Normalized burn ratios link fire severity with patterns of avian occurrence

Context

Remotely sensed differenced normalized burn ratios (DNBR) provide an index of fire severity across the footprint of a fire. We asked whether this index was useful for explaining patterns of bird occurrence within fire adapted xeric pine-oak forests of the southern Appalachian Mountains.

Objectives

We evaluated the use of DNBR indices for linking ecosystem process with patterns of bird occurrence. We compared field-based and remotely sensed fire severity indices and used each to develop occupancy models for six bird species to identify patterns of bird occurrence following fire.

Methods

We identified and sampled 228 points within fires that recently burned within Great Smoky Mountains National Park. We performed avian point counts and field-assessed fire severity at each bird census point. We also used Landsat? imagery acquired before and after each fire to quantify fire severity using DNBR. We used non-parametric methods to quantify agreement between fire severity indices, and evaluated single season occupancy models incorporating fire severity summarized at different spatial scales.

Results

Agreement between field-derived and remotely sensed measures of fire severity was influenced by vegetation type. Although occurrence models using field-derived indices of fire severity outperformed those using DNBR, summarizing DNBR at multiple spatial scales provided additional insights into patterns of occurrence associated with different sized patches of high severity fire.

Conclusions

DNBR is useful for linking the effects of fire severity to patterns of bird occurrence, and informing how high severity fire shapes patterns of bird species occurrence on the landscape.

     

Influence of high-resolution data on the assessment of forest fragmentation

Context

Remote sensing has been a foundation of landscape ecology. The spatial resolution (pixel size) of remotely sensed land cover products has improved since the introduction of landscape ecology in the United States. Because patterns depend on spatial resolution, emerging improvements in the spatial resolution of land cover may lead to new insights about the scaling of landscape patterns.

Objective

We compared forest fragmentation measures derived from very high resolution (1 m²) data with the same measures derived from the commonly used (30 m?×??30 m; 900 m²) Landsat-based data.

Methods

We applied area-density scaling to binary (forest; non-forest) maps for both sources to derive source-specific estimates of dominant (density ≥?60%), interior (≥?90%), and intact (100%) forest.

Results

Switching from low- to high-resolution data produced statistical and geographic shifts in forest spatial patterns. Forest and non-forest features that were “invisible” at low resolution but identifiable at high resolution resulted in higher estimates of dominant and interior forest but lower estimates of intact forest from the high-resolution source. Overall, the high-resolution data detected more forest that was more contagiously distributed even at larger spatial scales.

Conclusion

We anticipate that improvements in the spatial resolution of remotely sensed land cover products will advance landscape ecology through re-interpretations of patterns and scaling, by fostering new landscape pattern measurements, and by testing new spatial pattern-ecological process hypotheses.

     

Evaluating static and dynamic landscape connectivity modelling using a 25-year remote sensing time series

Context

Despite calls for landscape connectivity research to account for spatiotemporal dynamics, studies have overwhelmingly evaluated the importance of habitats for connectivity at single or limited moments in time. Remote sensing time series represent a promising resource for studying connectivity within dynamic ecosystems. However, there is a critical need to assess how static and dynamic landscape connectivity modelling approaches compare for prioritising habitats for conservation within dynamic environments.

Objectives

To assess whether static landscape connectivity analyses can identify similar important areas for connectivity as analyses based on dynamic remotely sensed time series data.

Methods

We compared degree and betweenness centrality graph theory metric distributions from four static scenarios against equivalent results from a dynamic 25-year remotely sensed surface-water time series. Metrics were compared at multiple spatial aggregation scales across south-eastern Australia’s 1 million km² semi-arid Murray–Darling Basin and three sub-regions with varying levels of hydroclimatic variability and development.

Results

We revealed large differences between static and dynamic connectivity metric distributions that varied by static scenario, region, spatial scale and hydroclimatic conditions. Static and dynamic metrics showed particularly low overlap within unregulated and spatiotemporally variable regions, although similarities increased at coarse aggregation scales.

