Development of a Drought Monitoring System for Winter Wheat in the Huang-Huai-Hai Region, China, Utilizing a Machine Learning–Physical Process Hybrid Model DOI Creative Commons

Qianchuan Mi,

Zhiguo Huo,

Meixuan Li

et al.

Agronomy, Journal Year: 2025, Volume and Issue: 15(3), P. 696 - 696

Published: March 13, 2025

Droughts, intensified by climate change and human activities, pose a significant threat to winter wheat cultivation in the Huang-Huai-Hai (HHH) region. Soil moisture drought indices are crucial for monitoring agricultural droughts, while challenges such as data accessibility soil heterogeneous necessitate use of numerical simulations their effective regional-scale applications. The existing simulation methods like physical process models machine learning (ML) algorithms have limitations: struggle with parameter acquisition at regional scales, ML face difficulties settings due presence crops. As more advanced complex branch ML, deep even greater limitations related crop growth management. To address these challenges, this study proposed novel hybrid system that merged model. Initially, we employed Random Forest (RF) regression model integrated multi-source environmental factors estimate prior sowing wheat, achieving an average coefficient determination (R2) 0.8618, root mean square error (RMSE) 0.0182 m3 m−3, absolute (MAE) 0.0148 m−3 across eight depths. RF provided vital parameters operation Water Balance Winter Wheat (WBWW) scale, enabling assessments combined Moisture Anomaly Percentage Index (SMAPI). Subsequent comparative analyses between system-generated results actual disaster records during two events highlighted its efficacy. Finally, utilized examine spatiotemporal variations patterns HHH region over past decades. findings revealed overall intensification conditions decline SMAPI rate −0.021% per year. Concurrently, there has been shift patterns, characterized increase both frequency extremity events, duration intensity individual decreased majority Additionally, identified northeastern, western, southern areas requiring concentrated attention targeted intervention strategies. These efforts signify notable application fusion techniques integration within big context, thereby facilitating prevention, management, mitigation

Language: Английский

Development of a Drought Monitoring System for Winter Wheat in the Huang-Huai-Hai Region, China, Utilizing a Machine Learning–Physical Process Hybrid Model DOI Creative Commons

Qianchuan Mi,

Zhiguo Huo,

Meixuan Li

et al.

Agronomy, Journal Year: 2025, Volume and Issue: 15(3), P. 696 - 696

Published: March 13, 2025

Droughts, intensified by climate change and human activities, pose a significant threat to winter wheat cultivation in the Huang-Huai-Hai (HHH) region. Soil moisture drought indices are crucial for monitoring agricultural droughts, while challenges such as data accessibility soil heterogeneous necessitate use of numerical simulations their effective regional-scale applications. The existing simulation methods like physical process models machine learning (ML) algorithms have limitations: struggle with parameter acquisition at regional scales, ML face difficulties settings due presence crops. As more advanced complex branch ML, deep even greater limitations related crop growth management. To address these challenges, this study proposed novel hybrid system that merged model. Initially, we employed Random Forest (RF) regression model integrated multi-source environmental factors estimate prior sowing wheat, achieving an average coefficient determination (R2) 0.8618, root mean square error (RMSE) 0.0182 m3 m−3, absolute (MAE) 0.0148 m−3 across eight depths. RF provided vital parameters operation Water Balance Winter Wheat (WBWW) scale, enabling assessments combined Moisture Anomaly Percentage Index (SMAPI). Subsequent comparative analyses between system-generated results actual disaster records during two events highlighted its efficacy. Finally, utilized examine spatiotemporal variations patterns HHH region over past decades. findings revealed overall intensification conditions decline SMAPI rate −0.021% per year. Concurrently, there has been shift patterns, characterized increase both frequency extremity events, duration intensity individual decreased majority Additionally, identified northeastern, western, southern areas requiring concentrated attention targeted intervention strategies. These efforts signify notable application fusion techniques integration within big context, thereby facilitating prevention, management, mitigation

Language: Английский

Citations

0