HR: 1340h
AN: H13A-0986    [Abstracts]
TI: UNH Data Cooperative: A Cyber Infrastructure for Earth System Studies
AU: Braswell, B H
EM: balazs.fekete@unh.edu
AF: University of New Hampshire, Complex Systems Research Center, Durham, NH 03824, United States
AU: * Fekete, B M
EM: balazs.fekete@unh.edu
AF: University of New Hampshire, Complex Systems Research Center, Durham, NH 03824, United States
AU: Prusevich, A
EM: alex.proussevitch@unh.edu
AF: University of New Hampshire, Complex Systems Research Center, Durham, NH 03824, United States
AU: Gliden, S
EM: Stanley.Glidden@unh.edu
AF: University of New Hampshire, Complex Systems Research Center, Durham, NH 03824, United States
AU: Magill, A
EM: alison.magill@unh.edu
AF: University of New Hampshire, Complex Systems Research Center, Durham, NH 03824, United States
AU: Vorosmarty, C J
EM: charles.vorosmarty@unh.edu
AF: University of New Hampshire, Complex Systems Research Center, Durham, NH 03824, United States
AB: Earth system scientists and managers have a continuously growing demand for a wide array of earth observations derived from various data sources including (a) modern satellite retrievals, (b) "in-situ" records, (c) various simulation outputs, and (d) assimilated data products combining model results with observational records. The sheer quantity of data, and formatting inconsistencies make it difficult for users to take full advantage of this important information resource. Thus the system could benefit from a thorough retooling of our current data processing procedures and infrastructure. Emerging technologies, like OPeNDAP and OGC map services, open standard data formats (NetCDF, HDF) data cataloging systems (NASA-Echo, Global Change Master Directory, etc.) are providing the basis for a new approach in data management and processing, where web- services are increasingly designed to serve computer-to-computer communications without human interactions and complex analysis can be carried out over distributed computer resources interconnected via cyber infrastructure. The UNH Earth System Data Collaborative is designed to utilize the aforementioned emerging web technologies to offer new means of access to earth system data. While the UNH Data Collaborative serves a wide array of data ranging from weather station data (Climate Portal) to ocean buoy records and ship tracks (Portsmouth Harbor Initiative) to land cover characteristics, etc. the underlaying data architecture shares common components for data mining and data dissemination via web-services. Perhaps the most unique element of the UNH Data Cooperative's IT infrastructure is its prototype modeling environment for regional ecosystem surveillance over the Northeast corridor, which allows the integration of complex earth system model components with the Cooperative's data services. While the complexity of the IT infrastructure to perform complex computations is continuously increasing, scientists are often forced to spend considerable amount of time to solve basic data management and preprocessing tasks and deal with low level computational design problems like parallelization of model codes. Our modeling infrastructure is designed to take care the bulk of the common tasks found in complex earth system models like I/O handling, computational domain and time management, parallel execution of the modeling tasks, etc. The modeling infrastructure allows scientists to focus on the numerical implementation of the physical processes on a single computational objects(typically grid cells) while the framework takes care of the preprocessing of input data, establishing of the data exchange between computation objects and the execution of the science code. In our presentation, we will discuss the key concepts of our modeling infrastructure. We will demonstrate integration of our modeling framework with data services offered by the UNH Earth System Data Collaborative via web interfaces. We will layout the road map to turn our prototype modeling environment into a truly community framework for wide range of earth system scientists and environmental managers.
DE: 1847 Modeling
DE: 1848 Monitoring networks
DE: 1855 Remote sensing (1640)
SC: Hydrology [H]
MN: 2007 Fall Meeting