Precipitation Forecasting via Multi-Scale Deconstructed ConvLSTM

TL;DR

This paper introduces a multi-scale deconstructed ConvLSTM model that enhances precipitation forecasting by efficiently capturing multi-scale spatiotemporal features, combining deep learning with traditional NWP data.

Contribution

The paper proposes a novel multi-scale deconstructed ConvLSTM that improves spatiotemporal modeling in weather prediction with low parameter consumption.

Findings

Effective precipitation forecasting on EC and CM datasets.

Improved spatiotemporal feature extraction with MSD-ConvLSTM.

Demonstrated data-driven approach enhances traditional NWP methods.

Abstract

Numerical Weather Prediction (NWP), is widely used in precipitation forecasting, based on complex equations of atmospheric motion requires supercomputers to infer the state of the atmosphere. Due to the complexity of the task and the huge computation, this methodology has the problems of inefficiency and non-economic. With the rapid development of meteorological technology, the collection of plentiful numerical meteorological data offers opportunities to develop data-driven models for NMP task. In this paper, we consider to combine NWP with deep learning. Firstly, to improve the spatiotemporal modeling of meteorological elements, a deconstruction mechanism and the multi-scale filters are composed to propose a multi-scale deconstructed ConvLSTM (MSD-ConvLSTM). The MSD-ConvLSTM captures and fuses the contextual information by multi-scale filters with low parameter consumption. Furthermore, an encoder-decoder is constructed to encode the features of multiple meteorological elements by deep CNN and decode the spatiotemporal information from different elements by the MSD-ConvLSTM. Our method demonstrates the data-driven way is significance for the weather prediction, which can be confirmed from the experimental results of precipitation forecasting on the European Centre Weather Forecasts (EC) and China Meteorological Forecasts (CM) datasets.