Conclusions

In regions that exhibit high spatiotemporal variability, static connectivity modelling approaches are unlikely to serve as effective surrogates for more data intensive approaches based on dynamic, remotely sensed data. Although this limitation may be moderated by spatially aggregating static connectivity outputs, our results highlight the value of remotely sensed time series for assessing connectivity in dynamic landscapes.

     

Effects of interactive scale-dependent variables on beetle diversity patterns in a semi-arid agricultural landscape

Understanding species-diversity patterns in heterogeneous landscapes invites comprehensive research on how scale-dependent processes interact across scales. We used two common beetle families (Tenebrionidae, detrivores; Carabidae, predators) to conduct such a study in the heterogeneous semi-arid landscape of the Southern Judean Lowland (SJL) of Israel, currently undergoing intensive fragmentation. Beetles were censused in 25 different-sized patches (500–40,000 m²). We used Fisher’s α and non-parametric extrapolators to estimate species diversity from 11,125 individuals belonging to 56 species. Patch characteristics (plant species diversity and cover, soil cover and degree of stoniness) were measured by field transects. Spatial variables (patch size, shape, physiognomy and connectivity) and landscape characteristics were analyzed by GIS and remote-sensing applications. Both patch-scale and landscape-scale variables affected beetle species diversity. Path-analysis models showed that landscape-scale variables had the strongest effect on carabid diversity in all patches. The tenebrionids responded differently: both patch-scale and landscape-scale variables affected species diversity in small patches, while mainly patch-scale variables affected species diversity in large patches. Most of the paths affected species diversity both directly and indirectly, combining the effects of both patch-scale and landscape-scale variables. These results match the biology of the two beetle families: Tenebrionidae, the less mobile and more site-attached family, responded to the environment in a fine-grained manner, while the highly dispersed Carabidae responded to the environment in a coarse-grained manner. We suggest that understanding abiotic and biotic variable interactions across scales has important consequences for our knowledge of community structure and species diversity patterns at large spatial scales.     

The feasibility of remotely sensed data to estimate urban tree dimensions and biomass

Accurately measuring the biophysical dimensions of urban trees, such as crown diameter, stem diameter, height, and biomass, is essential for quantifying their collective benefits as an urban forest. However, the cost of directly measuring thousands or millions of individual trees through field surveys can be prohibitive. Supplementing field surveys with remotely sensed data can reduce costs if measurements derived from remotely sensed data are accurate. This study identifies and measures the errors incurred in estimating key tree dimensions from two types of remotely sensed data: high-resolution aerial imagery and LiDAR (Light Detection and Ranging). Using Sacramento, CA, as the study site, we obtained field-measured dimensions of 20 predominant species of street trees, including 30–60 randomly selected trees of each species. For each of the 802 trees crown diameter was estimated from the aerial photo and compared with the field-measured crown diameter. Three curve-fitting equations were tested using field measurements to derive diameter at breast height (DBH) (r² = 0.883, RMSE = 10.32 cm) from the crown diameter. The accuracy of tree height extracted from the LiDAR-based surface model was compared with the field-measured height (RMSE = 1.64 m). We found that the DBH and tree height extracted from the remotely sensed data were lower than their respective field-measured values without adjustment. The magnitude of differences in these measures tended to be larger for smaller-stature trees than for larger stature species. Using DBH and tree height calculated from remotely sensed data, aboveground biomass (r² = 0.881, RMSE = 799.2 kg) was calculated for individual tree and compared with results from field-measured DBH and height. We present guidelines for identifying potential errors in each step of data processing. These findings inform the development of procedures for monitoring tree growth with remote sensing and for calculating single tree level carbon storage using DBH from crown diameter and tree height in the urban forest.     

