Project summary


Computational dynamic landscape manipulation and optimization for OPEN source GIS


H. Mitasova, L. Mitas, R. StAmant, T. Drake, North Carolina State University, Raleigh, NC


New mapping and monitoring technologies generate large volumes of high resolution spatio-temporal data and provide unique opportunities to dramatically improve land management and environmental protection. The major challenge is to develop a new generation of computational tools for processing and analysis of these data and their effective use for landscape modeling and optimization. The goal of this project is to develop virtual environment and computational approaches supporting sustainable landscape development, test it on case studies and provide the results to the community through the Open Source Geographic Information System (GIS) infrastructure. The proposed research strategy is based on a close collaboration between GIScience, computer science, earth science, and computational physics with the focus on the following developments :


  1. Multi-scale, 3D dynamic representation of landscapes directly linked with field sensors providing continuous updates on the landscape state and its changes. The field use of augmented reality will be explored for effective selection of monitoring sites and location of high risk areas based on virtual representations of future landscape development.

  2. Spatial simulation of landscape processes at multiple spatial and temporal scales with data adaptive capabilities. Landscape process simulations will provide the necessary diagnostic tools for assessment of the current state, prediction tools for evaluation of the planned land use impacts and tools supporting landscape manipulation and optimization with a potential to develop innovative conservation strategies. New robust and scalable methods for simulation of landscape processes will be explored including path sampling (path integral Monte Carlo) approaches based on the duality between particles and fields.

  3. Spatial optimization of complex systems. The methods will be aimed at finding optimal compromises between conflicting demands or finding landscape pattern solutions within prescribed constraints. In order to transform the optimization of multi-variate fields into a tractable form, basis set expansions, mapping on statistical mechanical models and robust minimization techniques based on simulated annealing or genetic algorithms will be employed.

  4. Intelligent virtual environment for simulation and interaction with landscapes. The development will focus on intelligent visual tools for interactive manipulation of landscape which will allow the user to select a general object (e.g. conservation measure) and locate it at a desired site with automatic adapting of its shape and properties using the data provided by topographic analysis and simulation of relevant processes. The system could provide layout of the initial landscape pattern as well as suggestions on possible alternatives.

  5. Open Source implementation. The developed methods and tools will be implemented within Open source GIS environment based on the latest release of GRASS5. GRASS (Geographic Resources Analysis Support System) is one of the top ten Open Source projects and it has played a pioneering role in integrating GIS with environmental modeling and 3D dynamic landscape visualization. The implementation will contribute to the rapidly growing open source geospatial computing infrastructure and benefit from modifications and enhancements by international team of GIS developers. It will be also leveraged by the NCSU/IBM/RedHat Open Source Initiative which includes seamless integration of LINUX into the NCSU academic computing environment.


The main expected outcome of the proposed project will be the integration of intelligent virtual environment with simulations supporting virtual dynamic landscape prototyping. It will enable the users visualize and interact with 3D dynamic landscapes and introduce modifications of the landscape features during an ongoing simulation. The interaction with landscape processes will allow users to "feel" the impact of their actions and explore various solutions. This would be especially important for the decision making process when the stake holders with very specific and often differing views (e.g. developers, local government, landowners, environmental groups) will be able to present their proposals and concerns and the entire group will be able to view and evaluate the impacts.

The developed concepts and prototypes will be tested using two different study areas: a) Centennial Campus - a new Technopolis community being developed by NCSU, b) section of highly dynamic North Carolina coastal region.

The proposed virtual landscape environment will be also used as a demonstration and educational tool for stake holders in watershed and coastal communities, local government, and NC museums, by providing visual representations of human impacts on landscape processes. The environment will be used to train students and professionals to use the advanced GIS, simulation, visualization and field computing to support land use management.