Export 5 results:
Sort by: Author Title Type [ Year  (Desc)]
Mazloff, MR, Cornuelle BD, Gille ST, Verdy A.  2018.  Correlation lengths for estimating the large-scale carbon and heat content of the Southern Ocean. Journal of Geophysical Research-Oceans. 123:883-901.   10.1002/2017jc013408   AbstractWebsite

The spatial correlation scales of oceanic dissolved inorganic carbon, heat content, and carbon and heat exchanges with the atmosphere are estimated from a realistic numerical simulation of the Southern Ocean. Biases in the model are assessed by comparing the simulated sea surface height and temperature scales to those derived from optimally interpolated satellite measurements. While these products do not resolve all ocean scales, they are representative of the climate scale variability we aim to estimate. Results show that constraining the carbon and heat inventory between 35 degrees S and 70 degrees S on time-scales longer than 90 days requires approximately 100 optimally spaced measurement platforms: approximately one platform every 20 degrees longitude by 6 degrees latitude. Carbon flux has slightly longer zonal scales, and requires a coverage of approximately 30 degrees by 6 degrees. Heat flux has much longer scales, and thus a platform distribution of approximately 90 degrees by 10 degrees would be sufficient. Fluxes, however, have significant subseasonal variability. For all fields, and especially fluxes, sustained measurements in time are required to prevent aliasing of the eddy signals into the longer climate scale signals. Our results imply a minimum of 100 biogeochemical-Argo floats are required to monitor the Southern Ocean carbon and heat content and air-sea exchanges on time-scales longer than 90 days. However, an estimate of formal mapping error using the current Argo array implies that in practice even an array of 600 floats (a nominal float density of about 1 every 7 degrees longitude by 3 degrees latitude) will result in nonnegligible uncertainty in estimating climate signals.

Hoteit, I, Hoar T, Gopalakrishnan G, Collins N, Anderson J, Cornuelle B, Kohl A, Heimbach P.  2013.  A MITgcm/DART ensemble analysis and prediction system with application to the Gulf of Mexico. Dynamics of Atmospheres and Oceans. 63:1-23.   10.1016/j.dynatmoce.2013.03.002   AbstractWebsite

This paper describes the development of an advanced ensemble Kalman filter (EnKF)-based ocean data assimilation system for prediction of the evolution of the loop current in the Gulf of Mexico (GoM). The system integrates the Data Assimilation Research Testbed (DART) assimilation package with the Massachusetts Institute of Technology ocean general circulation model (MITgcm). The MITgcm/DART system supports the assimilation of a wide range of ocean observations and uses an ensemble approach to solve the nonlinear assimilation problems. The GoM prediction system was implemented with an eddy-resolving 1/10th degree configuration of the MITgcm. Assimilation experiments were performed over a 6-month period between May and October during a strong loop current event in 1999. The model was sequentially constrained with weekly satellite sea surface temperature and altimetry data. Experiments results suggest that the ensemble-based assimilation system shows a high predictive skill in the GoM, with estimated ensemble spread mainly concentrated around the front of the loop current. Further analysis of the system estimates demonstrates that the ensemble assimilation accurately reproduces the observed features without imposing any negative impact on the dynamical balance of the system. Results from sensitivity experiments with respect to the ensemble filter parameters are also presented and discussed. (C) 2013 Elsevier B.V. All rights reserved.

Sarkar, J, Cornuelle BD, Kuperman WA.  2011.  Information and linearity of time-domain complex demodulated amplitude and phase data in shallow water. Journal of the Acoustical Society of America. 130:1242-1252.   10.1121/1.3613709   AbstractWebsite

Wave-theoretic ocean acoustic propagation modeling is used to derive the sensitivity of pressure, and complex demodulated amplitude and phase, at a receiver to the sound speed of the medium using the Born-Frechet derivative. Although the procedure can be applied for pressure as a function of frequency instead of time, the time domain has advantages in practical problems, as linearity and signal-to-noise are more easily assigned in the time domain. The linearity and information content of these sensitivity kernels is explored for an example of a 3-4 kHz broadband pulse transmission in a 1 km shallow water Pekeris waveguide. Full-wave observations (pressure as a function of time) are seen to be too nonlinear for use in most practical cases, whereas envelope and phase data have a wider range of validity and provide complementary information. These results are used in simulated inversions with a more realistic sound speed profile, comparing the performance of amplitude and phase observations. (C) 2011 Acoustical Society of America. [DOI: 10.1121/1.3613709]

Hoteit, I, Cornuelle B, Kim SY, Forget G, Kohl A, Terrill E.  2009.  Assessing 4D-VAR for dynamical mapping of coastal high-frequency radar in San Diego. Dynamics of Atmospheres and Oceans. 48:175-197.   10.1016/j.dynatmoce.2008.11.005   AbstractWebsite

The problem of dynamically mapping high-frequency (HF) radar radial velocity observations is investigated using a three-dimensional hydrodynamic model of the San Diego coastal region and an adjoint-based assimilation method. The HF radar provides near-real-time radial velocities from three sites covering the region offshore of San Diego Bay. The hydrodynamical model is the Massachusetts Institute of Technology general circulation model (MITgcm) with 1 km horizontal resolution and 40 vertical layers. The domain is centered on Point Loma, extending 117 km offshore and 120 km alongshore. The reference run (before adjustment) is initialized from a single profile of T and S and is forced with wind data from a single shore station and with zero heat and fresh water fluxes. The adjoint of the model is used to adjust initial temperature, salinity, and velocity, hourly temperature, salinity and horizontal velocities at the open boundaries, and hourly surface fluxes of momentum, heat and freshwater so that the model reproduces hourly HF radar radial velocity observations. Results from a small number of experiments suggest that the adjoint method can be successfully used over 10-day windows at coastal model resolution. It produces a dynamically consistent model run that fits HF radar data with errors near the specified uncertainties. In a test of the forecasting capability of the San Diego model after adjustment, the forecast skill was shown to exceed persistence for up to 20 h. (C) 2008 Elsevier B.V. All rights reserved.

Hursky, P, Porter MB, Cornuelle BD, Hodgkiss WS, Kuperman WA.  2004.  Adjoint modeling for acoustic inversion. Journal of the Acoustical Society of America. 115:607-619.   10.1121/1.1636760   AbstractWebsite

The use of adjoint modeling for acoustic inversion is investigated. An adjoint model is derived from a linearized forward propagation model to propagate data-model misfit at the observation points back through the medium to the medium perturbations not being accounted for in the model. This adjoint model can be used to aid in inverting for these unaccounted medium perturbations. Adjoint methods are being applied to a variety of inversion problems, but have not drawn much attention from the underwater acoustic community. This paper presents an application of adjoint methods to acoustic inversion. Inversions are demonstrated in simulation for both range-independent and range-dependent sound speed profiles using the adjoint of a parabolic equation model. Sensitivity and error analyses are discussed showing how the adjoint model enables calculations to be performed in the space of observations, rather than the often much larger space of model parameters. Using an adjoint model enables directions of steepest descent in the model parameters (what we invert for) to be calculated using far fewer modeling runs than if a forward model only were used. (C) 2004 Acoustical Society of America.