A Coupled Least Absolute Shrinkage and Selection Operator–Backpropagation Model for Estimating Evapotranspiration in Xizang Plateau Irrigation Districts with Reduced Meteorological Variables DOI Creative Commons
Qiang Meng, Jingxia Liu,

Fengrui Li

et al.

Agriculture, Journal Year: 2025, Volume and Issue: 15(5), P. 544 - 544

Published: March 3, 2025

This study addresses the challenge of estimating reference crop evapotranspiration (ETO) in Xizang Plateau irrigation districts with limited meteorological data by proposing a coupled LASSO-BP model that integrates LASSO regression BP neural network. The was applied to three districts: Moda (MD), Jiangbei (JB), and Manla (ML). Using ETO values calculated FAO-56 Penman–Monteith (FAO-56PM) as benchmark, performance applicability were assessed. Short-term predictions for also conducted using mean-generating function optimal subset algorithm. results revealed significant multicollinearity among six factors (maximum temperature, minimum average relative humidity, sunshine duration, wind speed), identified through tolerance, variance inflation factor (VIF), eigenvalue analysis. effectively captured interannual variation ETO, accurately identifying peaks troughs, trends closely aligned FAO-56PM model. demonstrated strong across all districts, evaluation metrics showing MAE, RMSE, NSE, R2 ranging from 4.26 9.48 mm·a−1, 5.91 11.78 0.92 0.96, 0.82 0.94, respectively. Prediction indicated statistically insignificant declining trend annual over period. Overall, is reliable accurate tool data.

Language: Английский

A Coupled Least Absolute Shrinkage and Selection Operator–Backpropagation Model for Estimating Evapotranspiration in Xizang Plateau Irrigation Districts with Reduced Meteorological Variables DOI Creative Commons
Qiang Meng, Jingxia Liu,

Fengrui Li

et al.

Agriculture, Journal Year: 2025, Volume and Issue: 15(5), P. 544 - 544

Published: March 3, 2025

This study addresses the challenge of estimating reference crop evapotranspiration (ETO) in Xizang Plateau irrigation districts with limited meteorological data by proposing a coupled LASSO-BP model that integrates LASSO regression BP neural network. The was applied to three districts: Moda (MD), Jiangbei (JB), and Manla (ML). Using ETO values calculated FAO-56 Penman–Monteith (FAO-56PM) as benchmark, performance applicability were assessed. Short-term predictions for also conducted using mean-generating function optimal subset algorithm. results revealed significant multicollinearity among six factors (maximum temperature, minimum average relative humidity, sunshine duration, wind speed), identified through tolerance, variance inflation factor (VIF), eigenvalue analysis. effectively captured interannual variation ETO, accurately identifying peaks troughs, trends closely aligned FAO-56PM model. demonstrated strong across all districts, evaluation metrics showing MAE, RMSE, NSE, R2 ranging from 4.26 9.48 mm·a−1, 5.91 11.78 0.92 0.96, 0.82 0.94, respectively. Prediction indicated statistically insignificant declining trend annual over period. Overall, is reliable accurate tool data.

Language: Английский

Citations

0