不同土地利用类型下土壤有机碳含量的高光谱反演

doi:10.13287/j.1001-9332.202003.014

摘要/Abstract

摘要： 不同土地利用类型下土壤光谱信息存在差异,了解不同土地利用类型下合适的建模方法可以高效准确地进行土壤有机碳含量反演。本研究以江西省奉新县中北部林地、耕地和园地3种土地利用类型共248个土壤样本为对象,首先对土壤原始光谱反射率曲线使用Savitzky-Golay(SG)滤波去噪并进行10 nm重采样减少数据冗余,之后采用偏最小二乘回归(PLSR)、基于网格搜索法的支持向量机回归(GRID-SVR)和基于粒子群算法的支持向量机回归(PSO-SVR)3种方法分别构建土壤有机碳含量的反演模型。结果表明: 构建单一土地利用类型反演模型时,PLSR方法在林地、耕地和园地的相对分析误差(RPD)分别为1.536、1.315和1.493,采用GRID-SVR方法时,其RPD分别提升0.150、0.183和0.502。采用PSO-SVR方法时精度最高,相较GRID-SVR方法,其林地、耕地和园地的RPD分别提高20.8%、10.0%和2.7%,林地和园地的RPD分别为2.036和2.049,可以极好地预测土壤有机碳含量,耕地的RPD为1.647,可以对土壤有机碳含量进行粗略估测。PSO-SVR方法对不同土地利用类型土壤有机碳反演效果最优,林地和园地土壤有机碳含量的反演精度相近且高于耕地。研究区不同土地利用类型对土壤有机碳含量的反演结果存在一定的影响,今后可以考虑在反演土壤有机碳时分不同土地利用类型进行建模。

Abstract: Soil spectral information differ across different land use types. Understanding the appropriate modeling methods for different land use types can efficiently and accurately invert soil organic carbon content. We collected 248 samples from forest, cultivated land and orchard in the north-central part of Fengxin County, Jiangxi Province. First, original spectral reflectance curves were reduced noises with Savitzky-Golay (SG) filter. Then 10 nm resampling method was used to reduce data redundancy. We used partial least squares regression (PLSR), support vector machine regression based on grid search method (GRID-SVR) and support vector machine regression based on particle swarm optimization (PSO-SVR) to construct the inversion models of soil organic carbon content. The results showed that when constructing a single land-use type inversion model, RPD of the PLSR method for forest, cultivated land and orchard was 1.536, 1.315 and 1.493 respectively. RPD of GRID-SVR method increased 0.150, 0.183 and 0.502 than that of PLSR method, respectively. The PSO-SVR method had higher accuracy, with RPD being 20.8%, 10.0% and 2.7% higher than GRID-SVR for forest, cultivated land and orchard, respectively. The RPD of forest and orchard were 2.036 and 2.049, which well predicts soil organic carbon. The RPD of cultivated land was 1.647, which can make a rough estimate of soil organic carbon. The PSO-SVR model had the best prediction effect on soil organic carbon of different land use types, with the prediction accuracy of soil organic carbon content in forest and orchard being close and higher than cultivated land. Soil nutrition diffed acorss different land use types, which affect the prediction of soil organic carbon content. Models for inversion of soil organic carbon should be constructed separately for different land use types.

国佳欣, 朱青, 赵小敏, 郭熙, 韩逸, 徐喆. 不同土地利用类型下土壤有机碳含量的高光谱反演[J]. 应用生态学报, 2020, 31(3): 863-871.

GUO Jia-xin, ZHU Qing, ZHAO Xiao-min, GUO Xi, HAN Yi, XU Zhe. Hyper-spectral inversion of soil organic carbon content under different land use types[J]. Chinese Journal of Applied Ecology, 2020, 31(3): 863-871.

