Overview

 In January 2016, Dr. Lynn K. (Nick) Shay at the University of Miami, Rosenstiel School of Marine and Atmospheric Science was awarded an RFP-V grant totaling $2,828,784 to lead the GoMRI project entitled Three-Dimensional Gulf Circulation and Biogeochemical Processes Unveiled by State-of-the-Art Profiling Float Technology and Data Assimilative Ocean Models, which consisted of 3 collaborative institutions and approximately 20 research team members (including students).   The overarching goal of this proposed research was to build a rapid response capability that can be deployed in the event of an oil spill. The capability consists of an integrated observation-prediction system to map the distribution and extent of hydrocarbons in the water column in real time and to quantify hydrocarbon removal and fate including short-term predictions of dispersion induced by the current field and transport of oil to the sea floor through scavenging by marine particles. Specific research objectives were (1) Observe fundamental physical and biogeochemical properties and processes using advanced state-of-the-art measurement sensors on new profiling floats; (2) Integrate physical and biogeochemical processes in a coupled model that assimilates real-time data streams in the presence of strong currents; (3) Develop a flexible and carefully evaluated “end-to- end” predictive capability that can be deployed rapidly in case of subsurface oil spills to improve mitigation approaches by emergency responders and policy makers; and, (4) Quantify data and model uncertainties via a robust suite of realistic scenario simulations so that the final forecasted probability has well-understood sources of uncertainty. The prediction system was then evaluated in retrospective assimilation experiments using data from the Deepwater Horizon spill and in forecast experiments that assimilate satellite and float data in real time. Both demonstrated the system’s capability and improved our understanding of physical mechanisms and their impacts on the biogeochemistry in the water column.  

Research Highlights:

 

As of December 31, 2019, this project’s research resulted in 8 peer-reviewed publications, 29 scientific presentations and 12 datasets being submitted to the GoMRI Information and Data Cooperative (GRIIDC), which are/will be made available to the public. The project also engaged 6 Masters and PhD students over its award period. Significant outcomes of this project’s research are highlighted below.

 

APEX-EM Float Performance:

 

The APEX-EM floats, developed at Teledyne Webb Research (TWR) and deployed in May 2017, continued to sample ocean conditions in the Gulf and along the eastern seaboard until September 2018. This is one of the longest deployment of APEX-EM floats encompassing 16 months in September 2018 (Table 1). Analyses of the records have provided valuable insights into how to improve float performance subject to a spectrum of stressors. Given the dispersion of the floats, it was impossible to retrieve these floats as a few of them made it through the Florida Straits and into the North Atlantic Ocean basin. Two floats died prematurely (within two months of deployment). One important engineering finding was the APF-9 controller boards have to be replaced by APF-11 boards at TWR. These upgraded

boards provide the float with more control that will feedback into an improved EM subsystems (Zsuts, 2018, personal communication). The APEX-EM platform represents an important tool for the ocean research community to measure deep ocean currents and the addition of the biochemical sensors expands the capabilities of the float. We are in the process of proposing such instrumentation for the NASEM Loop Current field program.

 

Data from the floats were quality controlled and uploaded to GRIIDC (Reference Number: R5x281.0000.0001). A paper was published in the Proceedings of the IEEE sponsored Current, Wave, and Turbulence Measurement Workshop on Physical and Biochemical Structures Measured by APEX-EM Floats. The paper was submitted to GRIIDC. This IEEE Proceedings paper is being expanded for eventual submission into the Journal of Ocean Engineering given the amount of technical and engineering data acquired during float deployments.

 

	7939	7940	7941	7942	7943	7944	7945	8081	8082	8083	Total
CTD	303	83	446	482	384	70	222	345	101	312	2748
Vel	303	83	291	482	384	70	141	341	101	312	2508
DO	303	83	440	482	384	70	225	345	101	312	2745
Chlor	240	83	338	482	384	70	211	345	101	312	2566
Total	1149	332	1515	1928	1536	280	799	1376	404	1248	10567

 

Table 1: Summary of profile measurements from the ten APEX-EM floats deployed in the Gulf of Mexico in May 2017 through Sept 2018 where Vel is velocity, DO is dissolved oxygen and Chlor Is Chlorophyll (Backscatter and CDOM has an equal number of profiles).

