基于地表温度的干旱平缓区土壤属性制图
|
摘要
环境变量是数字土壤制图的重要支撑。在平原等地形平缓区,地形、植被等易于观测获取的环境变量与土壤条件的协同程度通常比较低,难以用其推测土壤空间分布。如何探索开发新的环境协同变量是平缓地区土壤制图的一个重要研究问题。不同的土壤条件往往具有不同的热量过程和特征。基于这一点,本文探讨了遥感获取的地表温度能多大程度上揭示土壤条件空间差异的问题。选取西北干旱区的黑河下游额济纳旗平原为研究区,基于MODIS传感器获取的地表温度资料和野外土壤调查样点,一方面对地表温度和土壤多要素属性的相关性进行了分析,另一方面建立随机森林模型对土壤有机碳、砂粒与粉粒含量进行空间推测制图,采用留一交叉法验证制图精度,并比较了仅用地形变量或地表温度变量、地形变量加上地表温度变量三种变量组合方案的土壤制图效果。结果显示,地表温度变量与土壤有机碳和土壤质地均具有较好的相关性,地表温度可解释研究区土壤条件空间变异的33～40%,其中,有机碳为41%,粉粒含量为37%,砂粒含量为33%。这表明,地表温度变量能够较大程度上有效地揭示土壤条件的空间差异,这为进一步对地表温度数据进行提炼,研发更为有效的环境变量,提高平缓区数字土壤制图的准确性提供了基础。
Environmental covariates have crucial supports for digital soil mapping. In areas with low relief,easy-to-obtain environmental factors(terrain and vegetation) do not co-vary with soil conditions across space to the level that they can be effectively used in digital soil mapping. A challenge of predictive soil mapping in such areas is how to explore and develop a new environmental covariate. The different soil conditions have different heat transfer processes. Based on this point, the paper examined to what extent the land surface temperature obtained by remote sensing can reveal spatial differences of soil conditions. In this study, the Ejinaqi Plain was chosen as an example in the downstream Heihe river basin. Soil samples were collected from the soil survey, and land surface temperature images were from the moderate-resolution imaging spectroradiometer(MODIS) sensor. The random forest were used to model and map the spatial distribution of soil organic carbon, sand content and silt content. A leave-one-out cross validation method was used to assess prediction accuracy. Comparisons were conducted among the performances of three explanatory variables: terrain variable, land surface temperature, and terrain-temperature. The results showed that the land surface temperature strongly co-varied with the contents of soil organic carbon, sand and silt in this area.It could explain approximately 41% of soil organic carbon, 37% of silt content, and 33% of sand content. This indicated that the surface temperature variable can reveal the spatial variations of soil conditions to a big extent. This provides a basis for further developing more effective environmental variables and improving the accuracy of digital soil mapping over low-relief areas.
Environmental covariates have crucial supports for digital soil mapping. In areas with low relief,easy-to-obtain environmental factors(terrain and vegetation) do not co-vary with soil conditions across space to the level that they can be effectively used in digital soil mapping. A challenge of predictive soil mapping in such areas is how to explore and develop a new environmental covariate. The different soil conditions have different heat transfer processes. Based on this point, the paper examined to what extent the land surface temperature obtained by remote sensing can reveal spatial differences of soil conditions. In this study, the Ejinaqi Plain was chosen as an example in the downstream Heihe river basin. Soil samples were collected from the soil survey, and land surface temperature images were from the moderate-resolution imaging spectroradiometer(MODIS) sensor. The random forest were used to model and map the spatial distribution of soil organic carbon, sand content and silt content. A leave-one-out cross validation method was used to assess prediction accuracy. Comparisons were conducted among the performances of three explanatory variables: terrain variable, land surface temperature, and terrain-temperature. The results showed that the land surface temperature strongly co-varied with the contents of soil organic carbon, sand and silt in this area.It could explain approximately 41% of soil organic carbon, 37% of silt content, and 33% of sand content. This indicated that the surface temperature variable can reveal the spatial variations of soil conditions to a big extent. This provides a basis for further developing more effective environmental variables and improving the accuracy of digital soil mapping over low-relief areas.
引文
[1]MCBRATNEY A B,MENDONCA M L,MINASNY B.On digital soil mapping[J].Geoderma,2003,117(2003):3-52.
[2]AKUMU C E,JOHNSON J A,ETHERIDGE D et al.GIS-fuzzy logic based approach in modeling soil texture:using parts of the clay belt and Hornepayne region in Ontario Canada as a case study[J].Geoderma,2015,239-240(2015):13-24.
