University of Minnesota
University Relations

Minnesota Supercomputing Institute

Log out of MyMSI

Research Abstracts Online
January 2010 - March 2011

Main TOC ...... Next Abstract

University of Minnesota Twin Cities
College of Science and Engineering
Department of Civil Engineering
St. Anthony Falls Laboratory

PI: Efi Foufoula-Georgiou, Fellow

Modeling Variability in the Earth System: From Rainfall, to River Networks, to Landscape Processes

Understanding and quantifying the space-time variability of precipitation at local to global scales is essential for improving hydrologic predictions, parameterizing land-atmosphere interactions, and interpreting the effects of climate change on the redistribution of atmospheric and land surface fluxes. Datasets of multi-sensor rainfall observations are becoming richer and larger as a result of existing and planned Precipitation satellite missions, increasing thus the need to develop robust methodologies for taking full advantage of these data sets for scientific and operational studies. These researchers are working on developing multi-sensor merging techniques to optimally combine different sources (space-borne passive and active microwave sensors, ground- based radars and point gauges) of precipitation at various resolutions. As these sources of information are provided as very large size images, this often requires solving a complex and computationally intensive nonlinear optimization problem for multi-dimensional data sets. Therefore, having access to powerful computational resources and parallel computational capacity is essential for efficient implementation of this research.

Rainfall is the driver of land surface processes, including erosion and sedimentation and the flow of water and pollutants in streams, floodplains, and river networks. The availability of high-resolution topography from LIDAR (light detection and ranging, an optical remote sensing technology) calls for the development of new methodologies able to extract geomorphologic features (channel heads, channel banks, channel networks, slopes, curvatures, etc.) automatically. The researchers have developed a geometric framework that combines nonlinear diffusion for the pre-processing of the data and geodesic minimization principles for the extraction of channels. The nonlinear filtering operation allows them to focus the analysis on the scales of interest and to enhance features that are critical to the channel extraction. Channels are extracted as geodesics, or curves of minimal effort, where the effort is measured based on fundamental geomorphological characteristics such as flow accumulation and iso-height contours curvature. Because of the large size of LIDAR datasets, computationally efficient algorithms need to be developed to model flow and transport in large watersheds. Also new concepts, such as that of a dynamic tree (defined from the underlying static river network structure and the space-time variable processes operating on it) are explored to rescale fluxes from one scale to another for prediction purposes.

Group Members

Bidroha Basu, Graduate Student
Mohammad Ebtehaj, Graduate Student
Naga Vamsi K. Ganti, Graduate Student
Deborah Nykanen, Department of Mechanical and Civil Engineering, Minnesota State University, Mankato, Minnesota
Paola Passalaqua, Graduate Student
Jon Schwenk, Graduate Student
Arvind Kumar Singh, Graduate Student