Description:

Session discussing the modeling of coastal processes using AI methods applied to the nearshore waters, beach, dunes, marshes, for natural and anthropogenic processes and to include remote sensing data and the automated processing of coastal data.

Applications of Artificial Intelligence to the Coastal Environment:

Novel techniques are being developed and applied to modeling of coastal processes in nearshore waters. Artificial Intelligence (AI) methods can be applied to the analysis and simulation of nearshore waters, beach dunes, and marshes, for both natural and anthropogenic processes. AI techniques can be used to analyze remote sensing data and the automated processing of coastal data, and offer advantages over traditional techniques. This session focuses on the development and application of novel AI techniques to conditions in the coastal environment.

Artificial Intelligence Applied to Remote Sensing:

We present results of work to develop a dynamic over-ocean radiometric bias correction for the NOAA’s Microwave Integrated Retrieval System (MiRS) which accounts for spatial, temporal, spectral, and angular dependence of the systematic differences between observed and forward model-simulated radiances. The dynamic bias correction, which utilizes a deep neural network approach, is designed to incorporate dependence on the atmospheric and surface conditions that impact forward model biases. This approach utilizes collocations of observed SNPP/ATMS radiances and ECMWF model analyses, which are used as input to the Community Radiative Transfer Model (CRTM) forward model, to develop the training data of radiometric biases.

Analysis of the neural network performance on independent data indicates that in many channels, the dynamic bias is able to reproduce realistically both the spatial patterns of the original observed – simulated radiance variability and its probability distribution function. Furthermore, MiRS retrieval impact experiments show that, compared with the baseline static bias correction, using the dynamic bias correction can improve temperature and water vapor profile retrievals, particularly in regions with higher cloud liquid water (CLW) amounts. Additionally, ocean surface emissivity retrievals are improved, for example at 23.8 GHz showing an increase in correlation from 0.59 to 0.67 and a reduction of RMS error from 0.036 to 0.027. The Total Precipitable Water (TPW) retrieval bias over ocean is also reduced significantly, along with its angular dependence. Further testing is underway to assess the suitability of the dynamic bias correction approach for operational implementation in MiRS.