Neutral models for testing landscape hypotheses

Neutral landscape models were originally developed to test the hypothesis that human-induced fragmentation produces patterns distinctly different from those associated with random processes. Other uses for neutral models have become apparent, including the development and testing of landscape metrics to characterize landscape pattern. Although metric development proved to be significant, the focus on metrics obscured the need for iterative hypothesis testing fundamental to the advancement of the discipline. We present here an example of an alternative neutral model and hypothesis designed to relate the process of landscape change to observed landscape patterns. The methods and program, QRULE, are described and options for statistical testing outlined. The results show that human fragmentation of landscapes results in a non-random association of land-cover types that can be describe by simple statistical methods. Options for additional landscape studies are discussed and access to QRULE described in the hope that these methods will be employed to advance our understanding of the processes that affect the structure and function in human dominated landscapes.     

Edge effects on fire spread in a disturbed Northern Wisconsin landscape

The effect of area-of-edge influence (AEI) on fire size and movement was simulated by considering the distribution of single and multiple edges in the Chequamegon-Nicolet National Forest in Northern Wisconsin, USA. Six hypothetical landscapes with different delineations of AEIs were created for simulating fire spread using FARSITE to evaluate the influence of edges on the rate and direction of fire spread. The burned area differed significantly among the six landscapes. In the three scenarios with buffered edges, the burned area increased by 35% with high loading fuel in AEIs, while it decreased by 21 and 46% with medium and low fuel loading in the AEIs, respectively, as compared to the no edge scenario. In two scenarios we delineated the area-of-multiple-edge influence (AMEI) and placed more than one high loading fuel within it. This increased the burned area by 5% from the high buffered edge scenario and by 40% from the control. When the depth-of-edge influence (DEI) was doubled to 60 m using AMEI with high fuels, the burned area increased by 20% from the high buffered edge scenario and by 60% from the control. We found that low and medium fuel loading slowed the fire spread and over time, caused the fire front to change direction. In high fuel loading scenarios, AEIs acted as corridors facilitating the fire spread by providing a contiguous patch of fuel which allowed fires to increase in size and pulled the fire front in the same direction.     

Validating landtype associations using forest inventory and analysis data and neutral landscape models

One of the principal uses of land type associations (LTAs) is to provide information on ecological patterns and potentials useful for identifying alternatives and setting vegetation management objectives at landscape and watershed scales. Since LTA identification and delineation is an iterative process involving subjective decisions, it is imperative that LTAs accurately capture measurable distinctions in vegetation. In this study, we provided a framework for validating LTAs in Missouri’s. We chose a suite of variables from the Forest Inventory and Analysis database of the US Forest Service, an independent source, to capture a wide range of characteristics of the forest ecosystem. These variables included forest type, species composition, species diversity, species richness, site index, density of all trees 1 inch dbh and greater, and density of all trees 5 inch dbh and greater. First, the appropriateness of grouping LTAs into broader LTA types (identifications) was examined, and results suggest that species composition is more similar within LTA Type than among LTA Types. Second, a neutral model approach was used to evaluate the variables used in the study, and it was shown that forest type and the two density measures validated LTA adjacencies no more often than randomly delineated LTAs. Therefore, these three variables were removed from the analysis. Finally, comparisons of adjacent LTAs using statistical analysis of remaining variables resulted in the validation of 480 of the 623 compared adjacencies, showing a significant difference in at least one variable. Results of this study provide a quantitative measure of Missouri LTAs and reveal LTAs that have less distinction and need further refinements.     

Characterizing historical and modern fire regimes in Michigan (USA): A landscape ecosystem approach

We studied the relationships of landscape ecosystems to historical and contemporary fire regimes across 4.3 million hectares in northern lower Michigan (USA). Changes in fire regimes were documented by comparing historical fire rotations in different landscape ecosystems to those occurring between 1985 and 2000. Previously published data and a synthesis of the literature were used to identify six forest-replacement fire regime categories with fire rotations ranging from very short (<100 years) to very long (>1,000 years). We derived spatially-explicit estimates of the susceptibility of landscape ecosystems to fire disturbance using Landtype Association maps as initial units of investigation. Each Landtype Association polygon was assigned to a fire regime category based on associations of ecological factors known to influence fire regimes. Spatial statistics were used to interpolate fire points recorded by the General Land Office. Historical fire rotations were determined by calculating the area burned for each category of fire regime and dividing this area by fifteen (years) to estimate area burned per annum. Modern fire rotations were estimated using data on fire location and size obtained from federal and state agencies. Landtype Associations networked into fire regime categories exhibited differences in both historical and modern fire rotations. Historical rotations varied by 23-fold across all fire rotation categories, and modern forest fire rotations by 13-fold. Modern fire rotations were an order of magnitude longer than historical rotations. The magnitude of these changes has important implications for forest health and understanding of ecological processes in most of the fire rotation categories that we identified.This revised version was published online in May 2005 with corrections to the Cover Date.     

