From the shelves to the deep: satellite-derived profiles for operational ocean heat content applications in the North Atlantic and Pacific Ocean

PI: David Trossman, University of Maryland, College Park
Start Year: 2025 | Duration: 4  years
Partners: NOAA, NASA

Project Abstract: 

The proposers will develop the Next Generation Enterprise Ocean Heat Content (NGE OHC) algorithm to produce high-resolution (~10 km along-track and ¼° gridded) temperature and salinity (T/S) upper-ocean profiles operationally in the continental shelf and open ocean regions of the North Atlantic and Pacific Oceans using satellite data and machine learning techniques. These profiles will enable us to monitor and investigate the processes associated with extreme events, among other applications. The NGE OHC algorithm is rooted in the well-established gravest/geostrophic empirical mode (GEM) approach, which utilizes the widespread nature and dominance of lowmode baroclinic variability in the ocean to estimate subsurface profiles from surface information. Corrections to the GEM to account for temporal variability in boundary (i.e., sea surface) conditions are also derived from remotely sensed surface data, including sea surface temperatures and salinities, as well as additional parameters to address seasonal and latitudinal variability. The NGE OHC algorithm estimates T/S profiles at 2 dbar resolution from the surface to an appropriate maximum depth (typically ~1000 dbar in the open ocean and ~500 dbar in continental shelf zones) and related parameters of interest such as OHC and mixed layer depth.    In the open North Atlantic Ocean, a preliminary version of the algorithm is currently applied. This version of the NGE OHC algorithm can reproduce more than 80% of the variance in quality-controlled Argo temperature profile data over 0–600 dbar and more than 90% of the variance over 0–50 dbar. The algorithm reproduces more than 70% of the variance in quality-controlled Argo salinity profile data in the upper 100–600 dbar and more than 60% of the variance over 0–100 dbar, which we plan to improve upon in our proposed work with developments described below. Further assessment of the NGE OHC algorithm will be performed in this work using independent in-situ data, including those from deep gliders collected along the altimeter tracks. The team will both expand the domain of the current algorithm and enhance its fidelity with several novel developments. The existing open ocean North Atlantic NGE OHC algorithm is applied separately over areas with different stratification regimes, which are identified by applying a machine learning clustering method to historical temperature profile data. When adapting the algorithm to the continental shelves, we will need to estimate new clusters that capture the variability of this dynamic region. We will need to identify new clusters for the open Pacific Ocean and surrounding continental shelves applications of the algorithm as well. Further optimization of the stratification-based clusters (e.g., by also considering salinity) will also be explored. We plan to improve the algorithm itself in several ways, including: making separate GEM estimates within/below the mixed layer, using atmospheric forcing data for additional corrections to the GEM, and parameterizing the factors that impact surface S anomalies on time scales not resolved by satellite data. Upon finalizing the NGE OHC algorithm for each domain, we will quantify the uncertainties in our T/S profile estimates. Together with existing products (e.g., Argo-based velocity estimates), the profiles and associated uncertainties from the NGE OHC algorithm can then be used to analyze heat budgets that are critical to understanding the processes that control the life cycles of marine heat waves.  The team will then apply the NGE OHC algorithm to assess the benefits of using its outputs for tropical cyclone and marine heatwave forecast evaluation in the North Atlantic and Pacific Oceans using observing system experiments. The NGE OHC output can also be used to monitor the subsurface characteristics of these extreme events, which can have major impacts on fisheries and coral reefs.