利用CARS算法联合协变量估算盐碱农田土壤水分和有机质含量

doi:10.13287/j.1001-9332.202405.021

应用生态学报 ›› 2023, Vol. 35 ›› Issue (5): 1321-1330.doi: 10.13287/j.1001-9332.202405.021

利用CARS算法联合协变量估算盐碱农田土壤水分和有机质含量

丁启东¹, 王怡婧², 张俊华^1,3,4*, 贾科利², 黄华雨¹

¹宁夏大学生态环境学院, 银川 750021;
²宁夏大学地理科学与规划学院, 银川 750021;
³西北退化生态系统恢复与重建教育部重点实验室, 银川 750021;
⁴西部土地退化与生态恢复国家重点实验室培育基地, 银川 750021

收稿日期:2023-11-02 接受日期:2024-03-11 出版日期:2024-05-18 发布日期:2024-11-18
通讯作者: *E-mail: zhangjunhua728@163.com
作者简介:丁启东, 男, 1995年生, 硕士研究生。主要从事遥感监测与土壤质量提升研究。E-mail: 2535643120@qq.com
基金资助:
国家重点研发计划项目(2021YFD1900602)、国家自然科学基金项目(42067003,42061047)和宁夏科技创新领军人才项目(2022GKLRLX02)

Estimation of soil moisture and organic matter content in saline alkali farmland by using CARS algorithm combined with covariates

DING Qidong¹, WANG Yijing², ZHANG Junhua^1,3,4*, JIA Keli², HUANG Huayu¹

¹College of Ecology and Environmental Science, Ningxia University, Yinchuan 750021, China;
²College of Geographical Sciences and Planning, Ningxia University, Yinchuan 750021, China;
³Key Laboratory for Restoration and Reconstruction of Degraded Ecosystem in Northwestern China of Ministry of Education, Yinchuan 750021, China;
⁴Breeding Base for State Key Laboratory of Land Degradation and Ecological Restoration in Northwestern China, Yinchuan 750021, China

Received:2023-11-02 Accepted:2024-03-11 Online:2024-05-18 Published:2024-11-18

摘要/Abstract

摘要： 快速获取土壤含水率(SMC)和土壤有机质(SOM)含量对于盐碱农田土壤的改良利用至关重要。本研究基于河套平原农田土壤野外高光谱反射率和土壤属性实测数据,对原始光谱反射率(Ref)进行标准正态变量(SNV)转换后,采用竞争性自适应重加权采样算法(CARS)筛选敏感波段,然后分别以Ref、Ref-SNV和Ref-SNV+土壤协变量(SC)及数字高程模型(DEM)作为建模输入变量的策略Ⅰ、Ⅱ和Ⅲ,基于随机森林(RF)和轻梯度提升机(LightGBM)建立SMC和SOM估算模型,并对模型精度进行验证和对比。结果表明:经CARS筛选后,SMC和SOM敏感波段压缩至全波段的3.3%以下,有效优化波段选择,减少了冗余光谱信息。与LightGBM 模型相比,RF模型在SMC和SOM估算中精度更高,输入变量策略Ⅲ优于Ⅱ和Ⅰ,辅助变量的引入有效提升了模型的估算能力。综合分析,基于策略Ⅲ-RF的SMC估算模型验证决定系数(R_p²)、均方根误差(RMSE)和相对分析误差(RPD)分别为0.63、3.16和2.01,基于策略Ⅲ-RF的SOM估算模型R_p²、RMSE和RPD分别为0.93、1.15和3.52,策略Ⅲ-RF模型是估算土壤水分和土壤有机质的有效方法。研究结论可为盐碱农田土壤水分和有机质含量快速估算提供新方法。

