Digital Land Cover Data Research Articles

Modeling probability of waterfowl encounters from satellite imagery of habitat in the central Canadian arctic

We used aerial survey and corresponding digital land cover data to develop species-habitat models to describe breeding-ground distributions and landscape-level habitat associations of greater white-fronted geese (Anser albifrons), Canada (Branta canadensis) and cackling geese (B. hutchinsii), tundra swans (Cygnus columbianus), king eiders (Somateria spectabilis), and long-tailed ducks (Clangula hyemalis). We then used habitat associations in the Queen Maud Gulf Migratory Bird Sanctuary and the Rasmussen Lowlands, Nunavut, Canada, in models to predict distributions of focal species in each study area. We used the receiver operating characteristic (ROC) method and the area-under-the-curve (AUC) metric to evaluate predictive accuracy (hereafter, quality) of models. In the Queen Maud Gulf, AUC values suggested reasonable model discrimination for white-fronted geese, Canada geese, and tundra swans (i.e., AUC > 0.7). Quality of species-habitat models for king eiders and long-tailed ducks was less than other species considered, but these models still predicted encounters and non-encounters significantly better than the null model. For all species, quality of species-habitat models was lesser for the Rasmussen Lowlands than for the Queen Maud Gulf, although discrimination ability for Rasmussen Lowland distributions remained significantly better than corresponding null models for geese and swans, but not for seaducks. Our research suggested that species' distributions modeled with landscape-level habitat data is a tractable method to 1) identify habitat associations, 2) determine key habitats and regions, and 3) predict probable summer distributions of some waterfowl species over relatively large areas of the arctic from satellite imagery. © 2013 The Wildlife Society.

A synthesis of rates and controls on elemental mercury evasion in the Great Lakes Basin

Rates of surface-air elemental mercury (Hg 0) fluxes in the literature were synthesized for the Great Lakes Basin (GLB). For the majority of surfaces, fluxes were net positive (evasion). Digital land-cover data were combined with representative evasion rates and used to estimate annual Hg 0 evasion for the GLB (7.7 Mg/yr). This value is less than our estimate of total Hg deposition to the area (15.9 Mg/yr), suggesting the GLB is a net sink for atmospheric Hg. The greatest contributors to annual evasion for the basin are agricultural (∼55%) and forest (∼25%) land cover types, and the open water of the Great Lakes (∼15%). Areal evasion rates were similar across most land cover types (range: 7.0–21.0 μg/m 2-yr), with higher rates associated with urban (12.6 μg/m 2-yr) and agricultural (21.0 μg/m 2-yr) lands. Uncertainty in these estimates could be partially remedied through a unified methodological approach to estimating Hg 0 fluxes.

A river runs through it: Land-use and the composition of vegetation along a riparian corridor in the Cape Floristic Region, South Africa

Riparian zones are important for the many ecosystem services they supply. In settled areas, the vegetation of such zones is shaped by human land-use; this often creates conditions under which alien plant species thrive. Alien plants have been shown to induce large-scale changes in riparian habitats, and they pose a major threat to the continued provision of key ecosystem services. We used direct gradient analysis to assess correlations between land-use and the composition of vegetation along a riparian river corridor in the highly transformed landscape surrounding Stellenbosch in South Africa’s Western Cape Province. Vegetation plots were sampled along the entire length of the river from headwaters to estuary (ca. 40 km). Plant community composition was analyzed in relation to land-use data collected in the field, and additional land-use variables computed from digital land-cover data. Patterns of plant community structure were found to be directly related to land-use, with measures of cover, richness, and diversity differing significantly among land-use types. Portions of the riparian zone adjacent to agricultural land had the greatest level of alien plant cover, while areas bordered by urban land maintained the highest alien species richness. Areas adjacent to grazing and natural lands showed intermediate and low levels of invasion, respectively. Several native species were found to persist in areas with high abundance and diversity of invasive alien plants, suggesting that they will be valuable focal species for future restoration attempts. Due to the level of human-mediated change in many areas of the riparian zone, restoration to historic conditions over most of the river is not considered feasible. These areas should be recognized as examples of novel ecosystems, and management efforts should focus on restoring or creating desirable ecosystem functions, rather than on achieving assemblages comprising only native species.