 

We are continuing to work on underway data acquired during the cruise measurements and fluxes from the air-sea flux measuring system (CARTHE) and the ship’s sensors during the passage of of a cold front. These data were augmented with NOAA Data Buoy records in the Northern Gulf where surface pressure dipped to 1005 mb. These data are being used in a manuscript that discuss the cold front passage response measured by APEX-EM Floats that was presented at Teledyne Marine Workshop in Oct 2019 (invited). A key finding is that wind-driven shear induced- mixing event that deepened the oceanic mixed layer by 20 m. Underway quality-controlled data from the RV Smith (1-10 May) have been uploaded to GRIIDC (Ref Number R5.x 275.000:0004, 0005, 0006, 0007). These underway data were used in our initial evaluation of the floats for the first ten days (GRIIDC Reference Number R5x275.0000.0002/0003).

 

Hurricane Isaac and Nate:

 

Air-sea interactions research during hurricane Isaac (2012) was completed over the first year of this grant. These data were acquired as part of a Deep-C grant to the University of Miami (from FSU) resulting in the publication Observed Air-Sea Interactions in Tropical Cyclone Isaac Over Loop Current Mesoscale Eddy Features by Jaimes, Shay and Brewster in the Dynamics of Atmospheres and Oceans Special Issue of the Loop Current Dynamics Experiment. These data have been provided to GRIIDC (Reference Number: R1.x138.077:0015, R1.x138.077:0016, and R1.x138.077:0017). The APEX-EM Floats also measured a strong near-inertial response to hurricane Nate. A manuscript is in preparation for submission to a journal assessing the effects of the physical oceanographic response on the biochemical processes in collaboration with colleagues from Dalhousie University.

 

 

 

Impact of Loop Current Frontal Eddies on Loop Current Eddy Shedding:

 

During the 2010 Deepwater Horizon oil spill, an underestimated part of the oil was entrained into an intensified Loop Current Frontal Eddy (LCFE). Hiron et al. (2020) showed that these eddies, which are also known to play an essential role in the Loop Current Eddy (LCE) shedding, are difficult to predict, and the dynamics involving their intensification are still not fully understood. The Loop Current (LC) and its stronger LCFEs were continuously tracked during 2009–2011 using sea surface height (SSH) from AVISO+. The BOEM mooring array (Hamilton et al. 2016) and the oceanographic expendables from NOAA WP-3D flights (Shay et al. 2011) provided complementary information about the internal structure of this LC-LCFE interaction. The intensification of the tracked-LCFEs presented similar characteristics, independent of their location: a steep increase in kinetic energy, a corresponding decrease in SSH, and an increase in its area. As the LCFE grows, the flow at the interface with the LC becomes stronger and deeper, and the horizontal density gradient between the features increases. The intensification of the front and the LCFEs is driven by the advection (nonlinear) term, and the gradient pressure (linear) term in the momentum budget. Evidence of an inverse energy cascade suggests that LCFEs are extracting energy and mass from the submesoscale field to the zone of contact between the LC and the LCFE, strengthening the front, and allowing the LCFEs to grow during periods of intensification. Understanding the physics driving the LCFE intensification is a key step to improve LC forecast models, and to improve prediction of shedding events, as well as oil and particle transport around the LC. Float data are being used in the Hiron PhD dissertation research.

Oxygen:

Oxygen measurements from the 10 APEX floats deployed in May 2017 near the Deepwater Horizon (DwH) spill site were calibrated and quality controlled. A novel method for correcting measurement errors resulting from the relatively slow response time of the optode oxygen sensor was developed (Gordon 2019, Gordon et al. 2020). This method relies on an inverse determination of the effective in-situ response time of the oxygen sensor followed by an inverted filter and is broadly applicable to optode measurements from floats and gliders. Float data from continuous profiling periods (after deployment and during passage of Hurricanes Irma and Nate) were analyzed in detail. This analysis has shown that continuous profiling allows resolution of diurnal changes in oxygen inventory in the euphotic zone, which implies that continuously profiling floats can be used to measure rates of community production and respiration. The analysis also shows evidence of physically driven changes in dissolved oxygen due to near-inertial oscillations shifted off the local Coriolis frequency by 5 to 10%.

 

Data Assimilation and Modeling:

A data-assimilative physical-hydrocarbon model based on the Regional Ocean Modeling System (ROMS) for the Gulf of Mexico has been implemented and applied to the period of the 2010 DwH spill. Data from satellite (altimetry and temperature) and profilers (Shay et al. 2011) have been assimilated and shown to improve the model’s skill in the mesoscale (particularly the location and intensity of the Loop Current and Loop Current Eddies). In twin experiments it was shown that deep currents near the spill site are also improved by the assimilation system. A two-way nested version of the model has been implemented that represents the DwH spill site with a resolution of

~1 km. Two manuscripts describing the assimilation system as well as results from twin experiments and assimilation of real data have been published (Yu et al. 2018, 2019). The model