关键词: 高光谱, 遥感, 土壤协变量, 变量重要性投影, 随机森林, 轻梯度提升机, 反距离权重法

Abstract: Rapid acquisition of the data of soil moisture content (SMC) and soil organic matter (SOM) content is crucial for the improvement and utilization of saline alkali farmland soil. Based on field measurements of hyperspectral reflectance and soil properties of farmland soil in the Hetao Plain, we used a competitive adaptive reweighted sampling algorithm (CARS) to screen sensitive bands after transforming the original spectral reflectance (Ref) into a standard normal variable (SNV). Strategies Ⅰ, Ⅱ, and Ⅲ were used to model the input variables of Ref, Ref SNV, Ref-SNV+ soil covariate (SC), and digital elevation model (DEM). We constructed SMC and SOM estimation models based on random forest (RF) and light gradient boosting machine (LightGBM), and then verified and compared the accuracy of the models. The results showed that after CARS screening, the sensitive bands of SMC and SOM were compressed to below 3.3% of the entire band, which effectively optimized band selection and reduced redundant spectral information. Compared with the LightGBM model, the RF model had higher accuracy in SMC and SOM estimation, and the input variable strategy Ⅲ was better than Ⅱ and Ⅰ. The introduction of auxiliary variables effectively improved the estimation ability of the model. Based on comprehensive analysis, the coefficient of determination (R_p²), root mean square error (RMSE), and relative analysis error (RPD) of the SMC estimation model validation based on strategy Ⅲ-RF were 0.63, 3.16, and 2.01, respectively. The SOM estimation models based on strategy Ⅲ-RF had R_p², RMSE, and RPD of 0.93, 1.15, and 3.52, respectively. The strategy Ⅲ-RF model was an effective method for estimating SMC and SOM. Our results could provide a new method for the rapid estimation of soil moisture and organic matter content in saline alkali farmland.

Key words: hyperspectral, remote sense, soil covariate, variable importance projection, random forest, light gradient boosting machine, inverse distance weighting

丁启东, 王怡婧, 张俊华, 贾科利, 黄华雨. 利用CARS算法联合协变量估算盐碱农田土壤水分和有机质含量[J]. 应用生态学报, 2023, 35(5): 1321-1330.

DING Qidong, WANG Yijing, ZHANG Junhua, JIA Keli, HUANG Huayu. Estimation of soil moisture and organic matter content in saline alkali farmland by using CARS algorithm combined with covariates[J]. Chinese Journal of Applied Ecology, 2023, 35(5): 1321-1330.

