Information Exploration

Advances in forecasting and climate modeling require new computational methods to allow scientists to explore the complex high dimensional relationships in the remotely sensed data gathered by NOAA and its international partners.


Ncep Analysis  

Clustering : Non-Linear, Multi-Variate Analysis of Slow Climate Variations

Our ultimate goal is to determine the non-linear, multi-variate relationships that describe the processes coupling the atmosphere and ocean (and land) states and circulations and producing the slow modes of climate variation. To do that we need new, practical, analysis methodologies that can handle the volume of data and do not limit the analysis to particular space-time scales or to assumptions of linearity and/or low-dimensionality.

Algae Classification  

Classification : HarmFul Algae Bloom Classification

Algae are the most abundant photosynthetic organisms in marine ecosystems and are essential components of marine food webs harmful algal bloom states (HABs) can pose health hazards for humans or animals through the production of toxins or bioactive compounds. They also can cause deterioration of water quality through the buildup of high biomass, which degrades aesthetic, ecological, and recreational values. We have developed a classifier to predict the presence of HABs using satellite observations of the coast.



Data Segmentation: Clustering Analysis for Cloud and Surface Type Classification 

We have developed a clustering library that includes a variety of clustering algorithms, such as K-means and mean shift clustering, for use in the analysis of multi-modal MODIS level 1B data.

© 2010, NOAA-CREST CSDIRS, CCNY-Glasslab/Michael Grossberg/Irina Gladkova/Paul Alabi

Powered by LFC — Lightning Fast CMS

All rights reserved