NWI maps (onlv wetland tvpes). It is uncertain whether a similar 
modeling approach would be more successful in identifving 
forested wetlands compared to conventional methods (e.g. hybrid) 
in other states or regions of U.S.A.. 
The conventional classifications were based on class grouping 
derived from signature evaluation. The integrated model was 
based on class grouping derived from weighted GIS layer 
combinations. The analysis matrix technique can be used to 
analvze a multivariate discrete data set for optimal aggregation to 
category assignment, For example, an alternative approach can 
be simulated by analyzing spectral signatures derived from 
clustering or training samples to guide appropriate land cover 
class assignment. A reference data set is required to apply the 
analysis matrix technique. Consequently, the results avoid 
subjective signature evaluation and benefits from the formulated 
algorithm derived in the integrated model. 
The integrated model approach will require further research and 
development efforts. There are many important research issues 
related to the acquisition, processing, and joint analysis of remote 
sensing and GIS. Error is one of the primary issues of concern in 
the integrated model. For example, the geometric accuracy of the 
collective GIS layers is probably worse than the corrected 
Landsat TM data. Integration of remote sensing and GIS data 
required digitizing, converting, geo-referencing, and registration. 
These transformations have potential errors deeply embedded in 
the overall data processing flow. On the other hand, the current 
accuracy assessment procedures have been adapted from 
statistical procedures to give a quantity measure of overall 
accuracy. However, the quality (e.g. spatial distribution) error 
was not evaluated. Techniques need to be developed for 
assessing the spatial structure of error in the integrated model 
product. Further research into the integration of remote sensing 
and GIS along with improvements in spectral and spatial 
resolution of remotely sensed data may lead to advancements in 
wetland inventory and monitoring in the future. 
6.0 REFERENCES 
Abl, D.E, 1994, A Comparison of Satellite and GIS 
Classification Techniques for Delineating Forested 
Wetlands, Department of Forest Management, 
University of Maine, Orono, ME, pp. 51. 
Breiman, L., J.H. Friedman, R.A. Olshen, C.J. Stone, 1984, 
Classification and Regression Trees, Wadsworth, Inc., 
Belmont, California, pp. 18-171. 
Civco, D.L., and Y. Wang, 1994, Classification of Multispectral, 
Multitemporal, Multisource Spatial Data Using 
Artificial Neural Networks, ASPRS/ACSM Annual 
Convention & Exposition, Reno, Nevada, Volume One, 
pp. 123-133. 
Congalton, R.G., R.G. Oderwald, R.A. Mead, 1983, Assessing 
Landsat Classification Accuracy Using Discrete 
Multivariate Analysis Statistical Techniques, 
Photogrammetric Engineering and Remote Sensing, 
Vol. 49, No. 12, pp. 1671-1678. 
424 
Eastman, J.R., 1992, IDRISI, Technicial Reference Version 4) 
Clark University, IDRISI Production, Worcester 
Massachusetts. : 
Ferguson, RL, LL. Wood, and D. B. Graham, 193 
Monitoring Spatial Change in Seagrass Habitat with 
Aerial Photography, Photogrammetric Engineering and 
Remote Sensing Vol. 59, No. 5, pp. 1033-1038. 
Greene, W.H., 1992, LIMDEP User's Manual and Reference 
Guide, Version 6.0, Econometic Software, Inc 
Bellport, New York, pp. 423-441. 
Haddad, K.D., D.R. Ekberg, 1987, Potential of Landsat TM 
Imagery for Assessing the National Status and Trends 
of Coastal Wetlands, American Society of Civil 
Engineers, New York, Vol. 5, pp. 5182-5195, 
Hepner, G., T. Logan, N. Rittner and N. Bryant, 1990, Artificial 
Neural Network Classification Using A Minimum 
Training Set: Comparison to Conventional Supervised 
Classification, Photogrammetric Engineering and 
Remote Sensing, Vol. 56, No. 4, pp. 469-473. 
Jensen, J.R., S. Narumalani, O. Weatherbee, and H.E. Mackey, 
1993, Measurement of Seasonal and Yearly Catal 
and Waterlily Changes Using  Multidate SPOT 
Panchromatic data, Photogrammetric Engineering and 
Remote Sensing, Vol. 59, No. 2, pp. 519-525. 
Mitchell, L.C., et al. 1991, Wetland Value Assessment 
Methodology and Community Models, The 
Environmental Work Group, Coastal Wetland 
Planning, Protection, and Restoration Act Technical 
Committee, Lafayette, Louisiana: U.S. Fish and 
Wildlife Service. 
The Federal Geographic Data Committee, 1992, Wetlands 
Subcommittee, Application of Satellite Data for 
Mapping and Monitoring Wetlands, Technical Report 
+ 
Tiner Jr., R.W., 1990, Use of High-Altitude Aerial Photography 
for Inventorying Forested Wetlands in the United 
States, Forest Ecology and Management, Vol. 33, No 
34, pp. 593-604. 
Wei, Ji, J.B. Johnston, M.E. McNiff, and L.C. Mitchell, 1992, 
Knowledge-based GIS: An Expert System Approach 
for Managing Wetlands, Geo Info System, pp. 60-64. 
Wharton, C.H., W.M. Ketchens, E.C. Pendlton, and T.W. Sips, 
1982, The Ecology Bottomland Hardwood Swamps of 
the Southeast: A Community Profile. U.S. Fish and 
Wildlife Service, Biological Services Program, 
Washington D.C., FWS/OBS-81/37. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996