction model is stable,so it is more suitable for the estimation of carbon stocks.The results show that the digital soil mapping technology can be used to estimate the forest soil organic carbon storage with high accuracy,and the results are relatively reliable,and it does not require a large number of samples.The forecast map is stable,and the spatial changes and distribution of soil information can be expressed in a graphical form,which is more detailed and coherent.This study is a new exploration of the estimation method of soil organic carbon storage,which plays an exemplary role in the future similar studies,and has certain help in improving the efficiency and accuracy of the prediction mapping.