Robert H. Gardner

Indices of landscape pattern

Landscape ecology deals with the patterning of ecosystems in space. Methods are needed to quantify aspects of spatial pattern that can be correlated with ecological processes. The present paper develops three indices of pattern derived from information theory and fractal geometry. Using digitized maps, the indices are calculated for 94 quadrangles covering most of the eastern United States. The indices are shown to be reasonably independent of each other and to capture major features of landscape pattern. One of the indices, the fractal dimension, is shown to be correlated with the degree of human manipulation of the landscape.

Effects of changing spatial scale on the analysis of landscape pattern

The purpose of this study was to observe the effects of changing the grain (the first level of spatial resolution possible with a given data set) and extent (the total area of the study) of landscape data on observed spatial patterns and to identify some general rules for comparing measures obtained at different scales. Simple random maps, maps with contagion (i.e., clusters of the same land cover type), and actual landscape data from USGS land use (LUDA) data maps were used in the analyses. Landscape patterns were compared using indices measuring diversity (H), dominance (D) and contagion (C). Rare land cover types were lost as grain became coarser. This loss could be predicted analytically for random maps with two land cover types, and it was observed in actual landscapes as grain was increased experimentally. However, the rate of loss was influenced by the spatial pattern. Land cover types that were clumped disappeared slowly or were retained with increasing grain, whereas cover types that were dispersed were lost rapidly. The diversity index decreased linearly with increasing grain size, but dominance and contagion did not show a linear relationship. The indices D and C increased with increasing extent, but H exhibited a variable response. The indices were sensitive to the number (m) of cover types observed in the data set and the fraction of the landscape occupied by each cover type (Pk); both m and Pkvaried with grain and extent. Qualitative and quantitative changes in measurements across spatial scales will differ depending on how scale is defined. Characterizing the relationships between ecological measurements and the grain or extent of the data may make it possible to predict or correct for the loss of information with changes in spatial scale.

Neutral models for the analysis of broad-scale landscape pattern

The relationship between a landscape process and observed patterns can be rigorously tested only if the expected pattern in the absence of the process is known. We used methods derived from percolation theory to construct neutral landscape models,i.e., models lacking effects due to topography, contagion, disturbance history, and related ecological processes. This paper analyzes the patterns generated by these models, and compares the results with observed landscape patterns. The analysis shows that number, size, and shape of patches changes as a function of p, the fraction of the landscape occupied by the habitat type of interest, and m, the linear dimension of the map. The adaptation of percolation theory to finite scales provides a baseline for statistical comparison with landscape data. When USGS land use data (LUDA) maps are compared to random maps produced by percolation models, significant differences in the number, size distribution, and the area/perimeter (fractal dimension) indices of patches were found. These results make it possible to define the appropriate scales at which disturbance and landscape processes interact to affect landscape patterns.

Landscape patterns in a disturbed environment

Deciduous forest patterns were evaluated, using fractal analysis, in the U. S. Geological Survey 1: 250,000 Natchez Quadrangle, a region that has experienced relatively recent conversion of forest cover to cropland. A perimeter-area method was used to determine the fractal dimension; the results show a different dimension for small compared with large forest patches. This result is probably related to differences in the scale of human versus natural processes that affect this particular forest pattern. By identifying transition zones in the scale at which landscape patterns change this technique shows promise for use in developing hypotheses related to scale-dependent processes and as a simple metric to evaluate changes on the earths surface using remotely sensed data.

EFFECTS OF FIRE SIZE AND PATTERN ON EARLY SUCCESSION IN YELLOWSTONE NATIONAL PARK

The Yellowstone fires of 1988 affected >250 000 ha, creating a mosaic of burn severities across the landscape and providing an ideal opportunity to study effects of fire size and pattern on postfire succession. We asked whether vegetation responses differed between small and large burned patches within the fire-created mosaic in Yellowstone National Park (YNP) and evaluated the influence of spatial patterning on the postfire veg- etation. Living vegetation in a small (1 ha), moderate (70-200 ha), and large (500-3600 ha) burned patch at each of three geographic locations was sampled annually from 1990 to 1993. Burn severity and patch size had significant effects on most biotic responses. Severely burned areas had higher cover and density of lodgepole pine seedlings, greater abundance of opportunistic species, and lower richness of vascular plant species than less severely burned areas. Larger burned patches had higher cover of tree seedlings and shrubs, greater densities of lodgepole pine seedlings and opportunistic species, and lower species richness than smaller patches. Herbaceous species present before the fires responded in- dividually to burn severity and patch size; some were more abundant in large patches or severely burned areas, while others were more abundant in small patches or lightly burned areas. To date, dispersal into the burned areas from the surrounding unburned forest has not been an important mechanism for reestablishment of forest species. Most plant cover in burned areas consisted of resprouting survivors during the first 3 yr after the fires. A pulse of seedling establishment in 1991 suggested that local dispersal from these survivors was a dominant mechanism for reestablishment of forest herbs. Succession across much of YNP appeared to be moving toward plant communities similar to those that burned in 1988, primarily because extensive biotic residuals persisted even within very large burned areas. However, forest reestablishment remained questionable in areas of old (>400 yr) forests with low prefire serotiny. Despite significant effects of burn severity and patch size, the most important explanatory variable for most biotic responses was geographic location, particularly as related to broad-scale patterns of serotiny in Pinus contorta. We conclude that the effects of fire size and pattern were important and some may be persistent, but that these landscape-scale effects occurred within an overriding context of broader scale gra- dients.