Ectoparasite Infestations of Hedgehogs (Erinaceus europaeus) are Associated with Small-Scale Landscape Structures in an Urban–Suburban Environment

Animals that exploit heterogeneous and patchy environments encounter different local habitat conditions that influence their interaction with the environment, such as the acquisition of parasites. How and at which scales interaction processes between parasites, hosts, and the environment are realized remains largely unknown. We examined the infestation patterns of 56 hedgehogs (Erinaceus europaeus) with fleas and ticks at a small spatial scale within a 12 km(2) area along a suburban-urban gradient in southwestern Germany. The structure and type of habitats surrounding hedgehog capture locations were estimated from digital land cover data within radii of 20, 50, and 100 m. These were assumed to match the ranging area and underlying heterogeneous landscape matrix in which host-parasite interactions take place. Landscape-based models suggested that flea burdens significantly decreased with the diversity and heterogeneity of land cover, as well as with the areal coverage of roads within radii of 50 and 100 m. Overall tick infestation levels were mostly explained by the number of arable patches and the areal coverage of roads within radii of 50 and 100 m, as well as date of capture. Examination of the semivariance in model residuals revealed no evident spatial structure in any of the models with flea or tick infestation patterns as response variables. Our results, which are based on a sampling scheme within a relatively small spatio-temporal window, suggest that heterogeneous landscape matrices affect parasitization rates of animals in urban environments, with clear differences at the individual level.

Evaluating parameter effects in a DEM ‘burning’ process based on land cover data

AbstractOver the last years, various methods regarding the use of digital land cover data to change digital elevation models (DEMs) have been reported in the literature. Usually, these methods rely on digital river networks (DRNs) or digital river and lake networks (DRLNs) to force preferential flow directions of pixels composing a DEM to get more realistic hydrological information, and reduce the accumulated number of errors in the information acquisition process. More recently, a new method that takes advantage of spatial distribution of rivers and floodplains derived from classified radar images has been presented. The method is based on a double ‘burning’ process. The first process uses stream lines to reduce elevations in the DEM (traditional stream burning approach—SB approach) and the second uses the floodplains to create valleys forcing the drainage of these areas to the main river. This approach, referred to as floodplain burning approach (FB approach), was applied to a basin in the Amazon basin. It greatly improved the results compared to the SB approach. However, this method requires the calibration of various parameters to represent accurately land cover types. This study proposes a comprehensive evaluation of the influence of these parameters on the DEM‐based hydrological information acquisition and on the original DEM. Consequently, a method of parameters set selection is provided. The method was applied to a sub‐basin of the Negro River basin in Amazonia. Several parameter combinations were evaluated in the watershed delineation process. Particular attention was paid to whether those parameters with a greater impact on DEM yielded better results in the case study. However, these DEM changes can be reduced with the knowledge of vegetation's heights, the vegetation having a substantial effect on remote acquisition of DEMs in tropical forests. Copyright © 2009 John Wiley & Sons, Ltd.

Validation of the sigmoid wave curve fitting algorithm on a forest-tundra ecotone in the Northwest Territories, Canada

Ecotones are gradual transitions between two adjacent ecological systems. Ecotones are characterized by their spatial properties which are reflected in an ecotone width and location and the temporal variation across time during succession or environmental change on both a local and a global scale. If only one main environmental factor drives this gradual change the shape of the ecotone is evident as a sigmoid wave. In this paper we explore a two-dimensional sigmoid wave curve fitting algorithm that describes an ecotone and validate it on classified remote sensing data of a forest-tundra ecotone in the Northwest Territories of Canada. The estimated location and width of the forest-tundra ecotone are validated with digital land cover data. We conclude that the algorithm was able to delineate the forest-tundra ecotone based upon the classified remote sensing image and is robust for various algorithm parameter settings.

