Spatial Assessment of Soil Organic Carbon Density Through Random Forests Based Imputation
Spatial Assessment of Soil Organic Carbon Density Through Random Forests Based Imputation
复制标题
DOI：
10.1007/s12524-013-0332-x
复制
发表时间：
2014-02
影响因子：
2.5
通讯作者：
K. Sreenivas;G. Sujatha;K. Sudhir;Dr. U. V. Kiran;M. Fyzee;T. Ravisankar;V. Dadhwal
中科院分区：
文献类型：
--
作者：
K. Sreenivas;G. Sujatha;K. Sudhir;Dr. U. V. Kiran;M. Fyzee;T. Ravisankar;V. Dadhwal
土壤碳库的区域估算采用了各种方法，将联合收割机土壤图与样本数据库相结合。点土壤有机碳（SOC）密度的空间化采用的方法，如回归，空间插值，多边形求和等，目前的工作探讨了基于数据挖掘的空间插补土壤有机碳密度的空间评估。研究区域包括印度的安得拉邦和卡纳塔克邦。采用分层随机抽样方法进行实地采样，确定土地覆被/利用、土壤类型、农业生态区。在1公里分辨率的气候，植被指数，土地覆盖，土壤类型，地形的空间数据被用来作为输入建模的顶部30厘米土壤有机碳（SOC）密度。为了对SOC密度进行建模，采用了具有最佳参数和输入变量的基于随机森林（RF）的模型。试验结果表明，500棵树，每棵树5个变量，可以解释研究区土壤有机碳密度的最大变异。在用于模拟SOC密度的各种输入变量中，土地利用/覆盖是影响SOC密度的最重要因素，其重要性得分为34.7，其次是NDVI，得分为12.9。预测的平均SOC密度范围在2.22和13.2 Kg m− 2之间，并且在顶部30 cm深度处SOC的估计池大小为安得拉邦的923 Tg和卡纳塔克邦的1，029 Tg。使用该模型预测的SOC密度与实测观测值吻合良好（R= 0.86）。
Regional estimates of soil carbon pool have been made using various approaches that combine soil maps with sample databases. The point soil organic carbon (SOC) densities are spatialized employing approaches like regression, spatial interpolation, polygon based summation, etc. The present work investigates a data mining based spatial imputation for spatial assessment of soil organic carbon density. The study area covers Andhra Pradesh and Karnataka states of India. Field sampling was done using stratified random sampling method with land cover/use, soil type, agro-ecological regions for defining strata. The spatial data at 1 km resolution on climate, NDVI, land cover, soil type, topography was used as input for modeling the top 30 cm Soil Organic Carbon (SOC) density. To model the SOC density, a Random Forest (RF) based model with optimal parameters and input variables has been adopted. Experiment results indicate that 500 number of trees with 5 variables at each split could explain the maximum variability of soil organic carbon density of the study area. Out of various input variables used to model SOC density, land use / cover was found to be the most significant factor that influences SOC density with a distinct importance score of 34.7 followed by NDVI with a score of 12.9. The predicted mean SOC densities range between 2.22 and 13.2 Kg m−2and the estimated pool size of SOC in top 30 cm depth is 923 Tg for Andhra Pradesh and 1,029 Tg for Karnataka. The predicted SOC densities using this model were in good agreement with the measured observations (R= 0.86).