Statistical downscaling of precipitation using long short-term memory recurrent neural networks

• Published in: Theoretical and applied climatology Vol. 134; no. 3-4; pp. 1179 - 1196
• Main Authors: Misra, Saptarshi, Sarkar, Sudeshna, Mitra, Pabitra
• Format: Journal Article
• Language: English
• Published: Vienna Springer Vienna 01.11.2018; Springer; Springer Nature B.V
• Subjects: Analysis; Aquatic Pollution; Artificial neural networks; Atmospheric precipitations; Atmospheric Protection/Air Quality Control/Air Pollution; Atmospheric Sciences; Climate change; Climate science; Climatology; Computer simulation; Datasets; Earth and Environmental Science; Earth Sciences; General circulation models; Global climate; Global temperature changes; Hydrologic models; Hydrology; Hydrometeorology; Long short-term memory; Mathematical models; Methods; Modelling; Natural language processing; Neural networks; Original Paper; Precipitation; Rain; Rainfall; Recurrent neural networks; Regional analysis; Regression; River basins; Rivers; Short term; Statistical analysis; Statistical models; Waste Water Technology; Water Management; Water Pollution Control
• ISSN: 0177-798X; 1434-4483
• DOI: 10.1007/s00704-017-2307-2

Hydrological impacts of global climate change on regional scale are generally assessed by downscaling large-scale climatic variables, simulated by General Circulation Models (GCMs), to regional, small-scale hydrometeorological variables like precipitation, temperature, etc. In this study, we propose a new statistical downscaling model based on Recurrent Neural Network with Long Short-Term Memory which captures the spatio-temporal dependencies in local rainfall. The previous studies have used several other methods such as linear regression, quantile regression, kernel regression, beta regression, and artificial neural networks. Deep neural networks and recurrent neural networks have been shown to be highly promising in modeling complex and highly non-linear relationships between input and output variables in different domains and hence we investigated their performance in the task of statistical downscaling. We have tested this model on two datasets—one on precipitation in Mahanadi basin in India and the second on precipitation in Campbell River basin in Canada. Our autoencoder coupled long short-term memory recurrent neural network model performs the best compared to other existing methods on both the datasets with respect to temporal cross-correlation, mean squared error, and capturing the extremes.