Rice production and climate change: design and development of a GIS database to complement simulation models

A cooperative project between the International Rice Research Institute in Los Baños, Philippines, and the U.S. EPA Environmental Research Laboratory in Corvallis, Oregon, was initiated to estimate how rice yield in Asia might be affected by future climate change and enhanced UV-B irradiance following stratospheric ozone depletion. A radiative transfer model was used to estimate daily UV-B irradiance levels using remotely sensed ozone and cloud cover data for 1274 meteorological stations. A rice yield model using daily climatic data and cultivar-specific coefficients was used to predict changes in yield under given climate change scenarios. This paper gives an overview of the data required to run these two models and describes how a geographical information system (GIS) was used as a data pre- or postprocessor. Problems in finding reliable datasets such as cloud cover data needed for the UV-B radiation model and radiation data needed for the rice yield model are discussed. Issues of spatial and temporal scales are also addressed. Using simulation models at large spatial scales helped identify weaknesses of GIS data overlay and interpolation capabilities. Even though we focussed our efforts on paddy rice, the database is not intended to be system specific and could also be used to analyze the response of other natural systems to climatic change.     

Factors affecting intraspecific variation in home range size of a large African herbivore

Factors affecting intraspecific variation in home range size have rarely been examined using modern statistical and remote sensing methods. This is especially true for animals in seasonal savanna environments in Africa, despite this biome??s importance for both conservation and development goals. We studied the impacts of spatial and temporal variability in environmental conditions, along with individual and social factors, on home range sizes in African buffalo (Syncerus caffer) in northeastern Namibia. Our data set spans 4?years, is derived from 32 satellite tracking collars, and contains over 35,000 GPS locations. We used the local convex hull method to estimate home range size from 31 buffalo captured at 6 sites. We used a variety of remotely sensed data to characterize potential anthropogenic and natural boundaries, as well as seasonal and temporal heterogeneity in environmental conditions. Using an information-theoretic, mixed effects approach, our analyses showed that home ranges varied over two orders of magnitude and are among the largest recorded for this species. Variables relating to vegetation and habitat boundaries were more important than abiotic environmental conditions and individual or social factors in explaining variation in home range size. The relative contributions of environmental, individual, social, and linear boundary variables to intraspecific home range size have rarely been examined and prior to this had not been assessed for any species in seasonal savannas of Africa. Understanding the factors that condition space-use patterns of wildlife in this area will lead to better-informed conservation and sustainable development decisions.     

Gradient modeling of conifer species using random forests

Landscape ecology often adopts a patch mosaic model of ecological patterns. However, many ecological attributes are inherently continuous and classification of species composition into vegetation communities and discrete patches provides an overly simplistic view of the landscape. If one adopts a niche-based, individualistic concept of biotic communities then it may often be more appropriate to represent vegetation patterns as continuous measures of site suitability or probability of occupancy, rather than the traditional abstraction into categorical community types represented in a mosaic of discrete patches. The goal of this paper is to demonstrate the high effectiveness of species-level, pixel scale prediction of species occupancy as a continuous landscape variable, as an alternative to traditional classified community type vegetation maps. We use a Random Forests ensemble learning approach to predict site-level probability of occurrence for four conifer species based on climatic, topographic and spectral predictor variables across a 3,883 km² landscape in northern Idaho, USA. Our method uses a new permutated sample-downscaling approach to equalize sample sizes in the presence and absence classes, a model selection method to optimize parsimony, and independent validation using prediction to 10% bootstrap data withhold. The models exhibited very high accuracy, with AUC and kappa values over 0.86 and 0.95, respectively, for all four species. The spatial predictions produced by the models will be of great use to managers and scientists, as they provide vastly more accurate spatial depiction of vegetation structure across this landscape than has previously been provided by traditional categorical classified community type maps.     

