Complex spatial control problems can be computationally intensive. Timely response in urgent spatial control situations such as wildfire control poses great challenges for the efficient solving of spatial control problems. Web-based and service-oriented architectures of integrating geographic information system (GIS) clients and parallel computing resources have been suggested as an effective paradigm to solve computationally intensive spatial problems. Such real-time coupling framework is highly dependent upon interactivity and on-demand availability of dedicated parallel computing resources appropriate for the problem. We present an approach to enhancing the efficiency of solving spatial control problems while offering another coupling framework of integrating computing resources from desktop GIS and parallel computing environments to alleviate such dependency. Specifically, a model knowledge database is developed to bridge the gap between desktop GIS models and parallel computing resources. Desktop GIS models can iteratively improve themselves by steering rules retrieved from the model knowledge database. To examine its effectiveness, we applied the framework to a wildfire control case. Simulation results show dramatic reduction in computation time of the improved desktop GIS model, and indicate that desktop GIS models enhanced by model knowledge databases can be useful in providing timely assistance on computationally intensive spatial control problems.

This paper presents a general framework to utilize high performance computations in regional ecosystem simulation. First, a comprehensive modeling package is introduced to demonstrate the challenges encountered due to the multiple spatial and temporal scales which arise in regional ecosystem modeling. Second, a parallel simulation framework is presented to support multi-component ecosystem modeling on high performance computational platforms. Third, two ecological models are summarized and presented as an example of model integration in the simulation framework. This work presents the first component-based integrated regional ecosystem simulation for natural resource management on high performance computational platforms.

A distributed simulation framework is presented to enable natural resource managers to take advantage of both Geographic Information System (GIS) functionality and computationally intensive ecological modeling. Based on one of the newest products from a major vendor of GIS software (ESRI’s ArcGIS engine), a user interface was developed to manipulate and visualize digital maps and spatially explicit simulation results. An objective is to allow access to computationally intensive simulations while utilizing an interface system with which natural resource managers have extensive prior experience. Distributed communication services were also developed to support spatially explicit ecological modeling on both local and remote computational platforms. Such a capability is important for natural resource agencies which may have little in-house capacity for extensive computation and could greatly benefit from the use of a computing grid. To provide examples of the practical application of such a system, we present two ecological modeling cases. These models are components of a major ecological multimodel, the Across Trophic Level System Simulation (ATLSS), developed for application to one of the world’s largest ecological restoration projects, in the Everglades of South Florida. The first example is a sequential model to show the working procedure for short-time simulations on a single platform. The second example is a parallel model to demonstrate the procedure for long-time simulations on remote high performance computational platforms. In both cases, we illustrate how the framework provides user access to computationally intensive dynamic, spatially-explicit simulations, enhancing the capabilities available through a stand-alone GIS, while retaining many of the benefits of a GIS.

The Florida panther (Puma concolor coryi) is an endangered, wide-ranging predator whose habitat needs conflict with a rapidly growing human population. Our goal was to identify specific regions of the south Florida landscape that are of high conservation value to support a self-sustaining panther population. We used compositional and Euclidean distance analyses to determine relative importance of various land cover types as panther habitat and to investigate the role of forest patch size in habitat selection. A model of landscape components important to Florida panther habitat conservation was created. The model was used in combination with radio telemetry records, home range overlaps, land use/land cover data, and satellite imagery to delineate Primary and Secondary zones that would comprise a landscape mosaic of cover types sufficient to support a self-sustaining population. The Primary Zone generally supports the present population and is of highest conservation value, while the Secondary Zone is of lesser value but could accommodate expansion of the population given sufficient habitat restoration. Least-cost path models identified important landscape linkages, and model results were used to delineate a Dispersal Zone to accommodate future panther dispersal outside of south Florida. We determined that the three habitat zones could support 80–94 panthers, a population likely to persist and remain stable for 100 years, but that would be subject to continued genetic problems. The Primary, Dispersal and Secondary zones comprise essential components of a landscape-scale conservation plan for the protection of a viable Florida panther population in south Florida. Assessments of potential impacts of developments should strive to achieve no net loss of landscape function or carrying capacity for panthers within the Primary Zone or throughout the present range of the Florida panther.

A distributed simulation framework is presented to enable natural resource managers to take advantage of both Geographic Information System (GIS) functionality and computationally intensive ecological modeling. Based on one of the newest products from a major vendor of GIS software (ESRI’s ArcGIS engine), a user interface was developed to manipulate and visualize digital maps and spatially explicit simulation results. An objective is to allow access to computationally intensive simulations while utilizing an interface system with which natural resource managers have extensive prior experience. Distributed communication services were also developed to support spatially explicit ecological modeling on both local and remote computational platforms. Such a capability is important for natural resource agencies which may have little in-house capacity for extensive computation and could greatly benefit from the use of a computing grid. To provide examples of the practical application of such a system, we present two ecological modeling cases. These models are components of a major ecological multimodel, the Across Trophic Level System Simulation (ATLSS), developed for application to one of the world’s largest ecological restoration projects, in the Everglades of South Florida. The first example is a sequential model to show the working procedure for short-time simulations on a single platform. The second example is a parallel model to demonstrate the procedure for long-time simulations on remote high performance computational platforms. In both cases, we illustrate how the framework provides user access to computationally intensive dynamic, spatially-explicit simulations, enhancing the capabilities available through a stand-alone GIS, while retaining many of the benefits of a GIS.

