地球科学进展

地球科学进展 ›› 2012, Vol. 27 ›› Issue (2): 175 -184. doi: 10.11867/j.issn.1001-8166.2012.02.0175

研究论文 上一篇    下一篇

基于神经网络模型的中国表层土壤有机质空间分布模拟方法
李启权 1，2，王昌全 1，岳天祥 2，张文江 3，余勇 4   
  1. 1.四川农业大学资源环境学院，四川成都611130；
    2.中国科学院地理科学与资源研究所，北京100101；
    3.四川大学水力学与山区河流开发保护国家重点实验室，四川成都610065；
    4.四川农业大学林学院，四川雅安625014
  • 收稿日期:2011-10-20 修回日期:2011-12-03 出版日期:2012-02-10
  • 通讯作者: 李启权(1980-)，男，四川泸县人，讲师，主要从事土壤属性时空变化模拟研究. E-mail:liqq@lreis.ac.cn
  • 基金资助:

    国家杰出青年科学基金项目“资源环境模型与系统模拟”(编号：40825003)；国家自然科学基金青年科学基金项目“基于陆面水文过程的区域干旱遥感监测预报研究”(编号：40801175)资助.

Method for Spatial Simulation of Topsoil Organic Matter in China based on a Neural Network Model

Li Qiquan 1，2, Wang Changquan 1, Yue Tianxiang 2, Zhang Wenjiang 3, Yu Yong 4   

  1. 1.College of Resources and Environment, Sichuan Agricultural University, Chengdu611130, China;
    2.Institute of Geographic Sciences and Natural Resources Research,Chinese Academy of Sciences, Beijing100101,China；
    3.State Key Laboratory of Hydraulics and Mountain River Engineering,Sichuan University, Chengdu610065,China;
    4.College of Forestry, Sichuan Agricultural University, Yaan625014, China
  • Received:2011-10-20 Revised:2011-12-03 Online:2012-02-10 Published:2012-02-10
下载PDF ( 1438 )

基于全国第二次土壤普查得到的6 241个典型土壤剖面数据，采用主成分分析方法和径向基函数神经网络模型建立不同植被类型—土纲单元内土壤有机质与气候、地形和植被等环境因子间的非线性关系，模拟全国表层土壤有机质的空间分析格局。结果表明，该模型具有较准确的预测能力，性能指数达到1.94。与普通克里格法、反比距离法和多元回归模型相比，神经网络模型对621个验证点模拟结果与实测值的相关系数为0.799，分别提高了0.265、0.181和0.120，平均绝对误差分别降低了4.25、4.43和2.34 g/kg，平均相对误差分别降低了30.16%、32.66%和5.93%，均方根误差则分别降低了8.61、8.24和6.24 g/kg；从模拟结果图来看，神经网络模型能够提供更多的细节信息。该方法为大尺度土壤性质空间分布模拟提供了有益的参考。

关键词: 土壤有机质, 径向基函数神经网络, 空间模拟, 国家尺度

 Accurate simulation of spatial information of soil properties at large scale is essential to land use and environment management. Based on 6 241 typical soil profiles collected during the second national soil survey period (19791994), a radial basis function neural networks (RBFNN) method was used to capture the nonlinear relationship between topsoil organic matter and principal components converted from environmental variables including climate data, terrain attributes and vegetation index, to simulate the spatial pattern of topsoil organic matter at national scale in China. Results show that, the ratio of performance to deviation (RPD) produced by RBFNN is 1.94 which indicates that RBFNN has much stronger ability to simulate the spatial distribution of topsoil organic matter within a large scale. Error analysis based on 621 validation points show that, compared with ordinary kriging, inverse distance weighting and multiple linear regression model, correlation coefficients between the observed value increased by 0.265, 0.181 and 0.120, the mean absolute error decreased by 4.25, 4.43 and 2.34 g/kg, the mean relative error was reduced by 30.16%, 32.66% and 5.93%, and the root mean squared error decreased by 8.61, 8.24 and 6.24 g/kg. Moreover, RBFNN produces soil organic matter map with much higher level of detail, which is closer to the actual pattern of soil organic matter in the complex environment, especially for the area where the density of sample points is much smaller. The good performance of RBFNN method can be attributed to the ability of RBFNN method which can accurately capture the nonlinear relationship between soil organic matter and environmental factors. Thus, RBFNN method is considered as an effective tool to spatial simulation of soil properties at a large scale. Further researches are needed to employ more influencing factors into the method to improve the precision of simulation result.