Modelling the water balance of a mesoscale catchment basin using remotely sensed land cover data

Summary Hydrological modelling of mesoscale catchments is often adversely affected by a lack of adequate information about specific site conditions. In particular, digital land cover data are available from data sets which were acquired on a European or a national scale. These data sets do not only exhibit a restricted spatial resolution but also a differentiation of crops and impervious areas which is not appropriate to the needs of mesoscale hydrological models. In this paper, the impact of remote sensing data on the reliability of a water balance model is investigated and compared to model results determined on the basis of CORINE (Coordination of Information on the Environment) Land Cover as a reference. The aim is to quantify the improved model performance achieved by an enhanced land cover representation and corresponding model modifications. Making use of medium resolution satellite imagery from SPOT, LANDSAT ETM+ and ASTER, detailed information on land cover, especially agricultural crops and impervious surfaces, was extracted over a 5-year period (2000–2004). Crop-specific evapotranspiration coefficients were derived by using remote sensing data to replace grass reference evapotranspiration necessitated by the use of CORINE land cover for rural areas. For regions classified as settlement or industrial areas, degrees of imperviousness were derived. The data were incorporated into the hydrological model GROWA (large-scale water balance model), which uses an empirical approach combining distributed meteorological data with distributed site parameters to calculate the annual runoff components. Using satellite imagery in combination with runoff data from gauging stations for the years 2000–2004, the actual evapotranspiration calculation in GROWA was methodologically extended by including empirical crop coefficients for actual evapotranspiration calculations. While GROWA originally treated agricultural areas as homogeneous, now a consideration and differentiation of the main crops is possible. The accuracy was determined by runoff measurements from gauging stations. Differences in water balances resulting from the use of remote sensing data as opposed to CORINE were analysed in this study using a representative subcatchment. Resulting Nash–Sutcliff model efficiencies improved from 0.372 to 0.775 and indicate that the enhanced model can produce thematically more accurate and spatially more detailed local water balances. However, the proposed model enhancements by satellite imagery have not exhausted the full potential of water balance modelling, for which a higher temporal resolution is required.

Representation of Landcover Along Breeding Bird Survey Routes in the Northern Plains

ABSTRACT The North American Breeding Bird Survey (BBS) is used extensively to make inferences about populations of many North American bird species and is increasingly being used for avian conservation planning. How well BBS routes represent the landscape is poorly known, even though accuracy of representation could significantly affect inferences made from BBS data. We used digital landcover data to examine how well landcover within 400‐m buffers around BBS routes represented the surrounding landscape (the route neighborhood) for 52 routes in the Prairie Pothole Region of North Dakota and South Dakota. Differences in composition between landcover along BBS routes and the route neighborhood were not statistically significant for upland cover classes. The area of temporary and seasonal wetland basins was accurately represented by BBS routes in our study area, but the area of semipermanent and permanent wetland basins was significantly underrepresented along BBS routes. Number of wetland basins and upland patches was higher along routes. Area of urban, forest, and hay landcover classes was higher along routes, although differences were not statistically significant. Amount of bias in landcover representation was negatively correlated with the proportion of each landcover type in the study area, but bias was not correlated with area of the route neighborhoods. Differences between landcover along BBS routes and the route neighborhood were primarily attributable to increased anthropogenic activity along roads and siting of roads away from relatively large, deep water bodies. Our results suggest that inferences made from BBS data in our study region are likely biased for species that are associated with deeper‐water habitats or are strongly influenced by landscape fragmentation. Inferences made from BBS data for species associated with uplands or shallow wetlands are less likely to be biased because of differences in landcover composition.

