Classification and Ordination of Seral Plant Communities

Abstract

A conceptual model of secondary succession was tested with data from disturbed vegetation in theAgropyron/Poa habitat type using a combination of classification and ordination techniques. Individual stands were classifled into communities by an agglomerative method. Results of the Bray-Curtis polar ordination using three endpoint selection methods supported the validity of the model. The model is visualized as a solid cone in which all of the plant communities included in a habitat type are positioned relative to their degree of disturbance, inferring their probably secondary successional pattern within habitat types. Most of our nation's range vegetation is in some stage of secondary succession. One of the aims of modern resource management is to direct plant succession toward a desired seral stage in order to obtain maximum productivity or stability (Stoddart et al. 1975). Williams et al. (1969) point out that natural succession may be the most economical means, and in many inaccessible areas the only means, to restore the resource to a level of production that can be managed economically. In order to use secondary succession as a tool to attain this level, it is essential to have a better understanding of seral communities. Reviews on succession were made by Drury and Nisbet (1973) and Golley (1977). Tree age data are widely used in study of forest succession (Botkin et al. 1972; Zedler and Goff 1973; Horn 1975). Because age of perennial herbs is not determinable, the same approach cannot be applied to study secondary succession of grassland vegetation. This paper deals with testing of a conceptual model of secondary succession in which age of plant is not involved. The relatively recent availability of computer and multivariate analytical techniques adapted for vegetational analysis permit researchers to analyze and interpret large quantities of data. These techniques help reveal relationships of plant communities not previously possible. Reviews in use of multivariate techniques as applied to plant ecology have been published by Crovello (1970), Goodall (1970), Williams (1971), Orloci (1973), Clifford and Stephenson (1975), and others. The purpose of this study was to test the validity and soundness of a conceptualized model of secondary succession that was developed recently. The model is visualized as a solid cone with the climax plant community situated at the At the time of the research, authors were with the College of Forestry, Wildlife, and Range Sciences, University of Idaho, Moscow 83483. G. Huschle's present address is U.S. Fish and Wildlife Service, Roy, Montana 59471. The paper is published with the approval of the director, Forest, Wildlife, and Range Experiment Station, Univ. of Idaho, as Contribution No. 53. This study was part of a United States Army Corps of Engineers contract with the Idaho Cooperative Wildlife Research Unit, College of Forestry, Wildlife and Range Sciences, University of Idaho Daa in this study were collected by personnel of the Idaho Cooperative Research Unit. Special thanks are extended to Dr. Duane Asherin, Mr. James Claar and Mr. Jerry Lauer for their contributions to the study. Manuscript received December 15, 1978. apex and associated seral communities positioned in their respective positions in the remaining portion of the cone. The solid cone model may be illustrated by a habitat type, for example, which is defined as the collective area of land supporting or capable of supporting a specific climax plant community (Daubenmire 1970). Secondary succession within a habitat type is a continuum of plant communities whose endpoint is the climax vegetation. Within the habitat type there may be numerous plant communities in various stages of secondary succession. Conceptually, secondary succession can be viewed as a solid cone in which are contained all the plant communities associated with a single habitat type or having the same successional endpoint. The solid cone includes all of the disturbed communities within a habitat type, converging to the climax community that is positioned at the upper portion of the cone (Fig. 1). As succession returns 'oward climax, the vegetational compostion changes continuously until it reaches the climax association. Toward the base of the cone, numerous communities exist because different kinds and intensities of disturbance result in different vegetation. If the disturbance is severe enough, a community with a single pioneer species or finally the base level of the cone, bare ground, may result. While secondary succession is a vegetational continuum, it is convenient to recognize community types and seral stages for practical management. A community type of one habitat type may have greater resemblance to a seral community of another habitat type than to one of its own. All community types are not unique to a habitat type. This is shown by the overlap of community types of two adjacent habitat types in Figure 1. Secondary succession can be studied by investigating