Geographic Information System Data Layers Research Articles

RUSLE and GIS Based Soil Erosion Modelling in a Micro-Watershed (1E1C4a2) in the Budgam district of Jammu and Kashmir

Abstract: A comprehensive methodology that integrates Revised Universal Soil Loss Equation (RUSLE) model and Geographic Information System (GIS) techniques was adopted to determine that soil erosion vulnerability of micro-watershed in J&K India. The spatial pattern of annual soil erosion rate was obtained by integrating geo-environmental variables in a raster based GIS method. GIS data layers including, rainfall erosivity (R). Soil erodibility (K), slope length and steepness (LS), cover management (C) and conservation Practice (P) factors were computed to determine their effect on average annual soil loss in the area. The resultant map of annual soil erosion shows maximum soil loss of 2.9246 t h Y with a close relation to grass land area, agricultural lands and orchards (with low LS) . The spatial erosion maps generated with RUSLE method and GIS can serve as effective inputs in deriving strategies for land planning management in the environmentally sensitive agricultural areas.

Data mining and model-predicting a global disease reservoir for low-pathogenic Avian Influenza (AI) in the wider pacific rim using big data sets

Avian Influenza (AI) is a complex but still poorly understood disease; specifically when it comes to reservoirs, co-infections, connectedness and wider landscape perspectives. Low pathogenic (Low-path LP) AI in chickens caused by less virulent strains of AI viruses (AIVs)—when compared with highly pathogenic AIVs (HPAIVs)—are not even well-described yet or known how they contribute to wider AI and immune system issues. Co-circulation of LPAIVs with HPAIVs suggests their interactions in their ecological aspects. Here we show for the Pacific Rim an international approach how to data mine and model-predict LP AI and its ecological niche with machine learning and open access data sets and geographic information systems (GIS) on a 5 km pixel size for best-possible inference. This is based on the best-available data on the issue (~ 40,827 records of lab-analyzed field data from Japan, Russia, Vietnam, Mongolia, Alaska and Influenza Research Database (IRD) and U.S. Department of Agriculture (USDA) database sets, as well as 19 GIS data layers). We sampled 157 hosts and 110 low-path AIVs with 32 species as drivers. The prevalence across low-path AIV subtypes is dominated by Muscovy ducks, Mallards, Whistling Swans and gulls also emphasizing industrial impacts for the human-dominated wildlife contact zone. This investigation sets a good precedent for the study of reservoirs, big data mining, predictions and subsequent outbreaks of HPAI and other pandemics.

Next-generation geospatial-temporal information technologies for disaster management

Traditional geographic information systems (GIS) have been disrupted by the emergence of Big Data in the form of geo-coded raster, vector, and time-series Internet-of-Things data. This article discusses the application of new scalable technologies that go far beyond relational databases and file-based storage on spinning disk or tape to incorporate both storage and processing data in the same platform. The roles of the Apache Hadoop Distributed File Systems and NoSQL key-value stores such as the Apache Hbase are discussed, along with indexing schemes that optimally support geospatial-temporal use. We highlight how this new approach can rapidly search multiple GIS data layers to obtain insights in the context of early warning, impact evaluation, response, and recovery to earthquake and wildfire disasters.

Habitat Conservation, Ethanol, and Recent US Federal Farm Bills: A GIS Study Exploring Conservation Reserve Program Trends Pre- and Post-Ethanol Expansion in the Upper Midwest

Farmland in the Upper Midwest, USA, planted to corn has increased significantly in the past decade due to a rush toward ethanol production for fuel. Much of this increase came from fields that were previously protected under the Conservation Reserve Program (CRP), a Federal land retirement incentive. This Geographic Information Systems (GIS) study quantifies these land changes in Minnesota, a northern state in the U.S. Corn Belt, prior to the past decade using 1997 and 2008 CRP GIS data layers for individual fields. Currently available CRP enrollment data by county masks significant changes occurring on a field-level scale, data which is no longer publicly available. Results suggest that marginal lands better suited for riparian wildlife habitat and surface water protection were plowed under in favor of industrial corn production. This was especially notable with fields in close proximity to refineries or those in drinking water supply management or groundwater contamination susceptibility areas. Federal ethanol mandates, farm bill policies, and high corn prices contribute to the pressure to farm these sensitive lands. Currently, CRP enrollments are capped at 9.7 million hectares, reduced from a long-standing 13-million-hectare cap by the 2014 farm bill. Discussions are now underway in the U.S. Congress to craft the 2018 farm bill and farming and conservation interests are strongly urging the Federal government to restore the 13-million-hectare enrollment cap. Ecosystem functions and benefits are optimized when conservation efforts are maintained over long periods and not interrupted by short-term political or economic trends.