参考文献

[1] Jia PP, Zhang JH, He W, et al. Combination of hyperspectral and machine learning to invert soil electrical conductivity. Remote Sensing, 2022, 14: 2602
[2] Liu QS, Wu ZJ, Cui NB, et al. Estimation of soil moisture using multi-source remote sensing and machine learning algorithms in farming, and of Northern China. Remote Sensing, 2023, 15: 4214
[3] 张俊华, 尚天浩, 陈睿华, 等. 基于光谱FOD与优化指数的银川平原土壤有机质含量反演. 农业机械学报, 2022, 53(11): 379-387
[4] Jia PP, Zhang JH, He W, et al. Inversion of different cultivated soil types' salinity using hyperspectral data and machine learning. Remote Sensing, 2022, 14: 5639
[5] Gholizadeh A, Saberioon M, Ben-Dor E, et al. Monitoring of selected soil contaminants using proximal and remote sensing techniques:Background, state-of-the-art and future perspectives. Critical Reviews in Environmental Science and Technology, 2018, 48: 243-278
[6] Nawar S, Buddenbaum H, Hill J, et al. Estimating the soil clay content and organic matter by means of different calibration methods of vis-NIR diffuse reflectance spectroscopy. Soil and Tillage Research, 2016, 155: 510-522
[7] Wang XP, Zhang F, Ding JL, et al. Estimation of soil salt content (SSC) in the ebinur lake wetland national nature reserve (ELWNNR), northwest China, based on a bootstrap-BP neural network model and optimal spectral indices. Science of the Total Environment, 2018, 615: 918-930
[8] Li HD, Liang YZ, Xu QS, et al. Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration. Analytica Chimica Acta, 2009, 648: 77-84
[9] 王怡婧, 丁启东, 张俊华, 等. 基于无人机高光谱遥感和机器学习的土壤水盐信息反演. 应用生态学报, 2023, 34(11): 3045-3052
[10] 尚天浩, 陈睿华, 张俊华, 等. 基于分数阶微分联合光谱指数估算银川平原土壤有机质含量. 应用生态学报, 2023, 34(3): 717-725
[11] 周洋, 赵小敏, 郭熙. 基于多源辅助变量和随机森林模型的表层土壤全氮分布预测. 土壤学报, 2022, 59(2): 451-460
[12] 马国林, 丁建丽, 张子鹏. 基于土壤协变量与VIS-NIR光谱估算土壤有机质含量的研究. 激光与光电子学进展, 2020, 57(19): 265-275
[13] 高凤杰, 马泉来, 韩文文, 等. 黑土丘陵区小流域土壤有机质空间变异及分布格局. 环境科学, 2016, 37(5): 1915-1922
[14] 唐海涛, 孟祥添, 苏循新, 等. 基于CARS算法的不同类型土壤有机质高光谱预测. 农业工程学报, 2021, 37(2): 105-113
[15] Khalil M, AlSayed A, Liu Y, et al. Machine learning for modeling N₂O emissions from wastewater treatment plants: Aligning model performance, complexity, and interpretability. Water Research, 2023, 245: 120667
[16] Xu XT, Xiao C, Dong YB, et al. Machine learning algorithms realized soil stoichiometry prediction and its driver identification in intensive agroecosystems across a north-south transect of eastern China. Science of the Total Environment, 2024, 906: 167488
[17] 王思楠, 李瑞平, 吴英杰, 等. 基于环境变量和机器学习的土壤水分反演模型研究. 农业机械学报, 2022, 53(5): 332-341
[18] 顾永昇, 丁建丽, 韩礼敬, 等. 基于多源环境变量的渭-库绿洲土壤颗粒含量预测研究. 土壤, 2023, 55(2): 426-432
[19] 贾壮壮, 谭亚男, 管孝艳, 等. 宁夏盐碱地成因及分区治理措施综述. 灌溉排水学报, 2023, 42(5): 122-134
[20] Zhang JH, Ding QD, Wang YJ, et al. Soil quality assessment and constraint diagnosis of salinized farmland in the yellow river irrigation area in northwestern China. Geoderma Regional, 2023, 34: e00684
[21] 鲍士旦. 土壤农化分析(第三版). 北京: 中国农业出版社, 2000
[22] 刘焕军, 鲍依临, 徐梦园, 等. 基于SOM和NDVI的黑土区精准管理分区对比. 农业工程学报, 2019, 35(13): 177-183
[23] 纪文君, 史舟, 周清, 等. 几种不同类型土壤的VIS-NIR光谱特性及有机质响应波段. 红外与毫米波学报, 2012, 31(3): 277-282
[24] 李小雨, 贾科利, 魏慧敏, 等. 基于随机森林算法的土壤含盐量预测. 干旱区研究, 2023, 40(8): 1258-1267
[25] Li B, Liu K, Wang M, et al. High-spatiotemporal-resolution dynamic water monitoring using LightGBM model and Sentinel-2 MSI data. International Journal of Applied Earth Observation and Geoinformation, 2023, 118: 103278
[26] 贾萍萍, 孙媛, 尚天浩, 等. 基于高光谱和Landsat-8 OLI影像的盐渍化土壤水盐估算模型构建. 生态学杂志, 2020, 39(7): 2456-2466
[27] Gu XH, Wang YC, Sun Q, et al. Hyperspectral inversion of soil organic matter content in cultivated land based on wavelet transform. Computers and Electronics in Agriculture, 2019, 167: 105053
[28] Gholizadeh A, Boruvka L, Saberioon M, et al. Comparing different data preprocessing methods for monitoring soil heavy metals based on soil spectral features. Soil and Water Research, 2015, 10: 218-227
[29] 于雷, 朱亚星, 洪永胜, 等. 高光谱技术结合CARS算法预测土壤水分含量. 农业工程学报, 2016, 32(22): 138-145
[30] 蔡亮红, 丁建丽. 基于高光谱多尺度分解的土壤含水量反演. 激光与光电子学进展, 2018, 55(1): 406-415
[31] 尚天浩, 贾萍萍, 孙媛, 等. 宁夏银北地区盐碱化土壤水分光谱特征及模型拟合精度分析. 水土保持通报, 2020, 40(4): 183-189
[32] Mukhamediev RI, Merembayev T, Kuchin Y, et al. Soil salinity estimation for south Kazakhstan based on SAR Sentinel-1 and Landsat-8,9 OLI data with machine learning models. Remote Sensing, 2023, 15: 4269
[33] 张笑寒, 孟祥添, 唐海涛, 等. 优化光谱输入量的土壤有机质随机森林预测模型. 农业工程学报, 2023, 39(2): 90-99
[34] Qu LL, Lu HZ, Tian ZY, et al. Spatial prediction of soil sand content at various sampling density based on geostatistical and machine learning algorithms in plain areas. Catena, 2024, 234: 107572
[35] 刘婷玥, 代晶晶, 赵元艺, 等. 基于LightGBM的盐湖锂浓度遥感反演研究: 以西藏扎布耶盐湖北湖为例. 地质学报, 2021, 95(7): 2249-2256

