Abstract
Methods were developed to evaluate the performance of a decision-tree model used to predict landscape-level patterns of potential forest vegetation in central New York State. The model integrated environmental databases and knowledge on distribution of vegetation. Soil and terrain decision-tree variables were derived by processing state-wide soil geographic databases and digital terrain data. Variables used as model inputs were soil parent material, soil drainage, soil acidity, slope position, slope gradient, and slope azimuth. Landscapescale maps of potential vegetation were derived through sequential map overlay operations using a geographic information system (GIS). A verification sample of 276 field plots was analyzed to determine: (1) agreement between GIS-derived estimates of decision-tree variables and direct field measurements, (2) agreement between vegetation distributions predicted using GIS-derived estimates and using field observations, (3) effect of misclassification costs on prediction agreement, (4) influence of particular environmental variables on model predictions, and (5) misclassification rates of the decision-tree model. Results indicate that the prediction model was most sensitive to drainage and slope gradient, and that the imprecision of the input data led to a high frequency of incorrect predictions of vegetation. However, in many cases of misclassification the predicted vegetation was similar to that of the field plots so that the cost of errors was less than expected from the misclassification rate alone. Moreover, since common vegetation types were more accurately predicted than rare types, the model appears to be reasonably good at predicting vegetation for a randomly selected plot in the landscape. The error assessment methodology developed for this study provides a useful approach for determining the accuracy and sensitivity of landscape-scale environmental models, and indicates the need to develop appropriate field sampling procedures for verifying the predictions of such models.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Aicken, M. 1990. Maximum likelihood estimation of agreement in the constant predictive probability model and its relation to Cohen's kappa. Biometrics 46: 293–302.
Brennan, R.L. and Prediger, D.J. 1981. Coefficient kappa: some uses, misuses and alternatives. Ed. and Psych. Meas. 41: 687–699.
Burrough, P.A. 1986. Principles of geographical information systems for land resources assessment. Clarendon Press, Oxford. 194p.
Cline, M.G. and Marshall, R.L. 1977. Soils of New York Landscapes Information Bulletin #119. Cornell Cooperative Extension. Cornell University. Ithaca, New York. 62p.
Cohen, J. 1960. A coefficient of agreement for nominal scales. Educational and Psychological Measurement 20: 37–46.
ERDAS, Inc. 1991. Field Guide, Version 7.5. ERDAS, Inc. Atlanta, Georgia.
Fleiss, J. 1981. Statistical Methods for Rates and Proportions. New York: John Wiley & Sons, Inc. 321p.
Gardner, R.H., Huff, D.D., O'Neill, R.V., Mankin, J.B., Carney, J. and Jones, J. 1980. Applications of error analysis to a marsh hydrology model. Water Resources Research 16(4): 659–664.
Goodchild, M. and Gopal, S. 1989. Accuracy of Spatial Databases. New York: Taylor & Francis. 290p.
Goodchild, M., Guoqing, S. and Shiren, Y. 1992. Development and test of an error model for categorical data. Int. J. Geographical Information Systems 6(2): 87–104.
Heuvelink, G.B.M. 1993. Error propagation in quantitative spatial modelling: Applications in geographical information systems. Netherlands Geographic Studies #163. Univ. Utrecht, the Netherlands, 295p.
Hill, M.O., Bunce, R.G.H. and Shaw, M.W. 1975. Indicator species analysis, a divisive polythetic method of classification, and its application to a survey of native pinewoods in Scotland. J. Ecol. 63: 597–613.
Huenneke, L.E. 1982. Wetland forests of Tompkins County, New York. Bull. Torrey Bot. Club 109: 51–63.
Hutton, F.Z. Jr. 1971. Soil survey of Cayuga County, New York. USDA Soil Conservation Service. U.S. Govt. Printing Office. Washington, D.C. 205p.
Hutton, F.Z. Jr. 1972. Soil survey of Seneca County, New York. USDA Soil Conservation Service, U.S. Govt. Printing Office. Washington, DC. 143p.
