Assessing Tropical Pacific-induced Predictability of Southern California Precipitation Using Machine Learning
The Science
Traditional El Niño Southern Oscillation (ENSO) indices, like the Niño 3.4 index, although well-predicted themselves, fail to offer reliable sub-seasonal to seasonal predictions of Western US precipitation. In this study, researchers use a novel Multi-Input Multi-Output Autoencoder-decoder (MIMO-AE) machine learning network (Fig. 1) to capture the non-linear relationship between Tropical Pacific Ocean sea surface temperatures (SST) and Southern California precipitation, which is projected into an index – called the MIMO-AE index. MIMO-AE network was trained on a historical simulation with the DOE’s Energy Exascale Earth System Model (E3SMv1) and observational data and its skill of predicting sub-seasonal to seasonal Southern California precipitation was evaluated.
Figure 1. The image shows the novel Multi-Input Multi-Output AutoEncoder (MIMO-AE) network architecture that captures the non-linear relationship between Southern California precipitation and Tropical Pacific sea surface temperatures. The network provides enhanced sub-seasonal to seasonal predictability of Southern California precipitation.
The Impact
MIMO-AE provides significantly enhanced predictability of Southern California precipitation for a lead-time of up to four months as compared to ENSO indices, like Niño 3.4 index and ENSO Longitudinal Index. MIMO-AE learned SST anomaly patterns associated with Southern California precipitation strongly influence processes that drive precipitation over the region, allowing MIMO-AE to provide enhanced predictive skill. The study demonstrates that machine learning approaches could significantly improve the predictability of regional precipitation on sub-seasonal to seasonal time scales.
Figure 2: Predictive skill of MIMO-AE. Correlation skill of MIMO-AE index (blue line) at predicting Southern California precipitation as compared to other indices like Niño3.4 index (yellow line) and ELI (green line) for observational data. Note the enhanced skill of MIMO-AE.
Summary
A novel Multi-Input Multi-Output Autoencoder-decoder (MIMO-AE) machine learning network is constructed to capture the non-linear relationship of Southern California precipitation and Tropical Pacific Ocean SSTs. The MIMO-AE is trained on both monthly Tropical Pacific SST anomalies and Southern California precipitation anomalies simultaneously. The co-variability of the two fields in the MIMO-AE shared nonlinear latent space can be condensed into an index, termed the MIMO-AE index. A transfer learning approach was used to train a MIMO-AE on the combined dataset of 100 years of output from a historical simulation with DOE’s Energy Exascale Earth Systems Model (E3SMv1) and a segment of observational data. Long Short-Term Memory networks were used to assess sub-seasonal to seasonal predictability of Southern California precipitation from the MIMO-AE index. MIMO-AE provides enhanced predictability of Southern California precipitation for a lead-time of up to four months as compared to Niño 3.4 index and the El Niño Southern Oscillation Longitudinal Index (Fig. 2). This is likely because MIMO-AE learned SST anomaly patterns associated with Southern California precipitation strongly influence processes, like moisture fluxes into the region, which drive precipitation there – lending predictive skill to MIMO-AE.
Publication
- Passarella, Linsey S, and Salil Mahajan. 2023. “Assessing Tropical Pacific-Induced Predictability of Southern California Precipitation Using a Novel Multi-Input Multi-Output Autoencoder”. Artificial Intelligence for the Earth Systems. American Meteorological Society, 1-30. doi:10.1175/aies-d-23-0003.1.
Funding
- This research was supported by the Energy Exascale Earth System Model (E3SM) Project of the Earth System Model Development program area of the Department of Energy, Office of Science, Biological and Environmental Research program.
Contact
- Salil Mahajan, Oak Ridge National Laboratory
This article is a part of the E3SM “Floating Points” Newsletter, to read the full Newsletter check:
