Exploring low-resource weather forecasting with echo state network-based architectures and satellite data

Cloud forecasting plays a crucial role in various fields such as agriculture, energy systems, and air travel. An accurate forecasting system can offer significant benefits by improving decision-making efficiency in these areas. This study investigates the use of Echo State Network (ESN)-based architectures for weather forecasting, focusing on cloud prediction across Central Europe using the CloudCast benchmark, which integrates data from Meteosat satellites and the European Centre for Medium-Range Weather Forecasts (ECMWF) model. Two novel techniques are included in this study, evaluated in two different phases. First, the Multi-Reservoir Weighted ESN (MWESN) architecture is proposed, featuring optimized inter-reservoir connections that enhance both the effectiveness and adaptability of the model. This model is evaluated along with advanced ESN architectures, including Multi-Reservoir ESN, Deep ESN among others. Second, the Error-Guided Regional Training (ERT) method is introduced to minimize the computational resources required for forecasting at the pixel level while maintaining high accuracy. Combined, MWESN and ERT demonstrate a 1.41% improvement in accuracy, effectively capturing complex spatio-temporal dynamics while significantly reducing computational demands compared to existing state-of-the-art methods. Additionally, models are tested on low-resource devices such as Raspberry Pi units, illustrating their feasibility for real-world meteorological applications. • Comprehensive Evaluation of ESN-Based Architectures: The study evaluates multiple Echo State Network architectures, including Multi-Reservoir ESNs (MRESNs) and a newly developed Multi-Reservoir Weighted ESN (MWESN). • Superior Performance of MWESN: The MWESN architecture, with optimised connections between reservoirs, outperforms other models in terms of accuracy. It outperforms the state-of-the-art model by 1.41%. • Novel pixel-level video forecasting technique specially designed for resource-limited devices called Error-Guided Regional Training. • Low-Resource Training Potential: Highlights the feasibility of training these models on low-resource devices like Raspberry Pi units, making them accessible for broader applications. • Real-World Application: Highlights the applicability of MRESNs in meteorology and related fields.