Objective A model will be established to address the issues existing in current soil moisture prediction models, such as insufficient spatial feature mining, limitations in temporal dynamic modeling, and weak adaptability to data distribution differences in the collaborative optimization of complex spatiotemporal dynamic representation and cross-scenario sample generalization. This will provide a scientific decision-making tool for dynamic early warning of drought and flood disasters, precision allocation of irrigation resources, and improved agricultural disaster resistance. Methods To address these issues, an attention-guided spatiotemporal feature dynamic fusion network (AGSMP-Net) was proposed. The model integrated a long short term spatiotemporal prediction network with a feature-time-space attention mechanism module, enabling focused processing of time-series information and the capture of spatial distribution variations. It identified the long-term variation trends of soil moisture and optimized the utilization of information across spatiotemporal dimensions. Results Experiments validated the feasibility of the AGSMP-Net model in predicting soil moisture using meteorological factors (precipitation and soil temperature). In the soil moisture prediction task for Henan Province from 2015 to 2024, compared to ConvLSTM, AGSMP-Net model improved the accuracy (R2 ) from 0.758 to 0.806 and reduced the root mean square error (RMSE) from 0.069 to 0.057. Precipitation has a significant effect on the prediction accuracy of soil moisture model. Conclusion The proposed model dynamically allocates feature weights through the spatiotemporal attention mechanism, which can effectively capture both the abrupt responses and steady-state trends in soil moisture variations, thereby improving the accuracy of soil moisture prediction.
LiXiaoyan. Coupling, response and adaptation mechanism of soil-vegetation-hydrology in arid areas [J]. Scientia Sinica (Terrae), 2011,41(12):1721-1730.
[4]
LiQingliang, WangZiyu, ShangguanWei, et al. Improved daily SMAP satellite soil moisture prediction over China using deep learning model with transfer learning [J]. Journal of Hydrology, 2021,600:126698.
[5]
CelikM F, IsikM S, YuzugulluO, et al. Soil moisture prediction from remote sensing images coupled with climate, soil texture and topography via deep learning [J]. Remote Sensing, 2022,14(21):5584.
DongHongxia, ZhangQindi, WeiQiang. Dynamic simulation of soil moisture under typical land preparation in semi-arid loess area using HYDRUS-1D model [J]. Bulletin of Soil and Water Conservation, 2023,43(5):156-168.
WuZemian, QiuJianxiu, LiuSuxia, et al. Advances in agricultural drought monitoring based on soil moisture [J]. Progress in Geography, 2020,39(10):1758-1769.
[10]
《中国统计年鉴2024》编委会.中国统计年鉴[M].北京:中国统计出版社,2024:4-5.
[11]
National Bureau of Statistics of China. Cina Statistical yearbook 2024 [M]. Beijing: China Statistics Press, 2024:4-5.
[12]
EgeaG, Diaz-EspejoA, FernándezJ E. Soil moisture dynamics in a hedgerow olive orchard under well-watered and deficit irrigation regimes:Assessment, prediction and scenario analysis [J]. Agricultural Water Management, 2016,164:197-211.
FanTing, ZhangCunhou, ZhangDeguang, et al. Response of soil moisture of desert steppes to precipitation patterns in Damao Banner, Inner Mongolia [J]. Bulletin of Soil and Water Conservation, 2020,40(1):72-77.
[15]
MahmoodR, HubbardK G. An analysis of simulated long-term soil moisture data for three land uses under contrasting hydroclimatic conditions in the northern great Plains [J]. Journal of Hydrometeorology, 2004,5(1):160-179.
[16]
Carrasco-BenavidesM, Gonzalez ViejoC, TongsonE, et al. Water status estimation of cherry trees using infrared thermal imagery coupled with supervised machine learning modeling [J]. Computers and Electronics in Agriculture, 2022,200:107256.
