Ben Johnson
Perhaps the most fascinating aspect of geophysical fluid dynamics is the emergence of phenomena at different scales. As atmospheric models increase in resolution, they begin resolving Nor'easters, hurricanes, squall lines, tornadoes and clouds, respectively. Oceanographic models exhibit similar behavior, but require 100x higher resolution than atmospheric models in order for the analogous phenomena to emerge in the ocean. Only recently have global oceanographic models been able to resolve their analog to Nor'easters in the atmosphere: the ubiquitous occurrence of mesoscale eddies in the ocean. During the Yellowstone ASD period, CESM's Parallel Ocean Program (POP) was used to complete a pair of experiments to produce a global high-resolution simulation that allowed oceanographers to study the effects of these mesoscale eddies in the ocean. With Derecho's computing power, it is feasible to integrate multi-member ensembles of these ocean configurations and employ data assimilation methods to create what is known as a reanalysis, or a historical dataset that synthesizes real ocean observations into the model simulation. This ASD project uses off-the-shelf configurations of POP and the Data Assimilation Research Testbed (DART) which, when combined with Derecho's computing power, will create an unprecedented dataset: a hierarchical reanalysis using both low-resolution, eddy-parameterizing and high-resolution, eddy-resolving ocean model ensembles to understand the influence of emergent phenomena when predicting changes in Earth's climate. In particular, this reanalysis aims to study the Northern Pacific Basin as it transitions from its cool to warm phase to improve our understanding of the Pacific Decadal Oscillation, a major mode of climate variability.
Ben Johnson is a Project Scientist I in CISL's Data Assimilation Research Section. He has a BS in Physics and a PhD in Oceanography from the University of Maryland.