Determining landscape extent for succession and disturbance simulation modeling

Dividing regions into manageable landscape units presents special problems in landscape ecology and land management. Ideally, a landscape should be large enough to capture a broad range of vegetation, environmental and disturbance dynamics, but small enough to be useful for focused management objectives. The purpose of this study was to determine the optimal landscape size to summarize ecological processes for two large land areas in the southwestern United States. We used a vegetation and disturbance dynamics model, LANDSUMv4, to simulate a set of nine scenarios involving systematically varied topography, map resolution, and model parameterizations of fire size and fire frequency. Spatial input data were supplied by the LANDscape FIRE Management Planning System (LANDFIRE) prototype project, an effort that will provide comprehensive and scientifically credible mid-scale data to support the National Fire Plan. We analyzed output from 2,000 year simulations to determine the thresholds of landscape condition based on the variability of burned area and dominant vegetation coverage. Results show that optimal landscape extent using burned area variability is approximately 100 km² depending on topography, map resolution, and model parameterization. Variability of dominant vegetation area is generally higher and the optimal landscape sizes are larger in comparison to those features determined from burned area. Using the LANDFIRE project as a case study, we determined landscape size and map resolution for a large mapping project, and showed that optimal landscape size depends upon geographical, ecological, and management context. This paper was written and prepared by U.S. Government employees on official time, and therefore is in the public domain and not subject to copyright. The use of trade or firm names in this paper is for reader information and does not imply endorsement by the U.S. Department of Agriculture of any product or service.     

Behavior of class-level landscape metrics across gradients of class aggregation and area

Habitat loss and fragmentation processes strongly affect biodiversity conservation in landscapes undergoing anthropogenic land use changes. Many attempts have been made to use landscape structure metrics to quantify the independent and joint effects of these processes. Unfortunately, ecological interpretation of those metrics has been plagued by lack of thorough understanding of their theoretical behavior. We explored behavior of 50 metrics in neutral landscapes across a 21-step gradient in aggregation and a 19-step gradient in area using a full factorial design with 100 replicates of each of the 399 combinations of the two factors to assess how well metrics reflected changes in landscape structure. Metric values from real landscapes were used to determine the extent of neutral landscape space that is represented in real landscapes. We grouped metrics into three major behavioral classes: strongly related to focal class area (n=15), strongly related to aggregation (n=7), and jointly responding to area and aggregation (n=28). Metrics strongly related to class area exhibited a variety of distinct behaviors, and many of these metrics have unique interpretations that make each of them particularly useful in certain applications. Metrics strongly related to aggregation, independent of class area, are particularly useful in assessing effects of fragmentation. Moreover, metrics in this group exhibited a range of specific behaviors, highlighting subtle but different aspects of landscape aggregation even though we controlled only one aspect of aggregation. The non-linear behavior exhibited by many metrics renders interpretation difficult and use of linear analytical techniques inappropriate under many circumstances. Ultimately, comprehensive characterization of landscapes undergoing habitat loss and fragmentation will require using several metrics distributed across behavioral groups.This revised version was published online in May 2005 with corrections to the Cover Date.     

Effects of site,landscape features,and fire regime on vegetation patterns in presettlement southern Wisconsin

The presettlement tree cover (1831–33) of 3 townships in a southern Wisconsin landscape was analyzed using original survey records. Four forest types were identified: closed forest, open forest, savanna, and prairie. Comparisons of vegetation types and landscape pattern were made between the east and west sides of the Pecatonica River, which bisects the landscape and could have acted as a natural fire barrier. West of the river, presettlement tree species richness and diversity were lower and trees were smaller in diameter and less dense than to the east. The major vegetation types to the west were prairie (42% of landscape) and savanna (40%), both fire-susceptible types. Prairie was more common on gentle slopes than on other landforms. To the east, the landscape was 70% forested (closed plus open forest). Here, prairie was more frequent on steep dry sites. These vegetation differences, including the contrasting landscape placement of prairie, are attributed to distinct site characteristics and to disturbance (fire) regimes, with the west likely having more frequent fires. In terms of the four vegetation types, the east landscape was more homogeneous, being dominated by closed forest (50%). West of the Pecatonica River, the landscape was more heterogeneous because of the high proportion of both prairie and savanna; however, in terms of flammability of vegetation, the west was essentially homogeneous (82% prairie plus savanna).     