A comparison of satellite data and landscape variables in predicting bird species occurrences in the Greater Yellowstone Ecosystem,USA

We compare the accuracy of predicting the occurrence of 11 bird species in montane meadows of the Greater Yellowstone National Park ecosystem, in the states of Montana and Wyoming, USA. We used remotely sensed, landscape, and habitat data. The meadow type, as determined from the remotely sensed data, was highly correlated with abundances of six of the 11 bird species. Landscape variables significant in predicting occurrence were selected using a stepwise multiple regression for each bird species. These variables were then used in a multiple regression with the variable meadow type. As expected, the abundances of the generalist species (American Robin, Dark-eyed Junco, White-crowned Sparrow, Brewer's Blackbird, and Chipping Sparrow) were not strongly correlated with landscape variables or meadow type. Conversely, abundances of the Common Snipe, Common Yellowthroat, Lincoln's Sparrow, Savannah Sparrow, Vesper Sparrow, and Yellow Warbler were highly correlated with meadow type and landscape variables such as percent cover of willow (Salix spp.), graminoid, woody vegetation, sagebrush (Artemisia spp.), and graminoid and shrub biomass. The results from our study indicate that remotely sensed data are applicable for estimating potential habitats for bird species in the different types of montane meadows. However, to improve predictions about species in specific sites or areas, we recommend the use of additional landscape metrics and habitat data collected in the field.     

Associations of Bird Species Richness and Community Composition with Local and Landscape-scale Environmental Factors in Borneo

A comprehensive understanding of variables associated with spatial differences in community composition is essential to explain and predict biodiversity over landscape scales. In this study, spatial patterns of bird diversity in Central Kalimantan, Indonesia, were examined and associated with local-scale (habitat structure and heterogeneity) and landscape-scale (logging, slope position and elevation) environmental variables. Within the study area (c. 196 km²) local habitat structure and heterogeneity varied considerably, largely due to logging. In total 9747 individuals of 177 bird species were recorded. Akaike's information criterion (AIC) revealed that the best explanatory models of bird community similarity and species richness included both local- and landscape-scale environmental variables. Important local-scale variables included liana abundance, fern cover, sapling density, tree density, dead wood abundance and tree architecture, while important landscape-scale variables were elevation, logging and slope position. Geographic distance between sampling sites was not significantly associated with spatial variation in either species richness or similarity. These results indicate that deterministic environmental processes, as opposed to dispersal-driven stochastic processes, primarily structure bird assemblages within the spatial scale of this study and confirm that highly variable local habitat measures can be effective means of predicting landscape-scale community patterns.     

Ancient experiments: forest biodiversity and soil nutrients enhanced by Native American middens

The legacy of ancient human practices can affect the diversity and structure of modern ecosystems. Here, we examined how prehistoric refuse dumps (“middens”) impacted soil chemistry and plant community composition in forests along the Chesapeake Bay by collecting vegetational and soil nutrient data. The centuries- to millennia-old shell middens had elevated soil nutrients compared to adjacent sites, greater vegetative cover, especially of herb and grass species, and higher species richness. Not only are middens important archaeological resources, they also offer a remarkable opportunity to test ecological hypotheses about nutrient addition over very long time scales. We found no evidence, for example, that elevated nutrients enhanced invasion by non-native species as predicted by the fluctuating resource hypothesis. However, we did find that elevated nutrients shifted community structure from woody species to herbaceous species, as predicted by the structural carbon-nutrient hypothesis. These results highlight the long-lasting effects that humans can have on abiotic and biotic properties of the natural environment, and suggest the potential for modern patterns of species’ distributions and abundances to reflect ancient human activities.     

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.     

What multiscale environmental drivers can best be discriminated from a habitat index derived from a remotely sensed vegetation time series?