Technical Note: Method to Streamline Processing of Livestock Global Positioning System Collar Data

The use of global positioning system (GPS) technology to study livestock movement has been widely adopted in range and animal sciences; however, the methods for processing GPS collar data are varied among researchers and often involve repetitive, time-consuming steps to get data into a format available to view in geographic information system (GIS) software. The objective of this technical note is to present a method for quickly processing uniform datasets using a commonly available commercial GPS collar and Program R. Data generated by a Lotek 3300LR GPS collar were processed in seconds using “Lotek_Function,” which was developed using R coding. Traditional hand-processing of the same data generally requires 30 or more minutes and is prone to error due to the tedious, repetitive nature of the task. Due to the open-source nature of Program R, base codes can be modified to fit specific researchers’ needs when incorporating GIS data layers or models to assess behavior based on motion sensor data output from collars.

Determination of fire-access zones along road networks in fire-sensitive forests

Wild forest fires are one of the greatest environmental disasters affecting forest resources. Along the coastal zone of the Mediterranean region in Turkey, forested areas are classified as first-degree, fire-sensitive areas. Every year, thousands of hectares of forests have been destroyed in Turkey. In this study, fire-access zones were determined in the Mediterranean forests of Turkey, by utilizing geographic information systems (GIS) technology. The effective reach distance of fire hoses from both sides of roads was considered in order to delineate fire-access zones. The effective reach distance can vary based on the technical capabilities and hydraulic capacity of fire trucks (minimum and maximum pressures on water pump); terrain structures (uphill, downhill and flat); and ground slope. These factors and their influences were studied in fire sensitive forest areas located in the eastern Mediterranean city of Kahramanmaras in Turkey. First, terrain structures on both sides of the road network and ground-slope classes were determined based on GIS data layers. Then, fire access zones were delineated according to water pressure data, terrain structures, and ground-slope classes. The results indicated that 69.30 % of the forested areas were within the fire-access zones, while the rest of the forest was out of reach the fire hoses. The accessible areas were also calculated for forested areas with different fire-sensitivity degrees. The accessible areas were 69.59, 69.96, and 67.16 % for the forested areas that are sensitive to fires at the first, second, and third degrees, respectively. This finding has implications for the monitoring and management of fire threats in areas outside of the reach distance. The outside areas should receive extra attention and monitoring during the fire season so that fires are detected ahead of time and management has sufficient time to react. Besides, new roads should be considered for these areas in order to access more lands in a shorter amount of time.

Estimation of Soil Erosion Risk Using the Universal Soil Loss Equation (USLE) and Geo-Information Technology in Oued El Makhazine Watershed, Morocco

Soil erosion by water is one of the major threats to soils in the north of Morocco; soil erosion not only decreases agricultural productivity, but also reduces the water availability. In the current study, Oued El Makhazine watershed is selected to estimate annual soil loss using the Universal Soil Loss Equation (USLE), remote sensing (RS) and geographic information system (GIS). GIS data layers including, rainfall erosivity (R), soil erodibility (K), slope length and steepness (LS), cover management (C) and conservation practice (P) factors are computed to determine their effects on average annual soil loss in the area. The resultant map of annual soil erosion shows a maximum soil loss of 735 t·h-1·y-1, about 65.25% (1575 km2), of the watershed ranges between 0 and 95 t·h-1·y-1. Higher soil losses are observed at higher LS factor area. The spatial erosion maps generate with USLE method, remote sensing and GIS can serve as effective inputs in deriving strategies for land planning and management in the environmentally sensitive mountainous areas.

