1.College of Land Science and Technology, China Agricultural University, Beijing 100193, China
2.Land Consolidation Center of the Ministry of Natural Resources, Beijing 100035, China
Show less
文章历史+
Received
Published
2023-06-19
2024-02-01
Issue Date
2026-04-15
PDF (4288K)
摘要
为及时、准确地应用多源遥感数据提取干旱和半干旱区域土壤盐渍化反演特征及获取土壤盐渍化程度的空间分布数据,以吉林省西部大安市为研究区,利用Sentinel-1 SAR、Sentinel-2 MSI多源遥感数据和DEM数据,构建土壤盐分含量(Soil Salt Content,SSC)反演特征集,结合BorutaShap算法优选特征,通过耦合地理加权回归(Geographically weighted regression,GWR)与极限梯度提升树(XGBoost)回归构建土壤盐渍化反演模型,并与XGBoost回归、GWR反演结果对比分析。结果表明:SSC反演特征集中,盐分指数、植被指数在BorutaShap算法中取得了较高的重要性排名,是大安市SSC反演的重要特征。GWR模型的R2和RMSE分别为0.48和4.83 g/kg,XGBoost回归模型的R2和RMSE分别为0.54和4.35 g/kg,耦合GWR与XGBoost回归构建的土壤盐渍化反演模型预测精度得到显著提高,R2与RMSE分别达到0.63和3.71 g/kg。依据该模型反演结果,大安市各类盐渍土分布存在较强的空间异质性,土壤盐分含量呈现出由东南向西北逐渐递减的趋势,与实地调查基本一致。综上,耦合GWR与XGBoost回归模型充分考虑了反演特征的空间异质性和非线性关系,可有效提高SSC反演精度,可获得更符合实际的SSC空间分布,可用于干旱和半干旱地区土壤盐分含量的反演。
Abstract
In order to timely and accurately applying multi-source remote sensing data to extract soil salinization inversion characteristics in arid and semi-arid areas and obtain spatial distribution data of soil salinization degree. Taking Da'an City in western Jilin Province as the research area, Sentinel-1 SAR, Sentinel-2 MSI multi-source remote sensing data and DEM data were used to construct a soil salt content (SSC) inversion feature set. Combined with the BorutaShap algorithm to optimize features, the coupling Geographically Weighted Regression (GWR) with XGBoost regression was used to construct a soil salinization inversion model. The results were compared with XGBoost regression and GWR inversion results. The results show that: The SSC inversion features of this study are concentrated. Salinity index and vegetation index achieve high importance ranking in the BorutaShap algorithm and are important features of SSC inversion in Da’an City. In the soil salinization inversion model constructed, the R2 and RMSE of the GWR model are 0.48 and 4.83 g/kg, respectively, and the R2 and RMSE of the XGBoost regression model are 0.54 and 4.35 g/kg, respectively. The prediction accuracy of the soil salinization inversion model constructed by coupling GWR and XGBoost regression is significantly improved, with R2 and RMSE reaching 0.63 and 3.71 g/kg, respectively. According to the inversion results of this model, there is strong spatial heterogeneity in the distribution of various types of saline soil in Da’an City. The SSC shows a gradually decreasing trend from southeast to northwest, which is basically consistent with the field survey. In summary, the coupled GWR and XGBoost regression model constructed in this study fully considers the spatial heterogeneity and nonlinear relationship of the inversion characteristics, which effectively improves the SSC inversion accuracy. A more realistic SSC spatial distribution can be obtained. This model can be used for the inversion of soil salinity content in arid and semi-arid areas.
DaliakopoulosI N, TsanisI K, KoutroulisA, KourgialasN N, VarouchakisA E, KaratzasG P, RitsemaC J. The threat of soil salinity: A European scale review[J]. Science of the Total Environment, 2016, 573: 727-739
[2]
杨劲松.中国盐渍土研究的发展历程与展望[J].土壤学报,2008,45(5):837-845
[3]
YangJ S. Development process and prospect of the research on salt-affected soils in China[J]. Acta Pedologica Sinica, 2008, 45(5):837-845 (in Chinese)
YuW J, SunD F, QuW. Study of the characteristics of soil salinization based on redundancy analysis in Pingluo, Ningxia[J]. Journal of China Agricultural University, 2015, 20(5): 103-110 (in Chinese)
[8]
MetternichtG I, ZinckJ A. Remote sensing of soil salinity: potentials and constraints[J]. Remote Sensing of Environment, 2003, 85(1): 1-20
[9]
牛东玲,王启基.盐碱地治理研究进展[J].土壤通报,2002,33(6):449-455
[10]
NiuD L, WangQ J. Research progress on saline-alkali field control[J]. Chinese Journal of Soil Science, 2002,33(6): 449-455 (in Chinese)
WangJ Z, XueZ H, ZhangC, ChangY R. Spatial evolution off saline-alkali cultivated land and its impact on productivity in Hetao plain of inner Mongolia[J]. Geographical Science, 2019, 39(5):827-835 (in Chinese)
WangY J, JiaP P, ChenR H, ZhangJ H. Sensitivity analysis and quantitative inversion of multi-source remote sensing to soil salt content in dry and wet seasons in Ningxia[J/OL]. Journal of Ecology, (--). DOI:10.13292/j.1000-4890.202309.006 (in Chinese)
LuoB, LiuX, GuoP. Remote sensing drought monitoring of Hetao irrigation area based on MODIS data [J]. Journal of China Agricultural University, 2020, 25(10): 44-54 (in Chinese)
[17]