Understanding which environmental conditions are critical for species survival is a critical, ongoing question in ecology. These conditions can range from climate, at the broadest scale, through to elevation and other local landscape conditions, to fine scale landscape patterns of land cover and use. Remote sensing is an ideal technology to monitor and assess changes in these environmental conditions at a variety of spatial and temporal scales, with many studies focusing on the physiological state of vegetation derived from time series of satellite measurements. As vegetation occurs within specific climatic zones, over certain soil, terrain, and land cover types, it can be difficult to decipher the influence of the underlying role of climate, topography, soil, and land cover on the observed vegetation signal. In this article, we specifically addressed this problem by asking the question: what is the relative impact and importance of these different scales of environmental drivers on the temporal and spatial patterns observed on a habitat index derived from remotely sensed data? To find the solution, we utilized a SPOT VEGETATION-normalized difference vegetation index time series of Europe to create a remote-sensing-derived habitat index, which incorporates aspects of productivity, seasonality, and cover. We then compared the observed temporal and spatial variations in the index to a pan-Europe terrestrial classification system, which explicitly incorporates variations in climate, terrain, soil parent material, land cover, and use. Results indicated that the most accurate level of discrimination from the habitat index was at the broadest level of the hierarchy, climate, while the poorest degree of discrimination was associated with elevation. In terms of similarity on the index across time and space, we found that arable and forest cover classes were more similar across elevation and parent materials than across other land cover types within them. Analyzing the remote-sensing index, at multiple scales, provides significant insights into the drivers of satellite-derived greenness indices, as well as highlights the benefit and cautions associated with linking satellite-derived indirect indicators to species distribution modeling and biodiversity.     

Integrating single-species management and landscape conservation using regional habitat occurrence models: the northern goshawk in the Southwest,USA

Conservation planners and land managers are often confronted with scale-associated challenges when assessing the relationship between land management objectives and species conservation. Conservation of individual species typically involves site-level analyses of habitat, whereas land management focuses on larger spatial extents. New models are needed to more explicitly integrate species-specific conservation with landscape or regional scales. We address this challenge with an example using the northern goshawk (Accipiter gentilis), a forest raptor with circumpolar distribution that is the focus of intense debate regarding forest management on public lands in the southwestern USA. To address goshawk-specific habitat conservation across a management area of 22,800-km² in northern Arizona, we focused on the territory scale rather than individual nest sites. We compiled a 17-year database of 895 nest sites to estimate territory locations. We then estimated the likelihood of territory occurrence for the entire management area using multiple logistic regression within an expert-driven, spatially balanced, and information-theoretic framework. Our occurrence model incorporated forest structure variables that were derived from USFS Forest Inventory and Analysis plots and high-resolution satellite imagery. Results indicated that high canopy-bulk density, intermediate canopy-base heights, and low variation in tree density were strong predictors of territory occurrence. We used model-averaged parameter estimates for these variables to map and explore patterns of territory distribution across multiple land jurisdictions and ecological subregions. Our iterative modeling approach complements previous demographic studies in the region. It also provides a robust framework for integrating species conservation and landscape management in ongoing and future regional planning efforts.     

Quantifying patch distribution at multiple spatial scales: Applications to wildlife-habitat models

Multiscale analyses are widely employed for wildlife-habitat studies. In most cases, however, each scale is considered discrete and little emphasis is placed on incorporating or measuring the responses of wildlife to resources across multiple scales. We modeled the responses of three Arctic wildlife species to vegetative resources distributed at two spatial scales: patches and collections of patches aggregated across a regional area. We defined a patch as a single or homogeneous collection of pixels representing 1 of 10 unique vegetation types. We employed a spatial pattern technique, three-term local quadrat variance, to quantify the distribution of patches at a larger regional scale. We used the distance at which the variance for each of 10 vegetation types peaked to define a moving window for calculating the density of patches. When measures of vegetation patch and density were applied to resource selection functions, the most parsimonious models for wolves and grizzly bears included covariates recorded at both scales. Seasonal resource selection by caribou was best described using a model consisting of only regional scale covariates. Our results suggest that for some species and environments simple patch-scale models may not capture the full range of spatial variation in resources to which wildlife may respond. For mobile animals that range across heterogeneous areas we recommend selection models that integrate resources occurring at a number of spatial scales. Patch density is a simple technique for representing such higher-order spatial patterns.     