ESTIMATION OF ANNUAL AVERAGE SOIL LOSS, BASED ON RUSLE MODEL IN KALLAR WATERSHED, BHAVANI BASIN, TAMIL NADU, INDIA

Abstract. Soil erosion is a widespread environmental challenge faced in Kallar watershed nowadays. Erosion is defined as the movement of soil by water and wind, and it occurs in Kallar watershed under a wide range of land uses. Erosion by water can be dramatic during storm events, resulting in wash-outs and gullies. It can also be insidious, occurring as sheet and rill erosion during heavy rains. Most of the soil lost by water erosion is by the processes of sheet and rill erosion. Land degradation and subsequent soil erosion and sedimentation play a significant role in impairing water resources within sub watersheds, watersheds and basins. Using conventional methods to assess soil erosion risk is expensive and time consuming. A comprehensive methodology that integrates Remote sensing and Geographic Information Systems (GIS), coupled with the use of an empirical model (Revised Universal Soil Loss Equation- RUSLE) to assess risk, can identify and assess soil erosion potential and estimate the value of soil loss. GIS data layers including, rainfall erosivity (R), soil erodability (K), slope length and steepness (LS), cover management (C) and conservation practice (P) factors were computed to determine their effects on average annual soil loss in the study area. The final map of annual soil erosion shows a maximum soil loss of 398.58 t/ h-1/ y-1. Based on the result soil erosion was classified in to soil erosion severity map with five classes, very low, low, moderate, high and critical respectively. Further RUSLE factors has been broken into two categories, soil erosion susceptibility (A=RKLS), and soil erosion hazard (A=RKLSCP) have been computed. It is understood that functions of C and P are factors that can be controlled and thus can greatly reduce soil loss through management and conservational measures.

Geospatial Application of the Water Erosion Prediction Project (WEPP) Model

Abstract. At the hillslope profile and/or field scale, a simple Windows graphical user interface (GUI) is available to easily specify the slope, soil, and management inputs for application of the USDA Water Erosion Prediction Project (WEPP) model. Likewise, basic small watershed configurations of a few hillslopes and channels can be created and simulated with this GUI. However, as the catchment size increases, the complexity of developing and organizing all WEPP model inputs greatly increases due to the multitude of potential variations in topography, soils, and land management practices. For these types of situations, numerical approaches and special user interfaces have been developed to allow for easier WEPP model setup, utilizing either publicly available or user-specific geospatial information, e.g., digital elevation models (DEMs), geographic information system (GIS) soil data layers, and GIS land use/land cover data layers. We utilize the Topographic Parameterization (TOPAZ) digital landscape analysis tool for channel, watershed, and subcatchment delineation and to derive slope inputs for each of the subcatchment hillslope profiles and channels. A user has the option of specifying a single soil and land management for each subcatchment or utilizing the information in soils and land use/land cover GIS data layers to automatically assign those values for each grid cell. Once WEPP model runs are completed, the output data are analyzed, results interpreted, and maps of spatial soil loss and sediment yields are generated and visualized in a GIS. These procedures have been used within a number of GIS platforms including GeoWEPP, an ArcView/ArcGIS extension that was the first geospatial interface to be developed in 2001. GeoWEPP allows experienced GIS users the ability to import and utilize their own detailed DEM, soil, and/or land use/land cover information or to access publicly available spatial datasets. A web-based GIS system that used MapServer web GIS software for handling and displaying the spatial data and model results was initially released in 2004. Most recently, Google Maps and OpenLayers technologies have been integrated into the web WEPP GIS software to provide significant enhancements. This article discusses in detail the logic and procedures for developing the WEPP model inputs, the various WEPP GIS interfaces, and provides example real-world geospatial WEPP applications. Further work is ongoing in order to expand these tools to allow users to customize their own inputs via the internet and to link the desktop GeoWEPP with the web-based GIS system.

