Daolan Zheng

Edge effects in fragmented landscapes: a generic model for delineating area of edge influences (D-AEI)

We developed a generic model for delineating area of edge influences D-AEI for quantifying edge effects within a landscape by combining remote sensing, geographic information systems (GIS), moving window (3 3), and computer programming techniques. Our model provided a more realistic assessment of edge effects than those based on traditional methods. Unique characteristics of the D-AEI model included: (1) preservation of the spatial characteristics of the landscape structure; (2) incorporation of the most critical parameters controlling edge effects, such as edge orientation, edge contrast, prevailing direction of edge effects, decay value, and interior approximation; and (3) ability to quantify edge effects for various variables at multiple scales. The model is flexible so that the users can define key parameters and generate ecologically relevant output based on environmental and spatial characteristics of the study area and the study purpose. Our results demonstrated that: (a) edge effects were not symmetrically distributed in all directions around clearcuts; (b) AEI was not necessarily continuous around patches; and (c) boundary dynamics and multiple edge effects were clearly reflected across the landscape. Results from this research are important for current and future resource assessments, biological conservation and wildlife habitat management, biodiversity studies of flora and fauna, microclimatic research, future studies on edges and their importance in landscape design and analysis. The model has potential for broader applications in other research areas where human and natural disturbances are evident, at multiple scales from watershed, forest management district, to region.

Fire Spread Regulated by Weather, Landscape Structure, and Management in Wisconsin Oak- Dominated Forests and New Jersey Pinelands

This study is to predict fire spread behavior and burned area across two fire-prone landscapes with contrasting vegetation (Oak-dominated ecosystem in WI vs. pine-dominated ecosystem in NJ), fuel-type composition, and land-use history regulated by the effects of weather, landscape structure and land management by combining simulations from three models (FARSITE, HARVEST, AND FRAGSTATS) under different scenarios. The results demonstrate:\1- substantial differences in fire-spread patterns between the two landscapes were observed when holding weather conditions constant and excluding roads, indicating that landscape fragmentation is a main controlling factor on fire spread at the landscape level; ➁ roads functioning as barriers could significantly reduce the burned area from fire spread; and ➂ Harvesting effects showed different trends, depending on landscape fuel type composition and weather conditions. At 4% harvesting intensity, both clustered and dispersed methods showed no significant impact (a=0.01) on reducing the mean burned area across the more fragmented WI landscape, but showed significant effects on fire spread in the less fragmented NJ landscape in summer when weather was hot and dry.

Simulating Fire Spread with Landscape Level Edge Fuel Scenarios

Area-of-edge influence (AEI) is sometimes the dominant element of many forested landscapes. Patch vegetation dynamics can create a different fuel loading at the edge relative to the interior. We used the computer simulation model FARSITE to examine a fuel edge structural feature with scenarios from three levels of edge fuel loading to determine what impacts fuels in AEI have on fire spread by ranking all of the landscape scenarios. The mean burned area (ha) was significantly different among the landscapes after seven days. Fire spread increased by 38% with a high fuel loading assigned to the designated edge structure; while it decreased by 20% with medium edge fuel loading and 44% with low edge fuel loading. The landscape without edge structure (i.e., control) produced burned areas between the medium and the high edge fuel loading scenarios. The daily rate of fire spread was also significantly affected by edge fuel loading. We encourage model users to include edge fuel in FARSITE fuel maps of highly fragmented forests. This study suggests that with on the ground fuel treatments, AEI can be manipulated to change the spread potential of large fires.