A Simple Spatial Model to Explain the Distribution of Human Tick-Borne Encephalitis Cases in Hungary

Tick-borne encephalitis (TBE) is a common medical problem in Hungary and throughout much of Europe and Asia. This paper develops a geographic model that helps to predict the distribution of human tick-borne encephalitis cases in Hungary. The model is tested on a dataset of serologically confirmed TBE cases mapped by patients' residences. Case densities (incidence rates) are compared to predicted distributions of TBE derived from digital land-cover data. Maps are analyzed at the county level and on a smaller spatial scale. The analyses identified three major factors that shape the geographic distribution of human TBE cases in Hungary. The most important component is the distribution of forest habitat. TBE incidence correlates positively with the amount of forested habitat in each county. On a finer scale, the amount of forests within a 2500-meter radius of each town and village correlated significantly with TBE incidence rate. Based on these data, about 30% of the variation in TBE incidence is accounted for by the specific distribution of forest habitats in Hungary. Besides the distribution of forests, differences in human land-use practices among regions also affect the distribution of TBE cases. Additionally, because of the low transmission rate of the virus to humans, the perceived distribution of TBE cases is affected by random stochastic events. As a consequence of stochastic variation, meaningful patterns in the distribution of TBE cases can be only recognized when data are analyzed over broader temporal and spatial scales.

A cautionary tale regarding use of the National Land Cover Dataset 1992

Abstract Digital land-cover data are among the most popular data sources used in ecological research and natural resource management. However, processes for accurate land-cover classification over large regions are still evolving. We identified inconsistencies in the National Land Cover Dataset 1992, the most current and available representation of land cover for the conterminous United States. We also report means to address these inconsistencies in a bird-habitat model. We used a Geographic Information System (GIS) to position a regular grid (or lattice) over the upper midwestern United States and summarized the proportion of individual land covers in each cell within the lattice. These proportions were then mapped back onto the lattice, and the resultant lattice was compared to satellite paths, state borders, and regional map classification units. We observed mapping inconsistencies at the borders between mapping regions, states, and Thematic Mapper (TM) mapping paths in the upper midwestern United Sta...

Modelling the distribution of palsas in Finnish Lapland with logistic regression and GIS

AbstractThe location of palsas (peat mounds with a perennially frozen core) was mapped in an area of 3370 km2 in Finnish Lapland by interpreting aerial digital photographs. Using environmental variables derived from digital land cover data and an elevation model, the distribution of palsas was modelled using geographic information system (GIS) techniques and multiple logistic regression. The relative roles of eight environmental variables potentially affecting the distribution of the palsas were studied in a spatial grid system with 3370 grid squares of 1.0 km2, of which 172 were found to contain palsas. The altitudes of the palsas varied from 180 m to 390 m. In the logistic regression model the probability of the presence of a palsa in a 1.0 km2 square increased with 1) the area of mire, 2) the proportion of flat topography, 3) water cover and 4) elevation of the lowest point in the 1.0 km2 analysis square. The palsa distribution model was validated by fitting it to an independent test area of 300 squares bordering on the main study area: the overall classification rate was 97.67%. The application of GIS data and techniques with logistic regression modelling has potential for wide use in studies on distribution patterns in periglacial processes and landforms. Copyright © 2002 John Wiley & Sons, Ltd.

Using digital terrain analysis modeling techniques for the parameterization of a hydrologic model

This paper discusses the application of digital terrain analysis modeling techniques to the parameterization of a semi-distributed hydrologic model. Most current techniques for deriving physiographic parameters in watershed analyses, including those using commercial geographic information systems (GIS), are tedious, costly and time consuming. The demands of these techniques result in them usually being limited in practical application to deriving parameters at only one level of detail or for only one set of sub-basins. This paper presents a computerized interface (SLURPAZ) that was developed to combine the output of an established digital terrain analysis model (TOPAZ) with digital land cover data to derive all the necessary physiographic parameters required as input by a widely used semi-distributed hydrological model (SLURP). This interface makes it possible to derive physiographic parameters rapidly and accurately, at several different levels of detail and for varying numbers of sub-basins. This paper describes the methods by which the SLURPAZ interface integrates land cover data with the topographic parameters derived by TOPAZ from a digital elevation model (DEM). It also presents an example application of the interface to an intermediate-sized alpine basin in Yukon, Canada. Hydrological model outputs obtained using the computerized interface are compared with those obtained using manual techniques.

