Satellite Geodesy at the Scripps Institution of Oceanography, University of California San Diego

http://topex.ucsd.edu/

Geodesy is the field of science that is involved in the measurement of the size and shape of the earth as well as its gravity field. Modern geodetic tools such as the global positioning system (GPS), radar altimetry, laser altimetry, synthetic aperture radar, and satellite-to-satellite tracking are accurate enough to monitor time variations in the earth related to plate tectonics, post-glacial, ocean circulation and atmospheric circulation.

Modern geodesy attempts to solve geophysical problems by assimilating observable phenonema (such as variations in the Earth's rotation, gravity, geocenter, and surface deformations) into models. Today, these observations come from a variety of sources including Satellite Laser Ranging (SLR), Very-Long-Baseline Interferometry (VLBI), the Global Positioning System (GPS), and Doppler Orbitography and Radiopositioning Integrated by Satellite (DORIS).

Discovery and analysis of topographic features using learning algorithms: A seamount case study Andrew P. Valentine, Lara M. Kalnins, Jeannot Trampert

http://onlinelibrary.wiley.com/doi/10.1002/grl.50615/full

Abstract

Identifying and cataloging occurrences of particular topographic features are important but time-consuming tasks. Typically, automation is challenging, as simple models do not fully describe the complexities of natural features. We propose a new approach, where a particular class of neural network (the “autoencoder”) is used to assimilate the characteristics of the feature to be cataloged, and then applied to a systematic search for new examples. To demonstrate the feasibility of this method, we construct a network that may be used to find seamounts in global bathymetric data. We show results for two test regions, which compare favorably with results from traditional algorithms.

Sparse autoencoder Lecture Notes by Andrew Ng

http://www.stanford.edu/class/cs294a/sparseAutoencoder.pdf


These notes describe the sparse autoencoder learning algorithm, which
is one approach to automatically learn features from unlabeled data. In some
domains, such as computer vision, this approach is not by itself competitive
with the best hand-engineered features, but the features it can learn do turn
out to be useful for a range of problems (including ones in audio, text, etc).

Further, there're more sophisticated versions of the sparse autoencoder (not
described in these notes, but that you'll hear more about later in the class)
that do surprisingly well, and in many cases are competitive with or superior
to even the best hand-engineered representations.

Seamount Biogeosciences Network Catalog

The Seamount Catalog is a digital archive for bathymetric seamount maps that can be viewed and downloaded in various formats. This catalog contains morphological data, sample information, related grid and multibeam data files, as well as user-contributed files that all can be downloaded. Currently this catalog contains more than 1,800 seamounts from all the oceans.

http://earthref.org/SC/