Quantifying patch distribution at multiple spatial scales: Applications to wildlife-habitat models

Multiscale analyses are widely employed for wildlife-habitat studies. In most cases, however, each scale is considered discrete and little emphasis is placed on incorporating or measuring the responses of wildlife to resources across multiple scales. We modeled the responses of three Arctic wildlife species to vegetative resources distributed at two spatial scales: patches and collections of patches aggregated across a regional area. We defined a patch as a single or homogeneous collection of pixels representing 1 of 10 unique vegetation types. We employed a spatial pattern technique, three-term local quadrat variance, to quantify the distribution of patches at a larger regional scale. We used the distance at which the variance for each of 10 vegetation types peaked to define a moving window for calculating the density of patches. When measures of vegetation patch and density were applied to resource selection functions, the most parsimonious models for wolves and grizzly bears included covariates recorded at both scales. Seasonal resource selection by caribou was best described using a model consisting of only regional scale covariates. Our results suggest that for some species and environments simple patch-scale models may not capture the full range of spatial variation in resources to which wildlife may respond. For mobile animals that range across heterogeneous areas we recommend selection models that integrate resources occurring at a number of spatial scales. Patch density is a simple technique for representing such higher-order spatial patterns.     

Spatial dynamics of coastal forest bird assemblages: the influence of landscape context,forest type,and structural connectivity

Context

Complex structural connectivity patterns can influence the distribution of animals in coastal landscapes, particularly those with relatively large home ranges, such as birds. To understand the nuanced nature of coastal forest avifauna, where there may be considerable overlap in assemblages of adjacent forest types, the concerted influence of regional landscape context and vegetative structural connectivity at multiple spatial scales warrants investigation.

Objectives

This study determined whether species compositions of coastal forest bird assemblages differ with regional landscape context or with forest type, and if this is influenced by structural connectivity patterns measured at multiple spatial scales.

Methods

Three replicate bird surveys were conducted in four coastal forest types at ten survey locations across two regional landscape contexts in northeast Australia. Structural connectivity patterns of 11 vegetation types were quantified at 3, 6, and 12 km spatial scales surrounding each survey location, and differences in bird species composition were evaluated using multivariate ordination analysis.

Results

Bird assemblages differed between regional landscape contexts and most coastal forest types, although Melaleuca woodland bird assemblages were similar to those of eucalypt woodlands and rainforests. Structural connectivity was primarily correlated with differences in bird species composition between regional landscape contexts, and correlation depended on vegetation type and spatial scale.

Conclusions

Spatial scale, landscape context, and structural connectivity have a combined influence on bird species composition. This suggests that effective management of coastal landscapes requires a holistic strategy that considers the size, shape, and configuration of all vegetative components at multiple spatial scales.

     

Dispersion of kangaroo rat mounds at multiple scales in New Mexico,USA

Burrowing mammals create disturbances that increase the ecological heterogeneity of landscapes. In desert systems, banner-tailed kangaroo rats (Dipodomys spectabilis) construct large mounds that greatly influence the spatial patterning of soils, plants, and animals. The overall effects of the patches generated by D. spectabilis depend on the dispersion patterns of the mounds; these patterns may be sensitive to scale and landscape position. We examined the distribution of D. spectabilis mounds across multiple scales in four 40-ha grassland plots in New Mexico, USA. We used Ripley's K-function for our point-pattern analysis. The dispersion patterns of mounds were generally scale-sensitive but depended somewhat on plot-level densities, which were related to topographic position and grazing history. Mound spacing was either regular or random at small scales (0–50 m), random or aggregated at intermediate scales (50–300 m), and aggregated at large scales (300–3000 m). This scale-dependency of pattern reflected spatial domains in which different biotic (territoriality, dispersal, grazing) and abiotic (soil texture and drainage) factors exerted strong influences. How other organisms perceive the spatial patterning of mounds will depend on the extent of their movements. Patches may appear regular to one species but aggregated to another. The dispersion of D. spectabilis mounds also has implications for the spatial structuring of desert rodent communities. D. spectabilis excludes smaller species of kangaroo rats from areas around their mounds; they create spatial heterogeneity in behavioral dominance that may influence the distribution of subordinate species at multiple scales.     

