PE&RS November 1996

VOLUME 62, NUMBER 11
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
JOURNAL OF THE AMERICAN SOCIETY FOR PHOTOGRAMMETRY AND REMOTE SENSING

Peer-Reviewed Articles

1245 An Automated Method for Digitizing Color Thematic Maps
Rick L. Lawrence, Joseph E. Means, and William J. Ripple

Abstract
Download Full Article
There is an increasing need for methods of rapidly entering analog data into geographic information systems. Traditional methods of hand digitizing or hand tracing followed by scanning are costly and time consuming. The authors have developed a rapid, easy to use method for digitizing color thematic maps that makes use of standard image processing techniques. The method uses a digital camera followed by supervised spectral classification and post- classification smoothing. Although overall accuracy for an extremely challenging test map was 93%, results indicate that for most applications expected accuracy is very high. 

1249 Mapping Ecological Land Systems and Classification Uncertainties from Digital Elevation and Forest-Cover Data Using Neural Networks
P. Gong, R. Pu,and J. Chen

Abstract
Download Full Article
Emphasizes mainly the technical aspects of the land-systems classification problem with neural networks. Using digital elevation, its derivatives, and forest cover data as input, the authors constructed neural networks to classify 27 land- system classes at Duck Mountain, Manitoba, Canada. Training and testing of those neural networks were done using an existing land-systems map prepared through airphoto interpretation and field studies. Two types of data structure were evaluated: polygon and raster forms. Types of data sets contained the elevation, slope, aspect, dominant forest species and corresponding crown closures, and more general site information on cover type, subtype, site, cutting class, and crown closure. Results indicate that a random sampling strategy for training sample selection led to better classification results than a contiguous sampling method. Approximately 10% of the total samples were sufficient for network training. The best overall classification accuracy was 52.0% when the neural network classification result was compared with the existing map. 

1261 Spatial and Compositional Pattern of Alpine Treeline, Glacier National Park, Montana
Thomas R. Allen and Stephen J. Walsh

Abstract
Download Full Article
Quantifies the complex patterns of alpine treeline across an extensive area of Glacier National Park, Montana. Satellite image classification, digital terrain modeling, and GIS measurements of landscape structure provided important tools for the analysis. The study area was topographically partitioned into watersheds and hillslope units in which to measure treeline patterns. Cluster analysis of selected spatial and compositional pattern metrics was used to infer major alpine treeline forms. Six significant treeline types were differentiated using patch richness, contagion, contrast, number of patches, fractal dimension, relative edge density, and forest-tundra juxtaposition. Clusters were validated using split-sample replication and discriminant analysis. Patterns were found to differ among types of terrain, affirming hypothesized sensitivities to topoclimatic gradients, natural disturbances, and geologic substrate. 

1269 Predicting Rare Orchid (Small Whorled Pogonia) Habitat Using GIS
Molly B. Sperduto and Russell G. Congalton

Abstract
Download Full Article
Iostria medeoloides (Pursh) Raf., commonly known as the small whorled pogonia, is the rarest orchid in eastern North America, north of Florida. A geographical information system (GIS) was used to facilitate locating potential habitat for Isotria medeoloides in New Hampshire and Maine. Two predictive GIS overlay models were developed: an equal weight model and a chi-square model. As a result of the chi-square evaluation, the following general characteristics were determined to be positively associated with small whorled pogonia sites: soils with a pan layer, percent slopes between 11 and 17%, and a digital reflectance greater than 68 for the near infra-red band of Landsat Thematic Mapper data. Nine previously undiscovered populations of Isotria medeoloides were found. The GIS provided a useful tool for combining ecological habitat characteristics to successfully predict potential habitat for Isotria medeoloides

1281 Application of a Modified Habitat Suitability Index Model for Moose
Jeffrey A. Hepinstall, LLoyd P. Queen, and Peter A. Jordan

Abstract
Download Full Article
This paper explores alternative approaches for calculating moose Habitat Suitability Index (HSI) values using a GIS. We modified an existing moose HSI model and implemented it using moving windows and various boundary value estimation methods. The habitat window and boundary analyses indicate that a 50% window overlap is sufficient to capture variation in the landscape. A mirror data set for areas outside the study area, used to estimate boundary habitat values from a sample grid within a vector GIS, is proposed as a useful alternative for supporting landscape-scale resource management. 

1287 Using Genetic Learning Neural Networks for Spatial Decision Making in GIS
Jiang Zhou and Daniel L. Civco

Abstract
Download Full Article
Traditional approaches for suitability analysis in GIS are overlay and the more complicated multicriteria evaluation (MCE). Despite being widely used, these methods have at least three problems: difficulties in handling spatial data possessing inaccuracy; multiple measurement scales, and factor interdependency; requirements of prior knowledge in identifying criteria, assigning scores, determining criteria preference, and selecting aggregation functions; and typically, an 'unfriendly' user interface. To solve these problems, in this paper a neural network approach is presented. The neural network uses a genetic algorithm as its learning mechanism. A set of experiments revealed that the aforementioned difficulties are overcome by the evolutionary learning of neural networks. Genetic learning neural networks can provide an alternative for and improvement over traditional suitability analysis methods in GIS. 

1297 Extending the Applicability of Viewsheds in Landscape Planning
Peter F. Fisher

Abstract
Download Full Article
The determination of the visible area or viewshed from a viewing point looking out on a landscape is a widely available function in a GIS. A reconsideration of the queries which may be made of the viewshed, however, revals that often the function does not address them correctly. This has led to the specification of alternative viewshed functions intended to give flexible outcomes which can be used to respond to the queries directly. The alternatives include the horizons viewshed, the local offset viewshed, the global offset viewshed, and reverse viewing versions of all three. Applications of these alternative viewshed functions to answer queries about the landscape and the view which the binary viewshed is not able to respond to either precisely or flexibly are examined. 

1303 A Comprehensive Managed Areas Spatial Database for the Conterminous United States
R. Gavin McGhie, Joseph Scepan,and John E. Estes

Abstract
Download Full Article
The compilation of a comprehensive, spatially referenced, digital database of managed areas in the conterminous United States is described. As concern over ecosystem degradation increases, so does the need for accurate, up-to-date information on the spatial location and aerial extent of currently managed and protected areas. This need represents the fundamental motivation for creation of this Managed Areas Database (MAD). MAD includes information on the level of protection each management designation provides, sources used for compilation, and a number of additional attributes. MAD can be used with supplementary data sets for conservation planning, and to determine protection status. The authors believe that this database can and will support a wide variety of environmental studies.