Accuracy of landsat-TM and GIS rule-based methods for forest wetland classification in Maine

https://doi.org/10.1016/0034-4257(95)00085-FGet rights and content

Abstract

An investigation was undertaken to compare satellite image classification techniques to delineate forest wetlands in Maine. Four classification techniques were compared, including a GIS rule-based model. Accuracy assessments of the four methods on two study sites, Orono and Acadia, revealed very similar results. Overall accuracy for four super groups (forest wetland, other wetland, forest upland, other upland) ranged from 72% to 81% at Orono and 74% to 82% at Acadia. Pairwise significance tests indicated that the GIS model was significantly better than unsupervised classification at both study sites, and significantly better than tasseled cap (Acadia) in classifying the four super groups. Although Kappa coefficients were slightly higher for the GIS model compared to hybrid classification, there was no significant difference between the two methods at either study site. Forest wetland user's and producer's accuracy was in the 80% range for the highest accuracy achieved either by the GIS model or hybrid classification. Hydric soils, National Wetland Inventory data, and slope percentage were the most important variables in the GIS model. From this study, it appears that a combination of hybrid and GIS rule-based classification methods are the most promising for further investigations of forest wetland delineation.

References (34)

  • R.W. Tiner

    Use of high-altitude aerial photography for inventorying forested wetlands in the United States

    Forest Ecol. Management

    (1990)
  • D.E. Ahl

    A Comparison of Satellite and GIS Classification Techniques for Delineating Forested Wetlands

    (1994)
  • J.R. Anderson et al.

    A land use and land cover classification system for use with remote sensor data

    U.S. Geological Survey Professional Paper 964

    (1976)
  • P.V. Bolstad et al.

    Rule-based classification models: flexible integration of satellite imagery and thematic spatial data

    Photogram. Engineering Remote Sens.

    (1992)
  • J.B. Campbell

    Introduction to Remote Sensing

    (1987)
  • D.L. Civco

    Knowledge-based land use and land cover mapping

    Tech. Papers, 1989 ACSM-ASPRS Annual Convention

    (1989)
  • J. Cohen

    A coefficient of agreement for nominal scales

    Educational Psychol. Measur.

    (1960)
  • R.G. Congalton et al.

    A quantitative method to test for consistency and correctness in photo-interpretation

    Photogram. Engineering Remote Sens.

    (1983)
  • R.G. Congalton et al.

    Accuracy of remotely sensed data: sampling and analysis procedures

  • R.G. Congalton et al.

    Assessing Landsat classification accuracy using discrete multivariate analysis statistical techniques

    Photogram. Engineering Remote Sens.

    (1983)
  • L.M. Cowardin et al.

    A Classification of Wetlands and Deepwater Habitats of the United States

  • E.P. Crist et al.

    A physically based transformation of thematic mapper data—the TM tasseled cap

    IEEE Trans. Geosci. Remote Sens.

    (1984)
  • E.P. Crist et al.

    Application of the tasseled cap concept to simulated thematic mapper data

    Photogram. Engineering Remote Sens.

    (1984)
  • J.E. Dobson et al.

    Coastwatch—detecting change in coastal wetlands

    Geo. Info Systems

    (1991)
  • ERDAS

    ERDAS PC Field Guide

    (1991)
  • ESRI

    PC ARC/INFO Starter Kit

    (1990)
  • Federal Geographic Data Committee

    Application of satellite data for mapping and monitoring wetlands

    Wetlands Subcommittee Technical Report 1

    (1992)
  • Cited by (152)

    • Impact of railways on land use and land cover change: Evidence from India

      2023, Transportation Research Part D: Transport and Environment
    • Classifying vegetation communities karst wetland synergistic use of image fusion and object-based machine learning algorithm with Jilin-1 and UAV multispectral images

      2022, Ecological Indicators
      Citation Excerpt :

      This paper used the testing samples data to quantitatively evaluate classification accuracy of different schemes and classifiers for mapping karst wetland vegetation communities. At the 95% confidence interval, the confusion matrix was established for nine classification schemes to calculate the overall accuracy and kappa coefficient, respectively (Sader et al., 1995). The user’s and producer’s accuracy, average accuracy (AA, the average value of user accuracy and producer accuracy) (Mui et al., 2015; Mahdavi et al., 2019) were also calculated to evaluate accuracy differences of vegetation communities.

    View all citing articles on Scopus

    This research was supported by NOAA Coastal Ocean Program, Coastwatch Change Analysis Program under Grant #NA26RG0420-01 to the Sea Grant Program of the University of Maine.

    1

    The authors extend their thanks to Dr. Robert Wall, Director of the UM Sea Grant Program, and his staff for administrative support. Cindy Paschal provided timely and efficient support in manuscript preparation, which was greatly appreciated. Constructive comments from anonymous reviewers also were appreciated and incorporated in the final manuscript. Maine Agriculture and Forestry Experiment Station External Publication #1882.

    View full text