XFMNet: Decoding Cross-Site and Nonstationary Water Patterns via Stepwise Multimodal Fusion for Long-Term Water Quality Forecasting
This addresses water quality prediction for environmental monitoring, particularly in multi-site river networks, but appears incremental as it builds on existing multimodal and decomposition techniques.
The paper tackled long-term water quality forecasting by developing XFMNet, a stepwise multimodal fusion network that integrates remote sensing precipitation imagery to model temporal and spatial dynamics, achieving substantial improvements over state-of-the-art baselines in experiments on real-world datasets.
Long-term time-series forecasting is critical for environmental monitoring, yet water quality prediction remains challenging due to complex periodicity, nonstationarity, and abrupt fluctuations induced by ecological factors. These challenges are further amplified in multi-site scenarios that require simultaneous modeling of temporal and spatial dynamics. To tackle this, we introduce XFMNet, a stepwise multimodal fusion network that integrates remote sensing precipitation imagery to provide spatial and environmental context in river networks. XFMNet first aligns temporal resolutions between water quality series and remote sensing inputs via adaptive downsampling, followed by locally adaptive decomposition to disentangle trend and cycle components. A cross-attention gated fusion module dynamically integrates temporal patterns with spatial and ecological cues, enhancing robustness to nonstationarity and site-specific anomalies. Through progressive and recursive fusion, XFMNet captures both long-term trends and short-term fluctuations. Extensive experiments on real-world datasets demonstrate substantial improvements over state-of-the-art baselines, highlighting the effectiveness of XFMNet for spatially distributed time series prediction.