ChenB L, ZhengH W, LuoG, ChenC S, BaoA, LiuT, ChenX. Adaptive estimation of multi-regional soil salinization using extreme gradient boosting with Bayesian TPE optimization[J]. International Journal of Remote Sensing, 2022, 43(3): 778-811
YangL B, ChenC B, ZhengH W, LuoG P, ShangB J, HellwichO. Retrieval of soil salinity content based on random forests regression optimized by Bayesian optimization algorithm and genetic algorithm[J]. Journal of Geo-Information Science, 2021,23(9):1662-1674 (in Chinese)
[20]
GeX Y, DingJ L, TengD X, WangJ Z, HuoT C, JinX Y, WangJ J, HeB Z, HanL J. Updated soil salinity with fine spatial resolution and high accuracy: The synergy of Sentinel-2 MSI, environmental covariates and hybrid machine learning approaches [J/OL]. CATENA, (--). DOI:10.1016/j.catena.2022.106054
[21]
WangY, HuP, ZhengH W, LiuY, CaoX W, HellwichO, LiuT, LuoG, BaoA M, ChenX. Integrative modeling of heterogeneous soil salinity using sparse ground samples and remote sensing images[J/OL]. Geoderma, (--). DOI:10.1016/j.geoderma.2022.116321
[22]
XuH T, ChenC B, ZhengH W, LuoG P, YangL A, WangW S, WuS X, DingJ L. AGA-SVR-based selection of feature subsets and optimization of parameter in regional soil salinization monitoring [J]. International Journal of Remote Sensing, 2020, 41(12): 4470-4495
[23]
ZhangQ Q, LiL, SunR Z, ZhuD H, ZhangC, ChenQ Q. Retrieval of the soil salinity from Sentinel-1 Dual-Polarized SAR data based on deep neural network regression [J/OL]. IEEE Geoscience and Remote Sensing Letters, (--).DOI:10.1109/LGRS.2020.3041059
[24]
BrungardC W, BoettingerJ L, DuniwayM C, WillsS A, Edwards JrT C. Machine learning for predicting soil classes in three semi-arid landscapes[J]. Geoderma, 2015, 239-240: 68-83
MengL N, DingJ L, WangJ Z, GeX Y. Spatial distribution of soil salinity in the Ugan -Kuqa River delta oasis based on environmental variables[J]. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(1): 175-181 (in Chinese)
[27]
GeorganosS, GrippaT, Niang GadiagaA, LinardC, LennertM, VanhuysseS, MbogaN, WolffE, KalogirouS. Geographical random forests: A spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling[J]. Geocarto International, 2021, 36(2): 121-136
[28]
LiL F. Geographically weighted machine learning and downscaling for high-resolution spatiotemporal estimations of wind speed[J/OL]. Remote Sensing, (--). DOI: 10.3390/rs11111378
[29]
QuiñonesS, GoyalA, AhmedZ U. Geographically weighted machine learning model for untangling spatial heterogeneity of type 2 diabetes mellitus (T2D) prevalence in the USA [J/OL]. Scientific Reports, (--). DOI: 10.1016/j.annepidem.2017.11.002
[30]
ShahnehM R, OymakS, MagdyA. A-gwr: Fast and accurate geospatial inference via augmented geographically weighted regression [C].//Proceedings of the 29th international conference on advances in geographic information systems. November 2-5, 2021, New York, New York State, USA. Association for Computing Machinery:2021
[31]
PengJ, BiswasA, JiangQ S, ZhaoR Y, HuJ, HuB F, ShiZ. Estimating soil salinity from remote sensing and terrain data in southern Xinjiang Province, China[J]. Geoderma, 2019, 337: 1309-1319
ChenH Y, ZhaoG X, ChenJ C, WangR Y, GaoM X. Remote sensing inversion of saline soil salinity based on modified vegetation index in estuary area of Yellow River[J]. Transactions of the Chinese Society of Agricultural Engineering, 2015, 31(5):107-114 (in Chinese)
WangF, YangS T, DingJ L, WeiY, GeX Y, LiangJ. Environmental sensitive variable optimization and machine learning algorithm using in soil salt prediction at oasis[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018,34(22):102-110 (in Chinese)
[38]
EoghanK. BorutaShap: A wrapper feature selection method which combines the Boruta feature selection algorithm with Shapley values[EB/OL]. [--]
[39]
BrunsdonC, FotheringhamS, CharltonM. Geographically weighted regression[J]. Journal of the Royal Statistical Society: Series D (The Statistician), 1998, 47(3): 431-443
[40]
ChenT, GuestrinC. XGBoost: a scalable tree boosting system[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Minin. August 13-17, 2016, San Francisco, California, USA. New York: Association for Computing Machinery, 2016
LiuQ M, ChengQ M, WangX, LiX G. Soil salinity inversion in Hetao Irrigation district using microwave radar[J].Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(16): 109-114 (in Chinese)
ZhangZ X, SongY T, ZhangH Z, LIX J, NiuB B. Spatiotemporal dynamics of soil salinity in the Yellow River Delta under the impacts of hydrology and climate[J]. Chinese Journal of Applied Ecology,2021,32(4):1393-1405 (in Chinese)