The effect of landscape heterogeneity on the probability of patch colonization

The effect of landscape heterogeneity on the dispersal of organisms between habitat islands is poorly understood. Preferred pathways for dispersal (i.e., corridors), as well as dispersal barriers, are difficult to identify when the landscape matrix is composed of a complex mixture of land cover types. We developed an individual-based dispersal model to measure immigration and emigration rates between habitat islands within hetero- geneous landscapes. Dispersing individuals of a model organism were simulated as self- avoiding random walkers (SAW) traversing a digital land cover map, with each habitat type assigned a priori a probability that the SAW would enter that habitat type. Each individual began the dispersal process on a random site at the edge of a deciduous forest patch and was allowed to move until it reached a different deciduous forest patch. Visu- alization of the movement patterns across the landscape was achieved by tabulating the frequency of visitation Qf successful dispersers to each grid cell on the map. The model was used to estimate the probabilities of disperser transfer between patches by varying the a priori probabilities of movement into each habitat type in order to: (1) estimate the effect of changing landscape heterogeneity on the transfer probabilities, and (2) visualize dispersal corridors and barriers as perceived by model organisms operating by specific movement rules and at specific scales. The results show that 89% of the variability in dispersal success can be accounted for by differences in the size and isolation of forest patches, with closer and larger patches having significantly greater exchange of dispersing organisms. However, changes in the heterogeneity of the landscape matrix could significantly enhance or decrease emigration success from an individual patch, depending on the landscape. Changes in emigration success from an individual patch resulting from changes in matrix heterogeneity were not predictable, and transfer rates between patches were not symmetrical due to differences in the proximity of neighboring patches, and differences in the funneling at- tributes of certain landscape patterns. Visualizations showed that corridors are often diffuse and difficult to identify from structural features of the landscape. A wide range of organisms with differing movement capabilities can be simulated using the approach presented to increase our understanding of how landscape structure affects organism dispersal.

Effects of fire on landscape heterogeneity in Yellowstone National Park, Wyoming

A map of burn severity resulting from the 1988 fires that occurred in Yellowstone National Park (YNP) was de- rived from Landsat Thematic Mapper (TM) imagery and used to assess the isolation of burned areas, the heterogeneity that resulted from fires burning under moderate and severe burning conditions, and the relationship between heterogeneity and fire size. The majority of severely burned areas were within close proximity (50 to 200 m) to unburned or lightly burned areas, suggesting that few burned sites are very far from potential sources of propagules for plant reestablishment. Fires that occurred under moderate burning conditions early during the 1988 fire season resulted in a lower proportion of crown fire than fires that occurred under severe burning condi- tions later in the season. Increased dominance and contagion of burn severity classes and a decrease in the edge: area ratio for later fires indicated a slightly more aggregated burn pattern compared to early fires. The proportion of burned area in different burn severity classes varied as a function of daily fire size. When daily area burned was relatively low, the propor- tion of burned area in each burn severity class varied widely. When daily burned area exceeded 1250 ha, the burned area contained about 50 % crown fire, 30 % severe surface burn, and 20 % light surface burn. Understanding the effect of fire on landscape heterogeneity is important because the kinds, amounts, and spatial distribution of burned and unburned areas may influence the reestablishment of plant species on burned sites.

Lacunarity indices as measures of landscape texture

Lacunarity analysis is a multi-scaled method of determining the texture associated with patterns of spatial dispersion (i.e., habitat types or species locations) for one-, two-, and three-dimensional data. Lacunarity provides a parsimonious analysis of the overall fraction of a map or transect covered by the attribute of interest, the degree of contagion, the presence of self-similarity, the presence and scale of randomness, and the existence of hierarchical structure. For self-similar patterns, it can be used to determine the fractal dimension. The method is easily implemented on the computer and provides readily interpretable graphic results. Differences in pattern can be detected even among very sparsely occupied maps.

Predicting across scales: Theory development and testing

Landscape ecologists deal with processes that occur at a variety of temporal and spatial scales. The ability to make predictions at more than one level of resolution requires identification of the processes of interest and parameters that affect this process at different scales, the development of rules to translate information across scales, and the ability to test these predictions at the relevant spatial and temporal scales. This paper synthesizes discussions from a workshop on ‘Predicting Across Scales: Theory Development and Testing’ that was held to discuss current research on scaling and to identify key research issues.

Predicting the spread of disturbance across heterogeneous landscapes

The expected pattern of disturbance propagation across a landscape was studied by using simple landscape models derived from percolation theory. The spread of disturbance was simulated as a function of the proportion of the landscape occupied by the disturbance-prone habitat and the frequency (probability of initiation) and intensity (probability of spread) of the habitat-specific disturbance. Disturbance effects were estimated from the proportion of habitat affected by the disturbance and changes in landscape structure (i.e., spatial patterns). Landscape structure was measured by the number of habitat clusters, the size and shape of the largest cluster, and the amount of edge in the landscape. Susceptible habitats that occupied less than 50% of the landscape were sensitive to disturbance frequency but showed little response to changes in disturbance intensity. Susceptible habitat that occupied more than 60% of the landscape were sensitive to disturbance intensity and less sensitive to disturbance frequency. These dominant habitats were also very easily fragmented by disturbances of moderate intensity and low frequency. Implications of these results for the management of disturbance-prone landscapes are discussed. The propagation of disturbance in heterogeneous landscapes depends on the structure of the landscape as well as the disturbance intensity and frequency.