Changes in ENSO Patterns Revealed from MERRA-2, TRMM/TMPA, and AIRS Data

We live and work in fortunate times: Given the count starts with the launch of the first TIROS-N (Television Infrared Observation Satellite Program) in 1978, NASA's Earth Observing System have accumulated four decades of quantitative satellite observations. Associated data records from reanalysis-assimilation efforts are closing four decades as well. It is easier than ever to find climate data records from NASA data centers, and all data are freely available for download. The burden of finding any data and computer resources for storage and analysis has given way to the more creative process of understanding the data potential and making the right choice among the many available data options.

In this presentation, we pick a sliver of popular data curated by the NASA’s Global Change Data Center, and reveal evidences of El Niño Southern Oscillation (ENSO) migrating to a more Central-Pacific mode of variability in recent decades. We use data from AIRS (Atmospheric Infrared Sounder) and MERRA-2 (Modern-Era Retrospective analysis for Research and Applications) that include atmospheric temperatures in the surface layer and outgoing longwave radiances (OLR). We also use precipitation data from the TRMM (Tropical Rainfall Measuring Mission) Multi-satellite Precipitation Analysis (TMPA) and MERRA-2. All data records are decomposed into Principal Components (PC) using Singular Value Decomposition, in the global spatial domain. The PC series are regressed on the Multivariate ENSO Index (MEI), and thus allow to easily identify the modes that contain most information on (coupled with) ENSO. The corresponding Empirical Orthogonal Functions (EOF) represent the spatial patterns of ENSO and its global teleconnections.

It is immediately apparent from the EOF of MERRA-2 OLR and precipitation that ENSO patterns in the equatorial Pacific have changed substantially in the past 15 years, as compared to 1980-1994. The recent patterns from MERRA-2 are in good agreement with AIRS and TMPA observations from the past 15 years. There are noticeable changes in the teleconnections, too. The El Niños (warm phase) of the past 15 years have stronger warming and weaker precipitation impacts in the East Pacific and the US West Coast; they caused wider cooling in the tropical Atlantic, all revealed by the EOFs. These results are confirmed at 95% confidence level by reconstructing the data series from the most informative modes, starting with the ENSO modes and dismissing the seasonal ones, and regressing the reconstructed series on MEI. Multiple implications can follow from these data. If these patterns persist, El Niño is likely to be not as effective in ending water shortages in California; even reverse conditions, i.e. drier and warmer, will be more probable, contrary to the patterns from decades ago. The descending branch of the Walker cell in the north tropical Atlantic have been more intensive during El Niño, thus having better potential to offset hurricane formation and intensity during El Niño (warm phase) years.