Long-Range Seasonal Forecasting of 2m-Temperature with Machine Learning

TL;DR

This paper explores the use of CNN and RNN neural networks for long-range seasonal 2m-temperature forecasting at diverse locations, showing improved accuracy up to 30-52 weeks in some regions, but with limitations in tropical areas.

Contribution

It demonstrates the potential of machine learning models, specifically CNN and RNN, to enhance long-range temperature forecasts compared to traditional methods.

Findings

Improved forecast accuracy up to 30 weeks for PCC.

Enhanced RMSESS up to 52 weeks in certain locations.

Limitations observed in tropical regions where climatology has low correlation.

Abstract

A significant challenge in seasonal climate prediction is whether a prediction can beat climatology. We hereby present results from two data-driven models - a convolutional (CNN) and a recurrent (RNN) neural network - that predict 2 m temperature out to 52 weeks for six geographically-diverse locations. The motivation for testing the two classes of ML models is to allow the CNN to leverage information related to teleconnections and the RNN to leverage long-term historical temporal signals. The ML models boast improved accuracy of long-range temperature forecasts up to a lead time of 30 weeks for PCC and up 52 weeks for RMSESS, however only for select locations. Further iteration is required to ensure the ML models have value beyond regions where the climatology has a noticeably reduced correlation skill, namely the tropics.