参考文献 39

[1]	黄昌勇, 徐建明. 土壤学. 北京: 中国农业出版社, 2010 [Huang C-Y, Xu J-M. Soil Science. Beijing: China Agricultural Press, 2010]
[2]	周慧珍, 龚子同. 土壤空间变异性研究. 土壤学报, 1996, 33(3): 232-241 [Zhou H-Z, Gong Z-T. Study on soil spatial variability. Acta Pedologica Sinica, 1996, 33(3): 232-241]
[3]	张娟娟, 田永超, 朱艳, 等. 不同类型土壤的光谱特征及其有机质含量预测. 中国农业科学, 2009, 42(9): 3154-3163 [Zhang J-J, Tian Y-C, Zhu Y, et al. Spectral characteristics and estimation of organic matter contents of different soil types. Scientia Agricultura Sinica, 2009, 42(9): 3154-3163]
[4]	孙问娟, 李新举. 煤矿区土壤有机碳含量的高光谱预测模型. 水土保持学报, 2018, 32(5): 346-351 [Sun W-J, Li X-J. Hyperspectral prediction model of soil organic carbon content in coal mining area. Journal of Soil and Water Conservation, 2018, 32(5): 346-351]
[5]	崔霞, 宋清洁, 张瑶瑶, 等. 基于高光谱数据的高寒草地土壤有机碳预测模型研究. 草业学报, 2017, 26(10): 20-29 [Cui X, Song Q-J, Zhang Y-Y, et al. Estimation of soil organic carbon content in alpine grassland using hyperspectral data. Acta Prataculturae Sinica, 2017, 26(10): 20-29]
[6]	Conforti M, Buttafuoco G, Leone AP, et al. Studying the relationship between water-induced soil erosion and soil organic matter using Vis-NIR spectroscopy and geomorphological analysis: A case study in southern Italy. Catena, 2013, 110: 44-58
[7]	于仕兴. 基于智能算法的支持向量机结合木材近红外光谱应用研究. 硕士论文. 北京: 北京林业大学, 2014 [Yu S-X. Support Vector Machine (SVM) Based on Intelligent Algorithm Combined with Wood NIR Application Research. Master Thesis. Beijing: Beijing Forestry University, 2014]
[8]	汪洪涛. 基于近红外光谱技术预测森林土壤有机碳含量的研究. 硕士论文. 哈尔滨: 东北林业大学, 2014 [Wang H-T. Predicting of Forest Soil Organic Carbon Content Based on Near Infrared Spectroscopy. Master Thesis. Harbin: Northeast Forestry University, 2014]
[9]	陈元鹏, 张世文, 罗明, 等. 基于高光谱反演的复垦区土壤重金属含量经验模型优选. 农业机械学报, 2019, 50(1): 170-179 [Chen Y-P, Zhang S-W, Luo M, et al. Empirical model optimization of hyperspectral inversion of heavy metal content in reclamation area. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(1): 170-179]
[10]	邹慧敏, 李西灿, 尚璇, 等. 粒子群优化神经网络的土壤有机质高光谱估测. 测绘科学, 2019, 44(5): 146-150 [Zou H-M, Li X-C, Shang X, et al. Hyperspectral estimation of soil organic matter based on particle swarm optimization neural network. Science of Surve-ying and Mapping, 2019, 44(5): 146-150]
[11]	刘焕军, 吴炳方, 赵春江, 等. 光谱分辨率对黑土有机质预测模型的影响. 光谱学与光谱分析, 2012, 32(3): 739-742 [Liu H-J, Wu B-F, Zhao C-J, et al. Effect of spectral resolution on black soil organic matter content predicting model based on laboratory reflectance. Spectroscopy and Spectral Analysis, 2012, 32(3): 739-742]
[12]	蒋烨林, 王让会, 李焱, 等. 艾比湖流域不同土地覆盖类型土壤养分高光谱反演模型研究. 中国生态农业学报, 2016, 24(11): 1555-1564 [Jiang Y-L, Wang R-H, Li Y, et al. Hyper-spectral retrieval of soil nutrient content of various land-cover types in Ebinur Lake Basin. Chinese Journal of Eco-Agriculture, 2016, 24(11): 1555-1564]