Key words: Soil organic matter, Radial basis function neural networks, Spatial simulation, National scale

中图分类号: 

[1]Pan Genxing. Soil organic carbon stock, dynamics and climate change mitigation of China[J].Advances in Climate Change Research, 2008,4(5): 282-289.[潘根兴. 中国土壤有机碳库及其演变与应对气候变化[J]. 气候变化研究进展, 2008, 4(5): 282-289.]
[2]Li Qiquan, Yue Tianxiang, Fan Zemeng, et al. Spatial simulation of topsoil TN at the national scale in China[J]. Geographical Research, 2010, 29(11): 1 981-1 992.[李启权, 岳天祥, 范泽孟, 等. 中国表层土壤全氮的空间模拟分析[J]. 地理研究, 2010, 29(11): 1 981-1 992.]
[3]Yang Lin, Zhu Axing, Qin Chengzhi, et al. Soil property mapping using fuzzy membership—A case study of a study area in Heshan farm of Heilongjiang province[J]. Acta Pedologica Sinica, 2009, 46(1): 9-15.[杨琳, 朱阿兴, 秦承志, 等. 运用模糊隶属度进行土壤属性制图的研究——以黑龙江鹤山农场研究区为例[J]. 土壤学报, 2009, 46(1): 9-15.]
[4]Phachomphon K, Dlamini P, Chaplot V. Estimating carbon stocks at a regional level using soil information and easily accessible auxiliary variables[J]. Geoderma, 2010, 155(3/4): 372-380.
[5]Grunwald S. Multi-criteria characterization of recent digital soil mapping and modeling approaches[J]. Geoderma, 2009, 152(3/4):195-207.
[6]Zhu A X, Hudson B, Burt J, et al. Soil mapping using GIS, expert knowledge, and fuzzy logic[J]. Soil Science Society of America Journal, 2001, 65(5): 1 463-1 472.
[7]Wu Lezhi, Cai Zucong. The relationship between the spatial scale and the variation of soil organic matter in China[J]. Advances in Earth Science, 2006, 21(9): 965-972.[吴乐知, 蔡祖聪. 中国土壤有机质含量变异性与空间尺度的关系[J]. 地球科学进展, 2006, 21(9): 965-972.]
[8]Mishra U, Lal R, Liu D S, et al. Predicting the spatial variation of the soil organic carbon pool at a regional scale[J]. Soil Science Society of America Journal, 2010, 74(3): 906-914.
[9]Bell J C, Grigal D F, Bates P C. A Soil Terrain Model for estimating spatial patterns of soil organic carbon[M]∥Wilson J P ed. Terrain Analysis: Principles and Applications.New York: John Wiley & Sons,  2000: 295-310.
[10]McBratney A B, Mendonca Santos M L, Minasny B. On digital soil mapping[J]. Geoderma, 2003, 17(1/2): 3-52. 
[11]Li Qiquan, Yue Tianxiang, Fan Zemeng, et al. Study on method for spatial simulation of topsoil SOM at national scale in China[J]. Journal of Natural Resources, 2010, 25(8): 1 385-1 399.[李启权, 岳天祥, 范泽孟,等.中国表层土壤有机质空间分布模拟分析方法研究[J]. 自然资源学报, 2010, 25(8): 1 385-1 399.]
[12]Al-Alawi S M, Abdul-Wahab S A, Bakheit C S. Combining principal component regression and artificial neural networks for more accurate predictions of ground-level ozone[J]. Environmental Modelling & Software, 2008, 23(4): 396-403.
[13]Sousa S I V, Martins F G, Alvim-Ferraz M C M, et al. Multiple linear regression and artificial neural networks based on principal components to predict ozone concentrations[J]. Environmental Modelling & Software, 2007, 22(1): 97-103.
[14]Erzin Y, Rao B H, Singh D N. Artificial neural network models for predicting soil thermal resistivity[J]. International Journal of Thermal Sciences, 2007, 47(10):1 347-1 358.
[15]Zou P, Yang J S, Fu J R, et al. Artificial neural network and time series models for predicting soil salt and water content[J]. Agricultural Water Management, 2010, 97(12): 2 009-2 019.
