Using USGS "Resource for Advanced Modeling" to connect climate drivers to biological responses

Related Research Areas
Climate Variability & Change
Project Description
Project Purpose This project will work to improve the USGS’s “Resource for Advanced Modeling”, which serves the Department of Interior (DOI) habitat modeling needs by bringing USGS and NASA resources together in a distributed modeling system that maximizes the complementary strengths of both NASA and the USGS. To demonstrate the utility of this capacity the project will focus on both invasive species and threatened and endangered species; two critical land management issues. The proposal is a hybrid; combining technical capacity with a natural resource applications focus. The Resource for Advanced Modeling: a USGS Decision Support Tool for species distribution modeling Over the past several years the USGS Fort Collins Science Center (the FORT) has established its “Resource for Advanced Modeling” (RAM). This was designed as a collaborative workspace where resource managers could meet with USGS scientists to develop predictive spatial habitat models. These models feed managers spatial information on where to focus their limited resources. The high standard for collaborative work fostered through the RAM was recognized by the Department of Interior in 2009 through its “Partners in Conservation Award”. Currently there are three major constraints limiting the ability for the RAM to serve the greater DOI community, given current resource and personnel limitations at the FORT: 1. The large amount of time, effort, and iterations needed to ensure the best possible spatial data layers are available for the modeling exercise, 2. The computational time required to run models, and 3. Maintaining a strategic and documented procedure to analyze the nearly infinite combination of input data layers, modeling techniques and their related parameters, and the different biological responses that are being modeled. Summary of Project Objectives The results of this project will allow the RAM to leverage off of: 1. Cutting-edge data sources available through NASA’s Terrestrial Observation and Prediction System (TOPS) 2. High-performance computational resources available through NASA Earth Exchange at ARC 3. VisTrails provenance-enabled scientific workflow middleware developed at the University of Utah with NSF and DOE funding. Component Overview The following figure on the left shows the main components of the project. The main concept is utilize both data and computational capacity available at NASA Ames to improve the accuracy and speed of models formulated at the USGS Resource for Advanced Modeling. The interaction between the RAM and Ames will be facilitated by the VisTrails software. The connection to the land management community will come through the interaction with CSU and the North Central Climate Science Center. The graphic on the right represents the impetus for the project. The USGS holds, within the RAM, the field data and modeling code. Many of the predictor layers used in the RAM come from NASA or could be generated within the NASA Ames TOPS project. So, instead of trying to bring all the predictor layers into the RAM, why not send the field data and model code to Ames? The project will work toward running the model in the most efficient and effective location. Study Area The following figure depicts a map showing the four main Landscape Conservation Cooperatives within the domain of the North Central Climate Science Center (note that the boundaries of the NC CSC domain are intentionally “fuzzy” to allow studies to consider areas that are ecologically relevant).
Project Administrator(s):
Petr Votava


Petr Votava