Assessing habitat suitability at multiple scales: A landscape-level approach

The distribution and abundance of many plants and animals are influenced by the spatial arrangement of suitable habitats across landscapes. We derived habitat maps from a digital land cover map of the ∼ 178,000 km 2 Chesapeake Bay Watershed by using a spatial filtering algorithm. The regional amounts and patterns of habitats were different for species which occur in ‘woody’, ‘herbaceous’, and ‘woody-edge’ habitats. Habitat for finer-scale species (∼ 5 ha home ranges) was twice as abundant and more evenly distributed than habitat for coarser-scale species (∼ 410 ha home ranges) in a 11,000 km 2 sub-region. Potential impacts of land cover changes on habitats in different parts of the region were assessed by the frequency distributions of habitat suitability for smaller (∼ 3000 km 2) embedded watersheds. The methods described in this paper can be applied to several scales of digital land cover data, and used to derive multiple-scale habitat suitabilities for a number of species or guilds.

A digital land cover map of Wyoming, USA: a tool for vegetation analysis

Abstract. A Landsat Thematic Mapper (TM) based digital land cover map has been created for the state of Wyoming, USA, at moderate spatial (l‐km2 minimum mapping unit) and high typal (41 land cover types) resolution as part of the Wyoming Gap Analysis Program (WGAP). This map presents opportunities for regional characterization of land cover, especially vegetation, and for examination of ecological phenomena that manifest themselves over large areas. Using the digital land cover data, we describe Wyoming vegetation and examine positions of three prominent physiognomic transitions in Wyoming: the elevation of upper and lower treeline, and the position of the biogeographic boundary between shruband grass‐dominated vegetation. By area, the three leading land cover types in Wyoming are Artemisia tridentata ssp. wyomingensis sagebrush (33.4 %), mixed grass prairie (17.5 %) and Pinus contorta forest (6.5 %). Average upper‐treeline elevation in Wyoming is 2947 m, and decreases with increasing latitude at an average rate of about 0.5 m/km, less than the rate of about 0.9 m/km reported by Peet (1978) for a gradient from Mexico to Canada. Lower‐treeline occurs at an average elevation of 2241 m, and decreases with increasing latitude and with southerly aspect. In Wyoming, shrub‐dominated communities are more likely to occur than grass‐dominated communities as summer precipitation decreases below 282 mm. All of these relationships are subtle, and it appears that for particular areas, local factors are more important than regional climatic trends in explaining the position of phytogeographic boundaries.

A comparison of three landscape classifications and investigation of the potential for using remotely sensed land cover data for landscape classification

Numerical landscape classifications group land units (e.g. 1-km grid squares) with similar characteristics into distinct landscape types (termed land classes). Landscape classifications have considerable potential as tools in rural land-use planning. To date, a plethora of numerical classifications have been developed, yet little attempt has been made to quantify differences between classifications which may arise through the use of different attribute data and classification techniques. The present paper quantifies differences between three hierarchical landscape classifications developed using closely related algorithms, but using contrasting attribute data. Two classifications were based on land cover; from field survey and remote sensing, respectively. The third was based primarily on abiotic environmental variables. Quantitative comparisons between the three classifications indicate that similar land classes are recognisable at different levels of environmental and ecological organisation. It is concluded that the use of remotely sensed land cover in the production of landscape classification has many advantages over previous approaches. Foremost among these benefits are the ease with which digital land cover data can be handled within Geographic Information Systems and the consequential ability to produce classifications rapidly at differing spatial resolutions. The potential roles of numerical landscape classifications in rural planning are reviewed.

Remote-sensing technology applied to forest assessment in California

Remote-sensing technology has been thoroughly evaluated for the analysis of California forest policy. A statewide, 1.6-acre-resolution, digital land-cover data base of Landsat Multispectral Scanner (MSS) classification has been produced. Three major resource regions have been analyzed in detail and one of them geographically integrated with 12 other physical and socioeconomic data layers to model fire and reforestation problems, using a geographic information system (GIS). A study of GIS design criteria has been conducted and the California Department of Forestry, the cooperator in all of these studies, is presently evaluating the alternatives and implementing certain aspects of them.