Estimation of soil erosion risk within a small mountainous sub-watershed in Kerala, India, using Revised Universal Soil Loss Equation (RUSLE) and geo-information technology

A comprehensive methodology that integrates Revised Universal Soil Loss Equation (RUSLE) model and Geographic Information System (GIS) techniques was adopted to determine the soil erosion vulnerability of a forested mountainous sub-watershed in Kerala, India. The spatial pattern of annual soil erosion rate was obtained by integrating geo-environmental variables in a raster based GIS method. GIS data layers including, rainfall erosivity (R), soil erodability (K), slope length and steepness (LS), cover management (C) and conservation practice (P) factors were computed to determine their effects on average annual soil loss in the area. The resultant map of annual soil erosion shows a maximum soil loss of 17.73t h-1 y-1 with a close relation to grass land areas, degraded forests and deciduous forests on the steep side-slopes (with high LS). The spatial erosion maps generated with RUSLE method and GIS can serve as effective inputs in deriving strategies for land planning and management in the environmentally sensitive mountainous areas.

Geographic information systems database, Walnut Gulch Experimental Watershed, Arizona, United States

The geographic information systems (GIS) database complementing the Walnut Gulch Experimental Watershed (WGEW) special section papers in this issue of Water Resources Research is described. Spatial data layers discussed here will be especially useful to modelers interested in simulating the spatial and temporal characteristics of rainfall, runoff, erosion, and sedimentation processes on the WGEW. All data are available as either images or individual GIS data layers (vector or raster format) via the U.S. Department of Agriculture Agricultural Research Service Southwest Watershed Research Center at http://www.tucson.ars.ag.gov/dap/. Standard metadata are provided with attending projection information and restrictions on use.

Epidemiologic Analysis of Factors Associated with Local Disappearances of Native Ranid Frogs in Arizona

We examined factors that may independently or synergistically contribute to amphibian population declines. We used epidemiologic case-control methodology to sample and analyze a large database developed and maintained by the Arizona Game and Fish Department that describes historical and currently known ranid frog localities in Arizona, U.S.A. Sites with historical documentation of target ranid species (n= 324) were evaluated to identify locations where frogs had disappeared during the study period (case sites) and locations where frog populations persisted (control sites). Between 1986 and 2003, 117 (36%) of the 324 sites became case sites, of which 105 were used in the analyses. An equal number of control sites were sampled to control for the effects of time. Risk factors, or predictor variables, were defined from environmental data summarized during site surveys and geographic information system data layers. We evaluated risk factors with univariate and multifactorial logistic-regression analyses to derive odds ratios (OR). Odds for local population disappearance were significantly related to 4 factors in the multifactorial model. Disappearance of frog populations increased with increasing elevation (OR = 2.7 for every 500 m, p < 0.01). Sites where disappearances occurred were 4.3 times more likely to have other nearby sites that also experienced disappearances (OR = 4.3, p < 0.01), whereas the odds of disappearance were 6.7 times less (OR = 0.15, p < 0.01) when there was a source population nearby. Sites with disappearances were 2.6 times more likely to have introduced crayfish than were control sites (OR = 2.6, p= 0.04). The identification of factors associated with frog disappearances increases understanding of declines occurring in natural populations and aids in conservation efforts to reestablish and protect native ranids by identifying and prioritizing implicated threats.

Spatial Modeling of Biological Soil Crusts to Support Rangeland Assessment and Monitoring

Biological soil crusts are a diverse soil surface community, prevalent in semiarid regions, which function as ecosystem engineers and perform numerous important ecosystem services. Loss of crusts has been implicated as a factor leading to accelerated soil erosion and other forms of land degradation. To support assessment and monitoring efforts aimed at ensuring the sustainability of rangeland ecosystems, managers require spatially explicit information concerning potential cover and composition of biological soil crusts. We sampled low disturbance sites in Grand Staircase–Escalante National Monument (Utah, USA) to determine the feasibility of modeling the potential cover and composition of biological soil crusts in a large area. We used classification and regression trees to model cover of four crust types (light cyanobacterial, dark cyanobacterial, moss, lichen) and 1 cyanobacterial biomass proxy (chlorophyll a), based upon a parsimonious set of GIS (Geographic Information Systems) data layers (soil types, precipitation, and elevation). Soil type was consistently the best predictor, although elevation and precipitation were both invoked in the various models. Predicted and observed values for the dark cyanobacterial, moss, and lichen models corresponded moderately well (R 2 ¼ 0.49, 0.64, 0.55, respectively). Cover of late successional crust elements (moss þ lichen þ dark cyanobacterial) was also successfully modeled (R 2 ¼ 0.64). We were less successful with models of light cyanobacterial cover (R 2 ¼ 0.22) and chlorophyll a (R 2 ¼ 0.09). We believe that our difficulty modeling chlorophyll a concentration is related to a severe drought and subsequent cyanobacterial mortality during the course of the study. These models provide the necessary reference conditions to facilitate the comparison between the actual cover and composition of biological soil crusts at a given site and their potential cover and composition condition so that sites in poor condition can be identified and management actions can be taken.