[13]	薛利红, 周鼎浩, 李颖, 等. 不同利用方式下土壤有机质和全磷的可见近红外高光谱反演. 土壤学报, 2014, 51(5): 993-1002 [Xue L-H, Zhou D-H, Li Y, et al. Prediction of soil organic matter and total phosphorus with VIS-NIR hyperspectral inversion relative to land use. Acta Pedologica Sinica, 2014, 51(5): 993-1002]
[14]	李龙, 秦富仓, 姜丽娜, 等. 土地利用方式和地形对半干旱区土壤有机碳含量的影响. 土壤, 2019, 51(2): 406-412 [Li L, Qin F-C, Jiang L-N, et al. Effects of land use type and terrain on soil organic carbon (SOC) content in semi-arid region. Soils, 2019, 51(2): 406-412]
[15]	鲍士旦. 土壤农化分析. 北京: 中国农业出版社, 1999 [Bao S-D. Soil and Agricultural Chemistry Analysis. Beijing: China Agriculture Press, 1999]
[16]	王惠文. 偏最小二乘回归方法及其应用. 北京: 国防工业出版社, 1999 [Wang H-W. Partial Least-squares Regression: Methods and Applications. Beijing: National Defense Industry Press, 1999]
[17]	Xu SX, Shi XZ, Wang MY, et al. Effects of subsetting by parent materials on prediction of soil organic matter content in a hilly area using Vis-NIR spectroscopy. PLoS One, 2016, 11(3): e0151536
[18]	于雷, 洪永胜, 耿雷, 等. 基于偏最小二乘回归的土壤有机质含量高光谱估算. 农业工程学报, 2015, 31(14): 103-109 [Yu L, Hong Y-S, Geng L, et al. Hyper-spectral estimation of soil organic matter content based on partial least squares regression. Transactions of the Chinese Society of Agricultural Engineering, 2015, 31(14): 103-109]
[19]	史舟. 土壤地面高光谱遥感原理与方法. 北京: 科学出版社, 2014 [Shi Z. Principle and Method of Soil Surface Hyperspectral Remote Sensing. Beijing: Science Press, 2014]
[20]	Cristianini N, Shawe TJ, Lodhi H. Latent Semantic Kernels. Journal of Intelligent Information Systems, 2002, 18: 127-152
[21]	王文剑, 门昌骞. 支持向量机建模及应用. 北京: 科学出版社, 2014 [Wang W-J, Men C-Q. Support Vector Machine Modeling and Application. Beijing: Science Press, 2014]
[22]	Yang DL, Liu YL, Li SB, et al. Gear fault diagnosis based on support vector machine optimized by artificial bee colony algorithm. Mechanism and Machine Theory, 2015, 90: 219-229
[23]	曲健, 陈红岩, 刘文贞, 等. 基于改进网格搜索法的支持向量机在气体定量分析中的应用. 传感技术学报, 2015, 28(5): 774-778 [Qu J, Chen H-Y, Liu W-Z, et al. Application of support vector machine based on improved grid search in quantitative analysis of gas. Chinese Journal of Sensors and Actuators, 2015, 28(5): 774-778]
[24]	赵乃刚, 邓景顺. 粒子群优化算法综述. 科技创新导报, 2015, 12(26): 216-217 [Zhao N-G, Deng J-S. Survey of particle swarm optimization algorithms. Science and Technology Innovation Herald, 2015, 12(26): 216-217]
[25]	王东, 吴湘滨. 利用粒子群算法优化SVM分类器的超参数. 计算机应用, 2008, 28(1): 134-135 [Wang D, Wu X-B. Utilizing particle swarm optimization to optimize hyper-parameters of SVM classifier. Journal of Computer Applications, 2008, 28(1): 134-135]
[26]	Kennedy J, Eberhart R. Particle swarm optimization. Proceedings of the IEEE International Conference on Neural Networks, Perth, Australia: IEEE, 1995, 4: 1942-1948
[27]	Chow CK, Tsui HT, Lee T. Surface registration using a dynamic genetic algorithm. Pattern Recognition, 2004, 37: 105-117
