Representation of an alpine treeline ecotone in SPOT 5 HRG data

An ecotone is a zone of vegetation transition between two communities,
 often resulting from a natural or anthropogenic environmental gradient.
 In remotely sensed imagery, an ecotone may appear as an edge, a boundary
 of mixed pixels or a zone of continuous variation, depending on the
 spatial scale of the vegetation communities and their transition
 zone in relation to the spatial resolution of the imagery. Often
 in image classification, an ecotone is either ignored if it falls
 within a width of one or two pixels, or part of it may be mapped
 as a separate vegetation
 
 community if it covers an area of several pixel widths. A soft classification
 method, such as probability mapping, is inherently appealing for
 mapping vegetation transition. Ideally, the probability of membership
 each pixel has to each vegetation class corresponds with the proportional
 composition of vegetation classes per pixel. In this paper we investigate
 the use of class probability mapping to produce a softened classification
 of an alpine treeline ecotone in Austria using a SPOT 5 HRG image.
 Here the transition with altitude is from dense subalpine forest
 to treeless alpine meadow and herbaceous vegetation. The posterior
 probabilities from a Maximum Likelihood algorithm are shown to reflect
 the land-cover composition of mixed pixels in the ecotone. The relationships
 between the posterior probability of class membership for the two
 end-member classes of ‘scrub and forest’ and ‘non-forest vegetation’
 and the percentage ground cover of these vegetation classes (enumerated
 in 15 quadrats from 1:1500 aerial photographs) were highly significant:
 r2=0.83 and r2=0.85 respectively ( pb0.001, n=15). We identify thresholds
 (alphacuts) in the posterior probabilities of class membership of
 ‘scrub and forest’ and ‘non-forest vegetation’ to map the alpine
 treeline ecotone as a transition zone of five intermediate vegetation
 classes between the end-member communities. In addition, we investigate
 the representation of the ecotone as a ratio between the posterior
 probabilities of ‘scrub and forest’ and ‘non-forest vegetation’.
 This displays the vegetation transition without imposing subjective
 boundaries, and has greater emphasis on the ecotone transition rather
 than on the end-member communities. We comment on the fitness for
 purpose of the different ways investigated for representing the alpine
 treeline ecotone.