🤖 AI Summary
This study addresses the need for accurate multivariate forecasting of key meteorological variables—reference evapotranspiration, vapor pressure deficit, wind speed, and wind direction—for agricultural irrigation and farming decisions. Leveraging ERA5 reanalysis data, the authors systematically evaluate GRU, LSTM, and their hybrid architectures combined with 1D-CNN for both short-term (24-hour) and medium-term (168-hour) predictions. Results demonstrate that hybrid CNN-recurrent models consistently outperform standalone recurrent networks: CNN-GRU achieves weighted quantile scores (WQS) of 0.8275 and 0.7829 at the two forecast horizons, representing improvements of 0.44%–1.63% over pure recurrent models. Moreover, CNN-LSTM attains comparable accuracy with fewer parameters, confirming the efficacy and computational efficiency of hybrid architectures in agrometeorological forecasting.
📝 Abstract
Accurate meteorological forecasting is essential for agricultural planning, irrigation management, and environmental decision support. This study conducts a comparative evaluation of recurrent and hybrid deep learning architectures for multivariate forecasting of reference evapotranspiration ($ET_0$), vapour pressure deficit (VPD), wind speed, and the sine and cosine components of wind direction. The analysis utilizes 134,376 hourly observations from Ioannina, Greece, spanning January 2011 to April 2026, sourced from ERA5 via the OpenMeteo Historical Weather API. Single and multi-layer GRU and LSTM networks are compared with hybrid 1D-CNN-GRU and 1D-CNN-LSTM models for two forecasting tasks: a 24-hour next-day forecast and a 168-hour week-ahead forecast. Performance is evaluated using normalized root mean squared error, the coefficient of determination, and a composite Weighted Quotient Score (WQS). The most effective purely recurrent models are a 64-unit LSTM for the 24-hour horizon, with a WQS of 0.816755, and a 1024-unit GRU for the 168-hour horizon, with a WQS of 0.779465. The hybrid CNN-GRU models achieved the highest overall scores of 0.827535 and 0.782863 for the 24-hour and 168-hour horizons, but with additionally more number of units respectively to LSTM models, while the CNN-LSTM models yield nearly identical results with substantially fewer parameters. Compared to the corresponding recurrent baselines, the hybrid models improve WQS by 1.22--1.63\% at 24 hours and by 0.44--0.45\% at 168 hours, indicating that convolutional feature extraction is more beneficial for short-term forecasting.