Work

Short-term soil moisture content forecasting with a hybrid informer model

Year: 2025

Type: article

Abstract: This study proposes a novel time-series forecasting approach that integrates the Informer model with the RAO − 1 optimization algorithm for soil water content (SWC) prediction. The method innovatively combines Informer’s long-range dependency modeling with RAO-1’s efficient hyperparameter optimization to enhance forecasting accuracy. Comparative experiments were conducted using Random Forest, Support Vector Regression, Long Short-Term Memory and Transformer as baseline models on SWC datasets from the Beijing region. The RAO-1-optimized Informer consistently outperforms these baselines in both deterministic and probabilistic forecasting tasks, while also achieving superior computational efficiency. These results highlight the robustness of the proposed method and its potential to support sustainable agricultural water management through accurate SWC prediction.

Source: Frontiers in Sustainable Food Systems

DOI: 10.3389/fsufs.2025.1636499

Authors: Long Wang, Shihan Yao, Chao Huang

Institutions: City University of Hong Kong, University of Science and Technology Beijing

Language: English

Topic: Hydrological Forecasting Using AI

Subfield: Environmental Engineering

Field: Environmental Science

Domain: Physical Sciences

SDGs: Zero hunger, Climate action, Life on land

Open Access status: gold

Estimated APC paid: $2,353