1 AESOP: Assessing the Effects of Submesoscale Ocean Parameterizations Scott Harper Code 322, Physical Oceanography Office of Naval Research [email protected].

1

AESOP: Assessing the Effects of Submesoscale Ocean Parameterizations

Scott HarperCode 322, Physical Oceanography

Office of Naval [email protected]

2

Naval Requirements for Ocean Information

The US Navy requires information about the ‘battlespace environment’ (BSE) on a variety of space and time scales, and would like predictions of the synoptic state of the BSE with up to a one week lead time.

While any particular application might require knowledge of only certain ocean characteristics on limited spatial scales, when one considers the breadth of Naval operations, the overall requirement can seem like “everything, anywhere.”

3

A few examples of BSE impact on the Navy

ASW (Anti-Submarine Warfare)– location of density fronts, depths of mixed layer and thermocline

for acoustic performance predictions

NSW (Naval Special Warfare)– temperature and currents for swimmer operations, optical

properties for diver visibility

MIW (Mine Warfare)– optical properties for mine hunting, drift predictions for floating

mines, currents and density conditions for AUV operations

Fleet Operations– Wave conditions for surface operations, currents for search and

rescue and ship routing

4

BSE Programs in Code 32 at ONR

One goal of Code 32 is to fund the research necessary to address the future BSE knowledge requirements of the Navy, through a combination of 6.1, 6.2, and 6.3 funding dollars.

Code 32Ocean, Atmosphere and Space

Code 321Sensing and Systems

Code 322Processes and Prediction

Physical Oceanography

Marine Meteorology

Optics and Biology

NOPP

Ocean Acoustics

Coastal Geosciences

5

Restructuring of Code 322 in 2003

In 2003, the Processes and Prediction Division was re-organized to reduce the number of overall programs, formally eliminating ‘Remote Sensing’ and ‘High Latitude Dynamics’.

‘Ocean Optics’ and ‘Biology and Chemistry’ were combined to form the Optics and Biology program.

Marine Meteorology incorporated the atmospheric elements of the ‘Remote Sensing’ program

‘Physical Oceanography’, ‘Ocean Modeling and Prediction’, ‘High Latitude Dynamics’ and oceanographic elements of ‘Remote Sensing’ were consolidated into the new Physical Oceanography program.

The division now consists of four programs: Physical Oceanography, Marine Meteorology, Optics and Biology, and NOPP.

6

Physical Oceanography Core Program

Three Major Thrust Areas:

• Boundary Layer ProcessesEmphasis: Air/Sea interactions, mixed layer dynamics, bottom boundary-layer

processes, flow interactions with topography, surface waves.

• Sub-Mesoscale Processes and ParameterizationEmphasis: frontal variability, internal waves, energy cascade processes,

temperature/salinity spiciness, turbulence, mixing.

• Prediction SystemsEmphasis: nowcast and forecast modeling, assimilation, filtering, and adjoint

methods, optimal and adaptive sampling, uncertainty.

We also consider the following two thrusts as areas of future growth:• Littoral Processes

Shallow water processes, highly nonlinear waves and internal waves, river plumes, estuaries.

• Ocean-Acoustic InteractionsCoupled ocean-acoustic modeling and transmission loss physics.

7

Additional Sources of Funding

In addition to funding we have available through the PO core program, we look for other opportunities to support our efforts…

• Departmental Research Initiatives within Code 32• Special Programs within ONR• Multidisciplinary University Research Initiatives• National Oceanographic Partnership Program

Note: our call for planning letters was delayed this year – the deadline for FY06 planning letters will likely be April 15th.

8

Programs associated with 322PO

Current DRI’s:

• ODDAS (Optimal Deployment of Drifting Acoustic Sensors)

• AESOP (Assessing the Effects of Submesoscale Ocean Parameterizations)

• NLIWI (Non-Linear Internal Wave Initiative)

New Starts for FY06: (just decided - all details still tentative)

• IndoEx (Predicting Variability in Choke Point Straits) Murray, Harper

• Radiance in a Dynamic Ocean (Understanding Radiance through the Air-Sea Interface) Ackleson, Drake, Vincent

• CBLAST Analysis (Supplement for focused collaboration and extended data analysis) Ferek, Friehe, Paluszkiewicz

9

Programs associated with 322PO

MURI’s:• ASAP (Adaptive Sampling and Prediction) Curtain• CSRE (Coherent Structures in Rivers and Estuaries) Vincent, Drake

• Real-time sea-state estimation for Ship Motion Prediction Vincent

NOPP Projects (just a few examples):

• GODAE/HYCOM (Global Ocean Data Assimilation Experiment / Hybrid Coordinate Ocean Model)

• PARADIGM (Partnership for Advancing Interdisciplinary Global Models)

New BAA: Assessment of Global Ocean Data Assimilation Experiment (GODAE) Boundary Conditions for Coastal Ocean Predictions

Proposals Due March 31st!

10

Undersea Persistent Surveillance

Special Program in Code 32 to explore the use of autonomous vehicles with cooperative behavior to sense and adapt to the environment to maximize detection of quiet submarines

Surveillance mission constraints:

• Unobtrusive undersea surveillance for targets in littoral waters of order 103 - 104 square nautical miles, shallow and deep, operating for months.