Wildfires and landscape patterns in the Eastern Iberian Peninsula

The relations between disturbance regime and landscape patterns have been developed from a theoretical perspective, but few studies have tested these relations when forces promoting opposing heterogeneity patterns are simultaneously operating on a landscape. This work provides quantitative evidence of these relations in areas dominated by human activity, showing that landscape heterogeneity decreases disturbance spread. In turn, disturbance introduces a source of landscape heterogeneity, but it is not enough to counterbalance the homogeneity trend due to agricultural abandonment. Land cover changes and wildfire occurrence (fires larger than 0.3 km²) have been monitored in the Tivissa municipality (208.4 km²) (Catalonia, NE Spain) from 1956 to 1993. Land cover maps were obtained from 1956, 1978 and 1993 and they were overlaid with fire occurrence maps obtained for the 1975–1995 period from 60 m resolution remote sensing images, which allow the identification of burned areas by sudden drops in Normalized Difference Vegetation Index (NDVI). Changes in landscape patterns in relation to fire regime have been analyzed considering several parameters: patch density, mean patch size, mean distance to the nearest neighbour of the same category, edge density, and the Shannon diversity index. In the 1956–1993 period there is a trend to increasing landscape homogenization due to the expansion of shrub­lands linked to a decrease in forest surface, and to the abandonment of agricultural lands. This trend, however, is not constant along all the period. Fires are more likely to occur in woody, homogenous areas, increasing landscape heterogeneity, as observed in the 1978–1993 period. This increase in heterogeneity does not counterbalance the general trend to landscape homogenization as a consequence of agricultural abandonment and the coalescence of natural vegetation patches.This revised version was published online in May 2005 with corrections to the Cover Date.     

Environmental,historical, and contextual determinants of vegetation cover: a landscape perspective

We formulated and tested models of relationships among determinants of vegetation cover in two agroforested landscapes of eastern North America (Haut Saint-Laurent, Quebec, Canada) that differed by the spatial arrangement of their geomorphic features and intensity of agricultural activities. Our landscape model compared the woody plots of each landscape in terms of the relative influence of environmental attributes, land use history (1958 – 1997), and spatial context (i.e., proximity of similar or contrasting land cover). Our vegetation model evaluated the relative contribution of the same sets of variables to the distributions of herbs, trees, and shrubs. Relationships were assessed using partial Mantel tests and path analyses. Significant environmental and contextual differences were found between the vegetation plots of the two landscapes, but disturbance history was similar. Our vegetation model confirms the dominant effect of historical factors on vegetation patterns. Whereas land-use history overrides environmental and contextual control for trees, herbaceous and shrub species are more sensitive to environmental conditions. Context is determinant only for understory species in older, less-disturbed plots. Results are discussed in relevance to vegetation dynamics in a landscape perspective that integrates interactions between environmental and human influences.     

The effects of landscape composition and physiognomy on metapopulation size: the role of corridors

We develop and analyze a model that examines the effects of corridor quality, quantity, and arrangement on metapopulation sizes. These ideas were formerly investigated by Lefkovitch and Fahrig (1985) and Henein and Merriam (1990). Our simulations provide results similar to the Henein and Merriam model, indicating that the quality of corridors in a landscape and their arrangement will influence the size of a metapopulation. We then go one step further, describing how corridor arrangement alters the metapopulation, and provide a method for predicting which corridor arrangements should support larger metapopulations. In contrast to the Henein and Merriam model, we find that the number of corridor connections has no influence on the size of a metapopulation in a landscape unless there is an accompanying change in the uniformity of the distribution of corridor connections among patches.     