Hierarchical relationships between landscape structure and temperature in a managed forest landscape

Management may influence abiotic environments differently across time and spatial scale, greatly influencing perceptions of fragmentation of the landscape. It is vital to consider a priori the spatial scales that are most relevant to an investigation, and to reflect on the influence that scale may have on conclusions. While the importance of scale in understanding ecological patterns and processes has been widely recognized, few researchers have investigated how the relationships between pattern and process change across spatial and temporal scales. We used wavelet analysis to examine the multiscale structure of surface and soil temperature, measured every 5 m across a 3820 m transect within a national forest in northern Wisconsin. Temperature functioned as an indicator – or end product – of processes associated with energy budget dynamics, such as radiative inputs, evapotranspiration and convective losses across the landscape. We hoped to determine whether functional relationships between landscape structure and temperature could be generalized, by examining patterns and relationships at multiple spatial scales and time periods during the day. The pattern of temperature varied between surface and soil temperature and among daily time periods. Wavelet variances indicated that no single scale dominated the pattern in temperature at any time, though values were highest at finest scales and at midday. Using general linear models, we explained 38% to 60% of the variation in temperature along the transect. Broad categorical variables describing the vegetation patch in which a point was located and the closest vegetation patch of a different type (landscape context) were important in models of both surface and soil temperature across time periods. Variables associated with slope and microtopography were more commonly incorporated into models explaining variation in soil temperature, whereas variables associated with vegetation or ground cover explained more variation in surface temperature. We examined correlations between wavelet transforms of temperature and vegetation (i.e., structural) pattern to determine whether these associations occurred at predictable scales or were consistent across time. Correlations between transforms characteristically had two peaks; one at finer scales of 100 to 150 m and one at broader scales of ^>300 m. These scales differed among times of day and between surface and soil temperatures. Our results indicate that temperature structure is distinct from vegetation structure and is spatially and temporally dynamic. There did not appear to be any single scale at which it was more relevant to study temperature or this pattern-process relationship, although the strongest relationships between vegetation structure and temperature occurred within a predictable range of scales. Forest managers and conservation biologists must recognize the dynamic relationship between temperature and structure across landscapes and incorporate the landscape elements created by temperature-structure interactions into management decisions.     

Spatial patterns of bird community similarity: bird responses to landscape composition and configuration in the Atlantic forest

Studies dealing with community similarity are necessary to understand large scale ecological processes causing biodiversity loss and to improve landscape and regional planning. Here, we study landscape variables influencing patterns of community similarity in fragmented and continuous forest landscapes in the Atlantic forest of South America, isolating the effects of forest loss, fragmentation and patterns of land use. Using a grid design, we surveyed birds in 41 square cells of 100 km² using the point count method. We used multivariate, regression analyses and lagged predictor autoregressive models to examine the relative influence of landscape variables on community similarity. Forest cover was the primary variable explaining patterns of bird community similarity. Similarity showed a sudden decline between 20 and 40% of forest cover. Patterns of land use had a second order effect; native bird communities were less affected by forest loss in landscapes dominated by tree plantations (the most suitable habitat for native species) than in landscapes dominated by annual crops or cattle pastures. The effects of fragmentation were inconclusive. The trade-off between local extinctions and the invasion of extra-regional species using recently created habitats is probably the mechanism generating the observed patterns of community similarity. Limiting forest loss to 30–40% of the landscape cover and improving the suitability of human-modified habitats will contribute to maintain the structure and composition of the native forest bird community in the Atlantic forest.