[16]Shen Zhangquan, Zhou Bin, Kong Fansheng, et al. Study on spatial variety of soil properties by means of generalized regression neural network[J]. Acta Pedologica Sinica, 2004, 41(3): 471-475.[沈掌泉, 周斌, 孔繁胜, 等. 应用广义回归神经网络进行土壤空间变异研究[J]. 土壤学报, 2004, 41(3): 471-475.]
[17]Behrens T, Förster H, Scholten T, et al. Digital soil mapping using artificial neural networks[J]. Journal of Plant Nutrition and Soil Science, 2005, 168(1): 21-33.
[18]Li Qiquan, Wang Changquan, Yue Tianxiang, et al. Method for spatial variety of soil organic matter based on radial basis function neural network[J]. Transactions of the CSAE, 2010, 26(1): 87-93.[李启权, 王昌全, 岳天祥, 等. 基于RBF神经网络的土壤有机质空间变异研究方法[J]. 农业工程学报, 2010, 26(1): 87-93.]
[19]Meersmans J, Van Wesemael B, Goidts E, et al. Spatial analysis of soil organic carbon evolution in Belgian croplands and grasslands, 1960-2006[J]. Global Change Biology, 2011, 17(1): 466-479.
[20]Pirie A, Singh B, Islam K. Ultra-violet, visible, near-infrared and mid-infrared diffuse reflectance spectroscopis techniques to predict several soil properties[J]. Australian Journal of Soil Research, 2005, 43(6): 713-721.
[21]Razakamanarivo R H, Grinand C, Razafindrakoto M A, et al. Mapping organic carbon stocks in eucalyptus plantations of the central highlands of Madagascar: A multiple regression approach[J]. Geoderma, 2011, 162(3/4): 335-346.
[22]Zhang Sumei, Wang Zongming, Zhang Bai, et al. Prediction of spatial distribution of soil nutrients using terrain attributes and remote sensing data[J]. Transactions of the CSAE, 2010, 26(5): 188-194.[张素梅, 王宗明, 张柏, 等. 利用地形和遥感数据预测土壤养分空间分布[J]. 农业工程学报, 2010, 26(5): 188-194.]
[23]Johnson C E, Ruiz-Méndez J J, Lawrence G B. Forest soil chemistry and terrain attributes in a Catskills Watershed[J]. Soil Science Society of America Journal, 2000, 64(5): 1 804-1 814.
[24]Schulp C J E, Verbrug P H. Effect of land use history and site factors on spatial variation of soil organic carbon across a physiographic region[J]. Agricultural, Ecosystems and Environment, 2009, 133(1/2): 86-97.
[25]Thompson J A, Eugenia M, Pena Y, et al. Soil-landscape modeling across a physiographic region: Topographic patterns and model transportability[J]. Geoderma, 2006, 133(1/2): 57-70.
[26]Meersmans J, De Ridder F, Canters F, et al. A multiple regression approach to assess the spatial distribution of Soil Organic Carbon (SOC) at the regional scale (Flanders, Belgium)[J]. Geoderma, 2008, 143(1): 1-13.
[1] 潘根兴, 丁元君, 陈硕桐, 孙景玲, 冯潇, 张晨, 郑聚锋, 张旭辉, 程琨, 刘晓雨, 卞荣军, 李恋卿. 从土壤腐殖质分组到分子有机质组学认识土壤有机质本质[J]. 地球科学进展, 2019, 34(5): 451-470.
[2] 赵同谦，张华，徐华山，贺玉晓. 黄河湿地孟津段不同植物群落类型土壤有机质含量变化特征研究[J]. 地球科学进展, 2008, 23(6): 638-643.
[3] 吴乐知,蔡祖聪. 中国土壤有机质含量变异性与空间尺度的关系[J]. 地球科学进展, 2006, 21(9): 965-972.
[4] 饶志国,朱照宇,陈发虎,张家武. 黄土有机质稳定碳同位素研究进展[J]. 地球科学进展, 2006, 21(1): 62-69.
[5] 陶贞;沈承德;易惟熙;高全洲. 土壤碳动力学同位素示踪研究进展[J]. 地球科学进展, 2004, 19(5): 793-801.
[6] 陈庆强，沈承德，易惟熙，彭少麟，李志安. 土壤碳循环研究进展[J]. 地球科学进展, 1998, 13(6): 555-563.
阅读次数
全文


摘要

网站版权所有 © 《地球科学进展》编辑部
地址：甘肃省兰州市天水中路8号
QQ：114723517
电话：0931-8762293　E-mail：adearth@lzb.ac.cn
陇ICP备2021001824号-5　 甘公网安备62010202000687