[28]	戴文智, 杨新乐. 基于惯性权重对数递减的粒子群优化算法. 计算机工程与应用, 2015, 51(17): 14-19 [Dai W-Z, Yang X-L. Particle swarm optimization algorithm based on inertia weight logarithmic decreasing. Computer Engineering and Applications, 2015, 51(17): 14-19]
[29]	史舟, 王乾龙, 彭杰, 等. 中国主要土壤高光谱反射特性分类与有机质光谱预测模型. 中国科学:地球科学, 2014, 44(5): 978-988 [Shi Z, Wang Q-L, Peng J, et al. Development of a national VNIR soil-spectral library for soil classification and prediction of organic matter concentrations. Science China: Earth Sciences, 2014, 44(5): 978-988]
[30]	O’Rourke SM, Holden NM. Determination of soil organic matter and carbon fractions in forest top soils using spectral data acquired from visible-near infrared hyperspectral images. Soil Science Society of America Journal, 2012, 76: 586-565
[31]	王丹丹, 史学正, 于东升, 等. 东北地区旱地土壤有机碳密度的主控自然因素研究. 生态环境学报, 2009, 18(3): 1049-1053 [Wang D-D, Shi X-Z, Yu D-S, et al. Main natural factors controlling soil organic carbon density in the uplands of Northeast China. Ecology and Environmental Sciences, 2009, 18(3): 1049-1053]
[32]	赵小敏, 杨梅花. 江西省红壤地区主要土壤类型的高光谱特性研究. 土壤学报, 2018, 55(1): 31-42 [Zhao X-M, Yang M-H. Hyper-spectral characteristics of major types of soils in red soil region of Jiangxi pro-vince. Acta Pedologica Sinica, 2018, 55(1): 31-42]
[33]	彭杰, 李曦, 周清, 等. 氧化铁对有机质光谱特性的影响分析. 遥感学报, 2013, 17(6): 1396-1412 [Peng J, Li X, Zhou Q, et al. Influence of iron oxide on the spectral characteristics of organic matter. Journal of Remote Sensing, 2013, 17(6): 1396-1412]
[34]	Radim V, Radka K, Lubo B, et al. Consideration of peak parameters derived from continuum-removed spectra to predict extractable nutrients in soils with visible and near-infrared diffuse reflectance spectroscopy (VNIR-DRS). Geoderma, 2014, 232-234: 208-218
[35]	徐锋, 邹早建, 徐小卡, 等. 基于支持向量机的船舶操纵运动黑箱建模. 北京航空航天大学学报, 2013, 39(11): 1553-1557 [Xu F, Zou Z-J, Xu X-K, et al. Black-box modeling of ship manoeuvring motion based on support vector machines. Journal of Beijing University of Aeronautics and Astronautics, 2013, 39(11): 1553-1557]
[36]	谭琨, 张倩倩, 曹茜, 等. 基于粒子群优化支持向量机的矿区土壤有机质含量高光谱反演. 中国地质大学学报, 2015, 40(8): 1339-1345 [Tan K, Zhang Q-Q, Cao Q, et al. Hyperspectral retrieval model of soil organic matter content based on particle swarm optimization-support vector machines. Journal of China University of Geosciences, 2015, 40(8): 1339-1345]
[37]	Stenberg B, Rossel RAV, Mouazen AM, et al. Visible and near infrared spectroscopy in soil science. Advances in Agronomy, 2010, 107: 163-215
[38]	曹升賡. 江西地区红壤性水稻土形成的特点. 土壤学报, 1964, 12(2): 155-163 [Cao S-G. The peculiarity of soil formation of the paddy soils derived from red earth in Jiangxi. Acta Pedologica Sinica, 1964, 12(2): 155-163]
[39]	邬登巍, 张甘霖. 母质与土地利用类型对土壤光谱反演模型的影响. 土壤, 2016, 48(1): 173-179 [Wu D-W, Zhang G-L. Effects of parent materials and land use types on inversion models by using soil spectral data. Soils, 2016, 48(1): 173-179]