The endangered Florida panther (Puma concolor coryi) shares its shrinking habitat with agriculture, surface mining, and rapid urban growth. Although panthers have extensive home ranges and use diverse land covers, methods that dominate panther habitat evaluation for Endangered Species Act (ESA) consultations and regional land use planning consider only forested day-use elements within the landscape mosaic. Maehr and Deason (2002) present a Panther Habitat Evaluation Model (PHEM) that, in addition to excluding nonforested habitat, reduces the assessed value of forest patches based on criteria for patch size, forest type, proximity to a "core" area, and connectivity to other patches. An examination of the foundations of PHEM is therefore warranted. Building oil earlier work that included an evaluation of panther habitat selection studies (Comiskey et al. 2002), we examine PHEM in light of data quality criteria and the panther's known life history requirements. We conclude that the precepts and rules of the PHEM methodology are based on unwarranted assumptions, nonstandard methods of analysis, and exclusion of relevant data, leading to an undue emphasis on day-use land cover and forest patches larger than 500 ha. Large areas of southern Florida that have abundant prey and are intensively used by panthers would score low in PHEM habitat assessments because they lack large forest patches. We discuss the conservation implications of applying a methodology that discounts substantial portions of Occupied panther habitat as unsuitable, and describe all alternative approach to habitat definition and evaluation that is both consistent with panther habitat requirements and applicable to conservation decision-making. Conserving sufficient habitat for recovery of the panther extends an umbrella of protection to the many species that dwell within its range.

As part of the effort to restore the ∼10 000-km2 Everglades drainage in southern Florida, USA, we developed spatially explicit species index (SESI) models of a number of species and species groups. In this paper we describe the methodology and results of three such models: those for the Cape Sable Seaside Sparrow and the Snail Kite, and the species group model of long-legged wading birds. SESI models are designed to produce relative comparisons of one management alternative to a base scenario or to another alternative. The model outputs do not provide an exact quantitative prediction of future biotic group responses, but rather, when applying the same input data and different hydrologic plans, the models provide the best available means to compare the relative response of the biotic groups. We compared four alternative hydrologic management scenarios to a base scenario (i.e., predicted conditions assuming that current water management practices continue). We ranked the results of the comparisons for each set of models. No one scenario was beneficial to all species; however, they provide a uniform assessment, based on the best available observational information, of relative species responses to alternative water-management plans. As such, these models were used extensively in the restoration planning.

A major environmental restoration effort is under way that will affect the Everglades and its neighboring ecosystems in southern Florida. Ecosystem and population-level modeling is being used to help in the planning and evaluation of this restoration. The specific objective of one of these modeling approaches, the Across Trophic Level System Simulation (ATLSS), is to predict the responses of a suite of higher trophic level species to several proposed alterations in Everglades hydrology. These include several species of wading birds, the snail kite, Cape Sable seaside sparrow, Florida panther, white-tailed deer, American alligator, and American crocodile. ATLSS is an ecosystem landscape-modeling approach and uses Geographic Information System (GIS) vegetation data and existing hydrology models for South Florida to provide the basic landscape for these species. A method of pseudotopography provides estimates of water depths through time at 28 × 28-m resolution across the landscape of southern Florida. Hydrologic model output drives models of habitat and prey availability for the higher trophic level species. Spatially explicit, individual-based computer models simulate these species. ATLSS simulations can compare the landscape dynamic spatial pattern of the species resulting from different proposed water management strategies. Here we compare the predicted effects of one possible change in water management in South Florida with the base case of no change. Preliminary model results predict substantial differences between these alternatives in some biotic spatial patterns.

When ecosystems are fragmented into patches, the whole can be worth less than the sum of its parts. Parallel methods can greatly speed up statistical analysis of clusters, in landscape ecology or other fields. The term 'landscape ecology' refers to the analysis of patterns and heterogeneity in natural landscapes and ecosystems. Computer modeling is used in landscape ecology applications to assess habitat fragmentation and its implications. Researchers in the Environmental Sciences Division at Oak Ridge National Laboratory developed a model called Noyelp that simulates the search, movement, and foraging activities of free-ranging elk and bison on winter range in northern Yellowstone National Park. The model helps to explore how the scale and patterns of fire affect winter foraging and survival of ungulate populations in the diverse, multihabitat landscape of the park. This model, written in Fortran-77, analyzes maps (2D grids) to determine the number, size, and geometry of habitat regions, or clusters, representing landscape patterns, resources, and animals.< >