Lewin, D.C. 1974. The vegetation of the ravines of the southern Finger Lakes, New York, region. Am. Midl. Nat. 91: 315–342.
Marks, P.L. and Gardescu, S. 1992. Vegetation of the central Finger Lakes Region of New York in 1790s. New York State Museum Bulletin #484 1–35. Albany, New York.
Mohler, C.L. 1991. Plant community types of the central Finger Lakes Region of New York: A synopsis and key. Proc. Rochester Acad. Sci. 17(2): 55–103.
Mohler, C.L. and Marks, P.L. Vegetation-environment relations in the central Finger Lakes Region of New York. (In Preparation).
Moore, D.M., Lees, B.G. and Davey, S.M. 1991. A new method for predicting vegetation distributions using decision tree analysis in a geographic information system. Environmental Management 15: 59–71.
Neeley, J.A. 1965. Soil survey of Tompkins County, New York. USDA Soil Conservation Service. U.S. Govt. Printing Office, Washington, DC. 241p.
O'Neill, R.V. 1973. Error analysis of ecological models. In Radionuclides in Ecosystems. pp. 898–908. Edited by D.J. Nelson. CONF-710501. National Technical Information Service, Springfield, Virginia, U.S.A.
Puglia, S.P. 1979. Soil survey of Schuyler County, New York. USDA Soil Conservation Service. U.S. Printing Office. Washington, DC. 192p.
Rastetter, E.B., King, A.W., Cosby, B.J., Hornberger, G.M., O'Neill, R.V. and Hobbie, J.E. 1992. Aggregating fine-scale ecological knowledge to model coarser-scale attributes of ecosystems. Ecological Applications 2(1): 55–70.
Reybold, W.U. and TeSelle, G.W. 1989. Soil geographic databases. J. Soil and Water Cons. 44: 28–29.
Rossi, R.E., Borth, P.W. and Tollefson, J.J. 1993. Stochastic simulation for characterizing ecological spatial patterns and appraising risk. Ecological Applications 3(4): 719–735.
Saltelli, A., Andres, T.H. and Homma, T. 1993. Sensitivity analysis of model output: an investigation of new techniques. Computational Statistics & Data Analysis 15: 211–238.
Seischab, F.K. 1985. An analysis of the forests of the Bristol Hills of New York. Am. Midl. Nat. 114: 77–83.
Seischab, F.K. 1990. Presettlement forests of the Phelps and Gorham Purchase in western New York. Bull. Torrey Bot. Club 117: 27–38.
Smith, B.E., Marks, P.L. and Gardescu, S. 1993. Two hundred years of forest cover changes in Tompkins County, New York. Bull. Torrey Botanical Club. 120: 229–247.
Snyder, J.P. 1982. Map projections used by the U.S. Geological Survey. Geol. Survey Bulletin #1532. U.S. Government Printing Office, Washington, D.C.
Soil Conservation Service. 1991. State Soil Geographic Data Base (STATSGO). Misc. Publ. #1492. U.S. Dept. Agriculture. Washington, DC. 88p.
Soil Conservation Service. 1981. Land resource regions and major land resource areas of the United States. USDA Soil Conservation Service Agric. Handbook #296. Washington, DC. 156p.
U.S. Geological Survey. 1987. Digital Elevation Models. Data Users' Guide #5. U.S. Geological Survey. Reston, Virginia. 38p.
Whittaker, R.H. 1960. Vegetation of the Siskiyou Mountains, Oregon and California. Ecol. Monogr. 30: 279–338.
Wilding, L.P., Smeck, N.E. and Hall, G.F. (ed.) 1983. Pedogenesis and soil taxonomy. I. Concepts and interactions. Elsevier, New York.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Lynn, H., Mohler, C.L., DeGloria, S.D. et al. Error assessment in decision-tree models applied to vegetation analysis. Landscape Ecol 10, 323–335 (1995). https://doi.org/10.1007/BF00130210
Issue Date:
DOI: https://doi.org/10.1007/BF00130210