WangTieying, WangYangren, ZhanGuolong, et al. Dynamic modeling and prediction of soil moisture based on real-time water content data [J]. Chinese Journal of Agrometeorology, 2021,42(1):13-23.
[19]
JacksonS H. Comparison of calculated and measured volumetric water content at four field sites [J]. Agricultural Water Management, 2003,58(3):209-222.
ShuSufang, QianHuafeng, QiuXiaowei. Soil moisture forecast model based on meteorological factors in Jinhua City [J]. Chinese Journal of Agrometeorology, 2009,30(2):180-184.
LiuDi, SunJiaqian, YuZhongbo. Multi-layer soil moisture inversion based on machine learning models [J]. Journal of Hohai University (Natural Sciences), 2024,52(3):7-14.
WangJingping, WuXiaodan, MaDujuan, et al. Remote sensing retrieval based on machine learning algorithm:Uncertainty analysis [J]. National Remote Sensing Bulletin, 2023,27(3):790-801.
LiMingwei. Estimating soil erosion under different soil and water conservation engineering measures using LSTM model:A case study in northwest Liaoning Province [J]. Bulletin of Soil and Water Conservation, 2023,43(4):162-169.
[28]
XuLei, LvYu, MoradkhaniH. Daily multistep soil moisture forecasting by combining linear and nonlinear causality and attention-based encoder-decoder model [J]. Stochastic Environmental Research and Risk Assessment, 2024,38(12):4979-5000.
[29]
LiuKexin, LuQi, GuoShilong, et al. CNN based approach for root zone soil moisture prediction by ground penetrating radar [J]. Journal of Physics: Conference Series, 2024,2887(1):012007.
[30]
LiXiaoning, ZhangZiyin, LiQingliang, et al. Enhancing soil moisture forecasting accuracy with REDF-LSTM:Integrating residual en-decoding and feature attention mechanisms [J]. Water, 2024,16(10):1376.
[31]
ParkS H, LeeB Y, KimM J, et al. Development of a soil moisture prediction model based on recurrent neural network long short-term memory (RNN-LSTM) in soybean cultivation [J]. Sensors, 2023,23(4):1976.
[32]
JiangShuwen, ChenGuifen, ChenDong, et al. Application and evaluation of an improved LSTM model in the soil moisture prediction of southeast Chinese tobacco-producing areas [J]. Journal of the Indian Society of Remote Sensing, 2023,51(9):1843-1853.
XuJinhong, LiQingquan, WeiChuntao, et al. Landslide susceptibility evaluation based on sample optimization and deep feature extraction [J]. Bulletin of Soil and Water Conservation, 2025,45(2):190-200.
CheYinchao, ZhengGuang, XiongShuping, et al. Construction of a multi-factor soil moisture prediction model DA-LSTM-soil [J]. Journal of Henan Agricultural University, 2025,59(4):698-710.
[37]
KaraA, PekelE, OzcetinE, et al. Genetic algorithm optimized a deep learning method with attention mechanism for soil moisture prediction [J]. Neural Computing and Applications, 2024,36(4):1761-1772.
YangXiaojie, QinMingzhou, ZhangPengyan, et al. Spatio-temporal analysis of coupling characteristics of cultivated land pressure and economic development in Henan Province [J]. Areal Research and Development, 2020,39(3):144-149.
LiuYing, YueHui, LiYao, et al. Remote sensing monitoring of spring drought based on MODIS in Henan Province [J]. Agricultural Research in the Arid Areas, 2018,36(3):218-223.
[42]
魏克循, 马国明. 河南省土壤普查办公室编著.河南土壤[M].北京:中国农业出版社,2004.
[43]
WeiKexun, MaGuoming. Compiled by the Soil Census Office of Henan Province. Soil in Henan [M]. Beijing: China Agriculture Press, 2004.
[44]
Copernicus Climate Data Store. ERA5-Land Hourly Data (1950-present) [EB/OL]. (2025-4-13)[2025-05-08].