Seasonal Use and Selection of Caves by Plethodontid Salamanders in a Karst Area of Arkansas

From December 1998 through May 2000, seasonal use and selection of caves by salamanders were documented in 93 small caves in Crawford County, Arkansas. Caves were surveyed seasonally between winter 1998–1999 and spring 2000, and the numbers and species of all salamanders present were documented. Cave ambient temperature and relative humidity were recorded. Also, each cave entrance was georeferenced using GPS (Global Positioning System) and landscape-level variables around each entrance were quantified using ArcView GIS (Geographic Information System) data layers. Using salamander and ArcView data, relations between salamander occurrence based upon landscape-level variables (dominant overstory vegetation, geology, slope, aspect, solar radiation and distance to perennial stream) and cave characteristics (ambient temperature, relative humidity, cave length and entrance size) were examined. Six species of salamanders were found during the survey. Eurycea lucifuga (cave salamander), Plethodon albagula (western slimy salamander) and P. angusticlavius (Ozark salamander) were commonly observed. Eurycea lucifuga salamanders were active in caves during spring, summer and autumn. Plethodon albagula were most frequently encountered during the summer and P. angusticlavius were most frequently encountered during spring surveys. Few salamanders of any species were detected during winter surveys. The use of caves by the various species of salamanders was influenced by the landscape-level variables and cave characteristics. However, cave ambient temperature and relative humidity appear to have the most influences of salamander use of caves. Both E. lucifuga and P. albagula were significantly more likely to be found in caves with cooler temperatures in summer and higher relative humidities in autumn. In addition to these factors, the probability of finding E. lucifuga significantly decreased with increasing distance to permanent streams, whereas P. albagula was significantly more likely to be found in caves with south and west facing aspects, especially during summer and autumn surveys. There were no significant patterns for P. angusticlavius use of caves. Seasonal and spatial patterns of cave use by salamanders were primarily influenced by environmental (temperature and relative humidity) factors and the ecological requirements of the salamanders.

Spatial Modeling of Biological Soil Crusts to Support Rangeland Assessment and Monitoring

Biological soil crusts are a diverse soil surface community, prevalent in semiarid regions, which function as ecosystem engineers and perform numerous important ecosystem services. Loss of crusts has been implicated as a factor leading to accelerated soil erosion and other forms of land degradation. To support assessment and monitoring efforts aimed at ensuring the sustainability of rangeland ecosystems, managers require spatially explicit information concerning potential cover and composition of biological soil crusts. We sampled low disturbance sites in Grand Staircase-Escalante National Monument (Utah, USA) to determine the feasibility of modeling the potential cover and composition of biological soil crusts in a large area. We used classification and regression trees to model cover of four crust types (light cyanobacterial, dark cyanobacterial, moss, lichen) and 1 cyanobacterial biomass proxy (chlorophyll a), based upon a parsimonious set of GIS (Geographic Information Systems) data layers (soil types, precipitation, and elevation). Soil type was consistently the best predictor, although elevation and precipitation were both invoked in the various models. Predicted and observed values for the dark cyanobacterial, moss, and lichen models corresponded moderately well (R2 = 0.49, 0.64, 0.55, respectively). Cover of late successional crust elements (moss + lichen + dark cyanobacterial) was also successfully modeled (R2 = 0.64). We were less successful with models of light cyanobacterial cover (R2 = 0.22) and chlorophyll a (R2 = 0.09). We believe that our difficulty modeling chlorophyll a concentration is related to a severe drought and subsequent cyanobacterial mortality during the course of the study. These models provide the necessary reference conditions to facilitate the comparison between the actual cover and composition of biological soil crusts at a given site and their potential cover and composition condition so that sites in poor condition can be identified and management actions can be taken.