Pocono mesic till barrens in retreat: topography,fire and forest contagion effects

The Pocono mesic till barrens (PMTB) are a unique assemblage of fire-maintained shrub communities that support numerous rare species. Historically these barrens covered a large area in the vicinity of Long Pond, Pennsylvania, USA. However, due largely to regional fire suppression instituted in the early 1960s, over 70% of the area covered by barrens succeeded to fire-intolerant forest that does not support the rare species. We investigated the influence of forest proximity on barrens succession across three geomorphic types during periods of high fire frequency and fire suppression, testing the hypothesis that forest processes such as seed rain, shading, and detrital enrichment of soils enhances barrens succession through a contagion effect. Evidence of a forest contagion effect should be shown by increased rates of barrens succession with increasing proximity to the nearest forest edge. In order to detect a forest contagion effect, barrens persistence and barrens succession were modeled in proximity zones of 0-50 m, 50-100 m, 100-200 m, and greater than 200 m from the nearest forest edge. We used existing GIS data layers for fire, geomorphology, and vegetation distribution in 1938, 1963, and 1992. The layers were modified and overlain using ArcView software to determine persistence and succession rates for each unique combination of layers in each proximity zone from 1938 to 1963 (pre-fire suppression) and 1963 to 1992 (post-fire suppression). ANCOVA results indicate that proximity to the nearest forest edge significantly affected barrens persistence rates in both time periods, but succession rates were significantly affected in 1938 to 1963 only. Twenty-eight percent of the 1938 barrens succeeded to forest by 1963; 56% of the 1963 barrens became forest by 1992. Results support previous findings that barrens persistence is enhanced by increased fire frequency, and that barrens persist longer where they overlie flat glacial till than on other geomorphology types.     

Influence of forest management alternatives and land type on susceptibility to fire in northern Wisconsin,USA

We used the LANDIS disturbance and succession model to study the effects of six alternative vegetation management scenarios on forest succession and the subsequent risk of canopy fire on a 2791 km² landscape in northern Wisconsin, USA. The study area is a mix of fire-prone and fire-resistant land types. The alternatives vary the spatial distribution of vegetation management activities to meet objectives primarily related to forest composition and recreation. The model simulates the spatial dynamics of differential reproduction, dispersal, and succession patterns using the vital attributes of species as they are influenced by the abiotic environment and disturbance. We simulated 50 replicates of each management alternative and recorded the presence of species age cohorts capable of sustaining canopy fire and the occurrence of fire over 250 years. We combined these maps of fuel and fire to map the probability of canopy fires across replicates for each alternative. Canopy fire probability varied considerably by land type. There was also a subtle, but significant effect of management alternative, and there was a significant interaction between land type and management alternative. The species associated with high-risk fuels (conifers) tend to be favored by management alternatives with more disturbances, whereas low disturbance levels favor low-risk northern hardwood systems dominated by sugar maple. The effect of management alternative on fire risk to individual human communities was not consistent across the landscape. Our results highlight the value of the LANDIS model for identifying specific locations where interacting factors of land type and management strategy increase fire risk.This revised version was published online in May 2005 with corrections to the Cover Date.     

Simulating fire frequency and urban growth in southern California coastal shrublands,USA

Fire is an important natural disturbance in the Mediterranean-climate coastal shrublands of southern California. However, anthropogenic ignitions have increased fire frequency to the point that it threatens the persistence of some shrub species and favors the expansion of exotic annual grasses. Because human settlement is a primary driver of increased ignitions, we integrated a landscape model of disturbance and succession (LANDIS) with an urban growth model (UGM) to simulate the combined effects of urban development and high fire frequency on the distribution of coastal shrublands. We tested whether urban development would contribute to an expansion of the wildland-urban interface (WUI) and/or change in average fire return intervals and compared the relative impacts of direct habitat loss and altered fire regimes on functional vegetation types. We also evaluated two methods of integrating the simulation models. The development pattern predicted by the UGM was predominantly aggregated, which minimized the expansion of the WUI and increase in fire frequency, suggesting that fire risk may be higher at intermediate levels of urbanization due to the spatial arrangement of ignition sources and fuel. The comparison of model coupling methods illustrated how cumulative effects of repeated fires may occur gradually as urban development expands across the landscape. Coastal sage scrub species and resprouting chaparral were more susceptible to direct habitat loss, but increased fire frequency was more of a concern to obligate seeder species that germinate from a persistent seed bank. Simulating different scenarios of fire frequency and urban growth within one modeling framework can help managers locate areas of highest risk and determine which vegetation types are most vulnerable to direct habitat loss, altered fire regimes, or both.     