[3]LIU F,GENG X Y,ZHU A X et al.Soil polygon disaggregation through similarity-based prediction with legacy pedons[J].Journal of Arid Land,2016,8(5):760-772.
[4]刘峰,朱阿兴,李宝林,等.利用陆面反馈动态模式来识别土壤类型的空间差异[J].土壤通报,2009,40(3):501-508.
[5]ZHU A X,LIU F,LI B L et al.Differentiation of soil conditions over low relief areas using feedback dynamic patterns[J].Pedology,2010,74(3):861-869.
[6]LIU F,GENG X Y,ZHU A X et al.Soil texture mapping over low relief areas using land surface feedback dynamic patterns extracted from MODIS[J].Geoderma,2012,171(2012):44-52.
[7]WANG D C,ZHANG G L,PAN X Z et al.Mapping soil texture of a plain area using fuzzy-c-means clustering method based on land surface diurnal temperature difference[J].Pedosphere,2012,22(3):394-403.
[8]ZHAO M S,ROSSITER D G,LI D C et al.Mapping soil organic matter in low-relief areas based on land surface diurnal temperature difference and a vegetation index[J].Ecological Indicators,2014,39:120-133.
[9]ZENG C Y,ZHU A X,LIU F et al.The impact of rainfall magnitude on the performance of digital soil mapping over low-relief areas using a land surface dynamic feedback method[J].Ecological Indicators,2016,72(2017):297-309.
[10]YANG R M,LIU F,ZHANG G L et al.Mapping soil texture based on field soil moisture observations at a high temporal resolution in an oasis agricultural area[J].Pedosphere,2016,26(5):699-708.
[11]李毅,邵明安,王文焰.质地对土壤热性质的影响[J].农业工程学报,2003,19(4):62-65.
[12]ZHAO J K,REN T S,DU Z L,WANG Y D.Effects of biochar amendment on soil thermal properties in the north China plain[J].Soil Science Society of America Journal,2016,80(5):1157-1166.
[13]韩刚,李瑞平,岳胜如,等.基于地表温度植被指数特征空间的荒漠化草原表层土壤含水量反演[J].四川农业大学学报,2015,33(4):385-391.
[14]李润林,董鹏程,王瑜,等.基于地表温度植被指数特征空间的土壤干旱监测[J].湖北农业科学,2017,56(16):3060-3066.
[15]周俊利,薛华柱,董国涛,等.基于AMSR2微波植被指数与地表温度的旱情分析[J].水土保持研究,2018,25(6):1-7.
[16]TAKTIKOU E,BOURAZANIS G,PAPAIOANNOU G et al.Prediction of soil moisture from remote sensing data[J].Procedia Engineering,2016,162(2016)309-316.
[17]SANCHEZ N,PILES M.A new soil moisture agricultural drought index(SMADI)integrating MODIS and SMOS products:a case of study over the Iberian Peninsula[J].Remote Sensing,2016,8(4):287.
[18]内蒙古土地勘测设计院.内蒙古自治区额济纳旗土壤[M].1988.
[19]YANG L,ZHU A X,ZHAO Y G et al.Regional soil mapping using multi-grade representative sampling and a fuzzy membership based mapping approach[J].Pedosphere,2017,27(2):344-357.
[20]张甘霖,龚子同.土壤调查实验室分析方法[M].北京:科学出版社,2012.
[21]BREIMAN L.Random forests[J].Machine Learning,2001,45(1):5-32.
[22]MALONE B P,MINASNY B,MCBRATNEY A B.Using R for digital soil mapping[M].Switzerland:Springer Nature,2017.
[23]PICARD R,DENNISCOOK R.Cross-validation of regression models[J].Journal of the American Statistical Association,1984,79(387):575-583.
[24]吴昊.秦岭山地松栎混交林土壤养分空间变异及其与地形因子的关系[J].自然资源学报,2015,30(5):858-869.
[25]谢娟英,高红超.基于统计相关性与K-means的区分基因子集选择算法[J].软件学报,2014,25(9):2050-2075.
[26]邵明安,王全九,黄明斌.土壤物理学[M].北京:高等教育出版社,2006.
[27]JURY W,HORTON R.Soil physics[M].New York:John Wiley&sons,Inc.,2004.
[28]DEVRIES D A.Thermal properties of soils[M].Physics of the Plant Environment.Amsterdam:North-Holland,1963.