The development of a spatially explicit model to estimate radiocaesium body burdens in raccoons ( Procyon lotor) for ecological risk assessment

A spatially explicit model of raccoon ( Procyon lotor) distribution for the U.S. Department of Energy's (DOE) Savannah River Site (SRS) in west-central South Carolina was developed using data from a raccoon radio-telemetry study and visualized within a Geographic Information System (GIS). An inductive approach was employed to develop three sub-models using the ecological requirements of raccoons studied in the following habitats: (1) man-made reservoirs, (2) bottomland hardwood/riverine systems, and (3) isolated wetland systems. Logistic regression was used to derive probabilistic resource selection functions using habitat compositional data and landscape metrics. The final distribution model provides a spatially explicit probability (likelihood of being in an area) surface for male raccoons. The model is a stand-alone tool consisting of algorithms independent of the specific GIS data layers to which they were derived. The model was then used to predict contaminant burdens in raccoons inhabiting a riverine system contaminated with radiocaesium ( 137Cs). The predicted 137Cs burdens were less than if one would assume homogeneous use of the contaminated areas. This modelling effort provides a template for DOE managed lands and other large government facilities to establish a framework for site-specific ecological assessments that use wildlife species as endpoints.

Using Wildlife as Receptor Species: A Landscape Approach to Ecological Risk Assessment

To assist risk assessors at the Department of Energy's Savannah River Site (SRS), a Geographic Information System (GIS) application was developed to provide relevant information about specific receptor species of resident wildlife that can be used for ecological risk assessment. Information was obtained from an extensive literature review of publications and reports on vertebrate- and contaminant-related research since 1954 and linked to a GIS. Although this GIS is a useful tool for risk assessors because the data quality is high, it does not describe the species' site-wide spatial distribution or life history, which may be crucial when developing a risk assessment. Specific receptor species on the SRS were modeled to provide an estimate of an overall distribution (probability of being in an area). Each model is a stand-alone tool consisting of algorithms independent of the GIS data layers to which it is applied and therefore is dynamic and will respond to changes such as habitat disturbances and natural succession. This paper describes this modeling process and demonstrates how these resource selection models can then be used to produce spatially explicit exposure estimates. This approach is a template for other large federal facilities to establish a framework for site-specific risk assessments that use wildlife species as endpoints.

Landscape characteristics associated with species richness and occurrence of small native mammals inhabiting a coastal heathland: a spatial modelling approach

In the coastal region of south-western Victoria, Australia, populations of native small mammal species are restricted to patches of suitable habitat in a highly fragmented landscape. The size and spatial arrangement of these patches is likely to influence both the occupancy and richness of species at a location. Geographic Information System (GIS)-based habitat models of the species richness of native small mammals, and individual species occurrences, were developed to produce maps displaying the spatial configuration of suitable habitat. Models were generated using either generalised linear Poisson regression (for species richness) or logistic regression (for species occurrences) with species richness or presence/absence as the dependent variable and landscape variables, extracted from both GIS data layers and multi-spectral digital imagery, as the predictor variables. A multi-model inference approach based on the Akaike Information Criterion was used and the resulting model was applied in a GIS framework to extrapolate predicted richness/likelihood of occurrence across the entire area of the study. A negative association between species richness and elevation, habitat complexity and sun index indicated that richness within the study area decreases with increasing altitude, vertical vegetation structure and exposure to solar radiation. Landform characteristics were important (to varying degrees) in determining habitat occupancy for all of the species examined, while the influence of habitat complexity was important for only one of the species. Performance of all but one of the models generated using presence/absence data was high, as indicated by the area under the curve of a receiver–operating characteristic plot. The effective conservation of the small mammal species in the area of concern is likely to depend on management actions that promote the protection of the critical habitats identified in the models.