利用CARS算法联合协变量估算盐碱农田土壤水分和有机质含量

Estimation of soil moisture and organic matter content in saline alkali farmland by using CARS algorithm combined with covariates

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	徐亮亮, 马开森, 王霞, 李东胜, 孙华. LA-UNet网络模型在城市绿地遥感分类中的应用 [J]. 应用生态学报, 2024, 35(4): 1101-1111.
[2]	王进, 周广胜, 何奇瑾, 周莉. 内蒙古克氏针茅草原净生态系统碳交换的物候特征及遥感监测 [J]. 应用生态学报, 2024, 35(3): 659-668.
[3]	赵允格, 吉静怡, 张万涛, 明姣, 黄琬雲, 高丽倩. 黄土高原生物土壤结皮分布时空分异特征 [J]. 应用生态学报, 2024, 35(3): 739-748.
[4]	袁佳玉, 熊立, 吴志伟, 朱诗豪, 康平, 李顺. 江西省赣州市南康区松材线虫病发生特征 [J]. 应用生态学报, 2024, 35(2): 507-515.
[5]	黄俊杰, 冯秀丽, 董毓伊, 张驰, 谢立建, 程俊恺, 高天宇. 基于遥感生态指数和图论知识的宁波市生态安全格局构建 [J]. 应用生态学报, 2023, 34(9): 2489-2497.
[6]	任孟林, 郭妍, 陈伯轩, 范佳乐, 胡同欣, 孙龙. 红松人工林地表可燃物火蔓延速率预测模型 [J]. 应用生态学报, 2023, 34(8): 2091-2100.
[7]	杨涛, 于颖, 杨曦光, 杜红萱. 无人机高光谱联合LiDAR估测林分与单木尺度叶绿素含量 [J]. 应用生态学报, 2023, 34(8): 2101-2112.
[8]	周文武, 舒清态, 王书伟, 杨正道, 罗绍龙, 胥丽, 肖劲楠. 基于多源遥感数据协同的滇西北森林郁闭度估测 [J]. 应用生态学报, 2023, 34(7): 1806-1816.
[9]	王怡婧, 陈睿华, 张俊华, 丁启东, 李小林. 基于分数阶微分技术的土壤水盐信息高光谱反演 [J]. 应用生态学报, 2023, 34(5): 1384-1394.
[10]	付绍桐, 贺晨曦, 马佳凯, 王犇, 甄志磊. 不同发展情景下黄河流域山西段生态环境质量 [J]. 应用生态学报, 2023, 35(5): 1337-1346.
[11]	顾晨, 梁建, 刘旭颖, 孙泊远, 孙同生, 俞坚钢, 孙晨曦, 万华伟, 高吉喜. 高光谱遥感在草原植物多样性监测中的应用与展望 [J]. 应用生态学报, 2023, 35(5): 1397-1407.
[12]	杨玉丽, 肖辉杰, 辛智鸣, 范光鹏, 李俊然, 贾肖肖, 汪立韬. 基于激光雷达与高光谱的荒漠绿洲农田防护林衰退程度评估 [J]. 应用生态学报, 2023, 34(4): 1043-1050.
[13]	瞿植, 罗漫雅, 赵永华, 杨舒媛, 韩磊, 穆琪. 黄河流域大型天然湖泊面积与岸线形态的时空动态 [J]. 应用生态学报, 2023, 34(4): 1102-1108.
[14]	冯平, 杨妮娟, 李建柱. 滦河流域遥感生态指数改进及生态环境质量评价 [J]. 应用生态学报, 2023, 34(12): 3195-3202.
[15]	陈悦, 赵庚星, 常春艳, 王卓然, 李因帅, 赵环三, 张术伟, 潘敬瑞. 基于Sentinel-2多光谱影像的小麦-玉米轮作耕地粮食产量估测——以曹县为例 [J]. 应用生态学报, 2023, 34(12): 3347-3356.