[2]AKUMU C E,JOHNSON J A,ETHERIDGE D et al.GIS-fuzzy logic based approach in modeling soil texture:using parts of the clay belt and Hornepayne region in Ontario Canada as a case study[J].Geoderma,2015,239-240(2015):13-24.
[3]LIU F,GENG X Y,ZHU A X et al.Soil polygon disaggregation through similarity-based prediction with legacy pedons[J].Journal of Arid Land,2016,8(5):760-772.
[4]刘峰,朱阿兴,李宝林,等.利用陆面反馈动态模式来识别土壤类型的空间差异[J].土壤通报,2009,40(3):501-508.
[5]ZHU A X,LIU F,LI B L et al.Differentiation of soil conditions over low relief areas using feedback dynamic patterns[J].Pedology,2010,74(3):861-869.
[6]LIU F,GENG X Y,ZHU A X et al.Soil texture mapping over low relief areas using land surface feedback dynamic patterns extracted from MODIS[J].Geoderma,2012,171(2012):44-52.
[7]WANG D C,ZHANG G L,PAN X Z et al.Mapping soil texture of a plain area using fuzzy-c-means clustering method based on land surface diurnal temperature difference[J].Pedosphere,2012,22(3):394-403.
[8]ZHAO M S,ROSSITER D G,LI D C et al.Mapping soil organic matter in low-relief areas based on land surface diurnal temperature difference and a vegetation index[J].Ecological Indicators,2014,39:120-133.
[9]ZENG C Y,ZHU A X,LIU F et al.The impact of rainfall magnitude on the performance of digital soil mapping over low-relief areas using a land surface dynamic feedback method[J].Ecological Indicators,2016,72(2017):297-309.
[10]YANG R M,LIU F,ZHANG G L et al.Mapping soil texture based on field soil moisture observations at a high temporal resolution in an oasis agricultural area[J].Pedosphere,2016,26(5):699-708.
[11]李毅,邵明安,王文焰.质地对土壤热性质的影响[J].农业工程学报,2003,19(4):62-65.
[12]ZHAO J K,REN T S,DU Z L,WANG Y D.Effects of biochar amendment on soil thermal properties in the north China plain[J].Soil Science Society of America Journal,2016,80(5):1157-1166.
[13]韩刚,李瑞平,岳胜如,等.基于地表温度植被指数特征空间的荒漠化草原表层土壤含水量反演[J].四川农业大学学报,2015,33(4):385-391.
[14]李润林,董鹏程,王瑜,等.基于地表温度植被指数特征空间的土壤干旱监测[J].湖北农业科学,2017,56(16):3060-3066.
[15]周俊利,薛华柱,董国涛,等.基于AMSR2微波植被指数与地表温度的旱情分析[J].水土保持研究,2018,25(6):1-7.
[16]TAKTIKOU E,BOURAZANIS G,PAPAIOANNOU G et al.Prediction of soil moisture from remote sensing data[J].Procedia Engineering,2016,162(2016)309-316.
[17]SANCHEZ N,PILES M.A new soil moisture agricultural drought index(SMADI)integrating MODIS and SMOS products:a case of study over the Iberian Peninsula[J].Remote Sensing,2016,8(4):287.
[18]内蒙古土地勘测设计院.内蒙古自治区额济纳旗土壤[M].1988.
[19]YANG L,ZHU A X,ZHAO Y G et al.Regional soil mapping using multi-grade representative sampling and a fuzzy membership based mapping approach[J].Pedosphere,2017,27(2):344-357.
[20]张甘霖,龚子同.土壤调查实验室分析方法[M].北京:科学出版社,2012.
[21]BREIMAN L.Random forests[J].Machine Learning,2001,45(1):5-32.
[22]MALONE B P,MINASNY B,MCBRATNEY A B.Using R for digital soil mapping[M].Switzerland:Springer Nature,2017.
[23]PICARD R,DENNISCOOK R.Cross-validation of regression models[J].Journal of the American Statistical Association,1984,79(387):575-583.
[24]吴昊.秦岭山地松栎混交林土壤养分空间变异及其与地形因子的关系[J].自然资源学报,2015,30(5):858-869.
[25]谢娟英,高红超.基于统计相关性与K-means的区分基因子集选择算法[J].软件学报,2014,25(9):2050-2075.
[26]邵明安,王全九,黄明斌.土壤物理学[M].北京:高等教育出版社,2006.
[27]JURY W,HORTON R.Soil physics[M].New York:John Wiley&sons,Inc.,2004.
[28]DEVRIES D A.Thermal properties of soils[M].Physics of the Plant Environment.Amsterdam:North-Holland,1963.