Seafloor and Sub-seafloor Data Interpolation and Minimal Data Density Analysis using Geometric and Multi-scale Markov Models

The objective of this project is to develop minimum data density and interpolation algorithms suitable for large volumes of Geospatial Information & Services (GI&S) survey data and for real-time or near-real-time use by decision makers. To accomplish this, we propose two different architectures to impose the Markov conditions -- the Geometric Markov model and the Multi-scale Markov model. The Markovian property in both architectures have direct connections to the underlying physical properties of natural processes. Furthermore, both architectures permit efficient computation algorithms to be implemented for both the data interpolation and the interpolation error estimation. This latter feature permits us to quickly assess different survey strategies and establish the ``best trade-off between resources invested and interpolation error.'' Phase I work will be conducted in the following areas: (1) Identification of Data Spatial Variability, (2) Data Interpolation, and (3) Minimal Data Density Analysis. The end-product of Phase I will be Matlab programs that implement the data interpolation and error estimation algorithms. Phase II end-product will be a prototype of the survey planning system readily to be integrated into Naval METOC Production Centers sensor systems. Our team consists of Scientific Systems Company, Inc. (SSCI) as the prime contractor, and Dr. John Goff of University of Texas Institute for Geophysics as a consultant for this project.