• Innovative technologies integrated into distributed scalable systems.

• Systems at all scales that are deployable, affordable and effective for large area, persistent coverage.

11

A simple view of ocean prediction

A better ocean prediction may result from improvements in:

• the initial conditions (via data assimilation, better climatology)• the boundary forcing (coupling with high fidelity atmospheric models,

application of appropriate lateral BC’s)• the robustness and reliability of the numerical model used for

integration.

InitialOceanState

PredictedOceanState

Numerical Model

Surface Forcing

Lateral Boundary Forcing

12

The AESOP DRI

In this DRI, we are only trying to address the fidelity of the numerical models - and further, we’re only examining issues concerning the subgrid-scale (SGS) parameterizations.

For the resolution of the models we’re considering, the SGS physics are in the submesoscale

on the order of ( 50m – 5km)

13

Some additional context

“Oceanic general circulation models must parameterize the effect of subgrid-scale motions and generally do so with diffusion terms. The following questions then arise: do oceanic observations allow inferences about mixing and diffusion coefficients – if so, how do ‘observed’ coefficients compare with those used in numerical models; can mixing rates be predicted from an understanding of the processes involved; does it matter; and are the results of numerical models sensitive to the choice of the diffusion coefficients?”

- from a report on the 1989 ‘Aha Huiliko’a meeting

14

Some additional context

“The parameterization of small-scale processes in numerical models is a basic science issue. It will only be resolved as numerical modelers and small-scale observers collaborate, pooling their expertise and resources and providing essential cross checks for each other. The meeting in Honolulu was an attempt to stimulate such collaboration, and it is encouraging that many of the participants left the meeting with joint projects on their minds – in the best of the Aloha spirit.”

- conclusion of the 1989 ‘Aha Huiliko’a meeting report

15

Some operational modeling plans at NAVO

FY07:

Global HYCOM at 1/12° (7 km) res

Regional NCOM or HYCOM at 2-4km res

Coastal model (tbd) at 500m-1500m res

FY09:

Global HYCOM at 1/25° (3.5 km) res

Regional model unnecessary

Coastal model (tbd) at 500m-1500m res

The need to assess the parameterizations that are being used in numerical models is more urgent than ever.

16

AESOP: An Analogy?

A Fable

The Goat and the Goatherd

A goatherd had sought to bring back a stray goat to his flock. He whistled and sounded his horn in vain; the straggler paid no attention to the summons. At last the Goatherd threw a stone, breaking the goat’s horn. He begged the Goat not to tell his master. The Goat replied, "Why, you silly fellow, the horn will speak though I be silent."

Moral: Do not attempt to hide things which cannot be hid

A Parameterization

Turbulence Closure Schemes

One can estimate the local eddy viscosity that arises from shear, dissipation, and buoyancy using the local Reynolds stresses and vertical shear along with a generic length scale for turbulent motions.

Moral: Do not attempt to hide things which cannot be hid (?)

17

Meeting Objectives

• Get everyone introduced, and understand what capabilities each group is bringing to the project

• Define and refine the specific project goals

• Determine the observations, model runs and diagnostics that will be required

• Draft an observation plan including– List of observational assets

– Platforms and vessels required

– Rough timeline of what is in the water and when

18

Meeting Objectives, continued

• Work out relevant partnering (Mixing Process Teams)

• Decide on the next meeting (host, location, date, goals, and what must be done by then)

• Determine if there are any crucial pieces of the puzzle missing from this group

19

Coordination with other programs

• ASAP MURI (What can we provide in real time? How can we use their operational predictions of the area?)

• UPS (observational assets and modeling work - e.g. Pierre Lermusiaux will be exploring different parameterizations in this area using HOPS)

• NSF (coordinate ship scheduling, coordinate with potential NSF-funded work in Monterey Bay)

• LOCO DRI (if they are in MB in ’06)• NPS observational program• CenCOOS observing system• MARS and the MBARI observational program• NOPP GODAE Boundary Condition group?• Other NOPP projects• Other ONR efforts from core programs (PO, MM, OA)• NRL modeling efforts (e.g. RTP for coupled modeling in MB)

20

Coordination with other programs

The coordination bit is important not only for the UNOLS ship requests, but also because there are a number of opportunities for parallel work, interesting science, and perhaps most importantly, developing a more comprehensive data set for making parameterization assessments.

The assumption behind the CPT initiative at NSF was that the data exists to verify the parameterizations for their class of problems. I’m not convinced we have the data set required for ours – that’s what we might be trying to build.

21

Straight from the workshop last year…

Some examples of questions we hope to address:

•What are the physical processes that are difficult to parameterize, limiting our ability to run realistic high-resolution simulations?

•How should one choose between parameterizations?

•What is an “improved” parameterization?

•Can we define the processes that dominate the dynamics at various resolutions?

•How sensitive are synoptic model predictions to the parameterized viscosity and diffusion?

22

Straight from the workshop last year…

Some desired products we hope to generate:

• new methods for assessing parameterizations

• new knowledge about submesoscale ocean processes

• insight into the relative importance of different processes at various resolutions

• information on how current parameterizations affect ocean predictions

• new methods to evaluate high-resolution models and their predictions