Spatial prediction of rufous bristlebird habitat in a coastal heathland: a GIS‐based approach

Summary To develop a conservation management plan for a species, knowledge of its distribution and spatial arrangement of preferred habitat is essential. This is a difficult task, especially when the species of concern is in low abundance. In south‐western Victoria, Australia, populations of the rare rufous bristlebird Dasyornis broadbenti are threatened by fragmentation of suitable habitat. In order to improve the conservation status of this species, critical habitat requirements must be identified and a system of corridors must be established to link known populations. A predictive spatial model of rufous bristlebird habitat was developed in order to identify critical areas requiring preservation, such as corridors for dispersal. Habitat models generated using generalized linear modelling techniques can assist in delineating the specific habitat requirements of a species. Coupled with geographic information system (GIS) technology, these models can be extrapolated to produce maps displaying the spatial configuration of suitable habitat. Models were generated using logistic regression, with bristlebird presence or absence as the dependent variable and landscape variables, extracted from both GIS data layers and multispectral digital imagery, as the predictors. A multimodel inference approach based on Akaike's information criterion was used and the resulting model was applied in a GIS to extrapolate predicted likelihood of occurrence across the entire area of concern. The predictive performance of the selected model was evaluated using the receiver operating characteristic (ROC) technique. A hierarchical partitioning protocol was used to identify the predictor variables most likely to influence variation in the dependent variable. Probability of species presence was used as an index of habitat suitability. Negative associations between rufous bristlebird presence and increasing elevation, ‘distance to creek’, ‘distance to coast’ and sun index were evident, suggesting a preference for areas relatively low in altitude, in close proximity to the coastal fringe and drainage lines, and receiving less direct sunlight. A positive association with increasing habitat complexity also suggested that this species prefers areas containing high vertical density of vegetation. The predictive performance of the selected model was shown to be high (area under the curve 0·97), indicating a good fit of the model to the data. Hierarchical partitioning analysis showed that all the variables considered had significant independent contributions towards explaining the variation in the dependent variable. The proportion of the total study area that was predicted as suitable habitat for the rufous bristlebird (using probability of occurrence at a ≥ 0·5 level) was 16%. Synthesis and applications. The spatial model clearly delineated areas predicted as highly suitable rufous bristlebird habitat, with evidence of potential corridors linking coastal and inland populations via gullies. Conservation of this species will depend on management actions that protect the critical habitats identified in the model. A multiscale approach to the modelling process is recommended whereby a spatially explicit model is first generated using landscape variables extracted from a GIS, and a second model at site level is developed using fine‐scale habitat variables measured on the ground. Where there are constraints on the time and cost involved in measuring finer scale variables, the first step alone can be used for conservation planning.

USING DISCRIMINANT ANALYSIS AND GIS TO DELINEATE SUBSURFACE DRAINAGE PATTERNS

The contamination of soil and water resources from nutrients, transported in subsurface drainage water havingdifferent drainage patterns, has important repercussions on the ecological environment and human health. This study wasdesigned to delineate subsurface drainage patterns using cluster analysis based on six years (1993 to 1998) of field measureddata on subsurface drainage flows from thirty-six 0.4 ha field experimental plots. These drainage patterns then were relatedspatially to the soil and topographic attributes using discriminant analysis and the map overlay capability of GeographicInformation Systems (GIS) to develop cause-effect relationships. The experimental field plots, under various tillage and nitrogenmanagement treatments, were located on glacial till derived soils at Iowa State Universitys Northeastern Research Centernear Nashua, Iowa. The field-measured subsurface drainage volumes were normalized to make comparisons over all plotsand years, and the normalized data were used in the subsequent statistical and GIS analyses. After performing cluster analysis,the output was generated as GIS data layers showing low, medium, and high drainage areas. Stepwise discriminant analysisidentified elevation, slope, and average normalized yield as the factors contributing significantly (P < 0.10) to theformation of subsurface drainage zones. GIS data layers of the factors, identified during discriminant analysis, were overlaidon the drainage patterns to study the spatial relationships. Map overlay analysis showed that high drainage areas were consistentlyfound at low elevation levels in the vicinity of Floyd soils over the 6-year study period. The combined use of discriminantanalysis and GIS was found to be effective in delineating subsurface drainage zones so that appropriate managementpractices can be applied to mitigate the environmental effects resulting from medium and higher subsurface drainage effluents.