Probabilistic modeling of lake surface water temperature using a Bayesian spatio-temporal graph convolutional neural network

TL;DR

This paper introduces a Bayesian spatio-temporal graph neural network for probabilistic lake surface temperature estimation, offering accurate, computationally efficient predictions across the entire lake surface with sparse data.

Contribution

It presents a novel combination of Bayesian recurrent and graph convolutional neural networks for lake temperature modeling, improving performance and efficiency over existing methods.

Findings

Homogeneous performance across the lake surface.

Effective with sparse training data.

Outperforms state-of-the-art Bayesian deep learning methods.

Abstract

Accurate lake temperature estimation is essential for numerous problems tackled in both hydrological and ecological domains. Nowadays physical models are developed to estimate lake dynamics; however, computations needed for accurate estimation of lake surface temperature can get prohibitively expensive. We propose to aggregate simulations of lake temperature at a certain depth together with a range of meteorological features to probabilistically estimate lake surface temperature. Accordingly, we introduce a spatio-temporal neural network that combines Bayesian recurrent neural networks and Bayesian graph convolutional neural networks. This work demonstrates that the proposed graphical model can deliver homogeneously good performance covering the whole lake surface despite having sparse training data available. Quantitative results are compared with a state-of-the-art Bayesian deep learning method. Code for the developed architectural layers, as well as demo scripts, are available on https://renkulab.io/projects/das/bstnn.