Effects of fire frequency on plant communities and landscape pattern in the Massif des Aspres (southern France)

Fire frequency can affect pattern and diversity in plant communities and landscapes. We had the opportunity to study changes due to recurring wildfires on the same sites over a period of 50 years in the Massif des Aspres (southern France). The study was carried out in areas occupied byQuercus suber andQ. ilex series. A comparison of historical and cartographical documents (vegetation maps covering a 50 year interval and an accurate map of major wildfires during this period) allowed us to determine the changes occurring over time with or without fire action. Plant communities were grouped into three main vegetation types: forests, treed shrublands and shrublands. The passage of three successive wildfires on the same site led to a decrease in forest areas and an increase in shrublands; however, shrublands were already present before the first fire of the period under consideration. Less frequent fire occurrence induced more complex heterogeneity and greater landscape diversity. In the study region as a whole, with or without fire action, a significant decrease in forest surfaces was recorded, whereas there was an increase of unforested communities such as treed shrublands and shrublands. In some parts of the Massif fires increased the homogeneity of the landscape, in other parts they created a greater heterogeneity and diversity of plant communities.     

Detecting dominant landscape objects through multiple scales: An integration of object-specific methods and watershed segmentation

Complex systems, such as landscapes, are composed of different critical levels of organization where interactions are stronger within levels than among levels, and where each level operates at relatively distinct time and spatial scales. To detect significant features occurring at specific levels of organization in a landscape, two steps are required. First, a multiscale dataset must be generated from which these features can emerge. Second, a procedure must be developed to delineate individual image-objects and identify them as they change through scale. In this paper, we introduce a framework for the automatic definition of multiscale landscape features using object-specific techniques and marker-controlled watershed segmentation. By applying this framework to a high-resolution satellite scene, image-objects of varying size and shape can be delineated and studied individually at their characteristic scale of expression. This framework involves three main steps: 1) multiscale dataset generation using an object-specific analysis and upscaling technique, 2) marker-controlled watershed transformation to automatically delineate individual image-objects as they evolve through scale, and 3) landscape feature identification to assess the significance of these image-objects in terms of meaningful landscape features. This study was conducted on an agro-forested region in southwest Quebec, Canada, using IKONOS satellite data. Results show that image-objects tend to persist within one or two scale domains, and then suddenly disappear at the next, while new image-objects emerge at coarser scale domains. We suggest that these patterns are associated to sudden shifts in the entire image structure at certain scale domains, which may correspond to critical landscape thresholds.This revised version was published online in May 2005 with corrections to the Cover Date.     

Interplay between physical and predator landscapes affects transferability of shorebird distribution models

Coastal landscapes with extensive intertidal mudflats provide non-breeding habitat for Arctic shorebirds. Few attempts have been made to develop and test landscape-level models predicting the intertidal distribution of these birds. We modelled the distribution of a Holarctic species, Dunlin (Calidris alpina), at a hemispherically important non-breeding site, the Fraser River Delta, British Columbia, Canada, in seasons with different predator landscapes. We trained the models during a season when nocturnal predators were common and tested temporal transferability of the models on independent datasets when nocturnal predators were absent. Snowy Owls (Nyctea scandiaca) influenced Dunlin distribution and thus model transferability. After accounting for their presence, models displayed good to excellent discrimination, i.e. prediction of the instantaneous and cumulative (over low tide period) probability of mudflat use by Dunlin, in fore- and backcasting applications. Model calibration was good or else, where over-prediction was observed, the reason for the bias was identified. The distribution models may predict mudflat use by Dunlin and possibly related species given relevant data describing the intertidal landscape. The models are amenable to GIS application, describe the amount of use per hectare of the intertidal zone and can be used to determine and visualise relative and absolute suitability of intertidal areas. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

A controlled,hierarchical study of habitat fragmentation: responses at the individual,patch, and landscape scale

We compared the performance of individuals and whole populations of meadow voles, Microtus pennsylvanicus, within and between experimentally created habitat fragments of three sizes (1.0, 0.25, and 0.0625 ha) and between a 20-ha fragmented and a 20-ha continuous habitat landscape. We recorded 10,020 captures of 3946 individuals over 17 censuses between June 1993 and October 1994. Five demographic parameters showed significantly different population responses between the two landscapes but no difference in tests comparing fragment size: i.e., mean and peak population densities (the latter, in each of the two growing seasons) averaged 149 to 172% higher, population growth rate averaged 219% higher, and adult recruitment 170% higher in fragmented than in the continuous control landscape. Observations at the individual level (body sizes, rates of reproduction, residence times) suggested that these landscape differences involved enhanced performance of adult females associated with edge habitats rather than differential immigration or emigration. If this turns out to be a common response to fragmentation, the detection of such responses will be greater when comparing fragmented and unfragmented landscapes with qualitatively different structure than for fragments of varied size with differing proportions of edge. That responses to habitat fragmentation may be more evident at the very small (individual) and very large (landscape) scales, but may be obscured at the intermediate spatial scale of fragments, is a proposition that clearly requires more attention.