New methods for the analysis of spatial pattern in vegetation

SUMMARY (1) A new method of analysis is suggested that modifies the interpretation of quadrat variance methods for detecting the scale of spatial pattern in vegetation. (2) A complementary method is presented for distinguishing the contributions made by the sizes of the gaps (regions of low density) and of the patches (regions of high density) to that pattern. (3) These methods are tested on a variety of artificial data and it is shown that they can recover the important characteristics of artificial patterns. (4) The application of these methods to a study of primary succession on proglacial deposits indicates that the scale of pattern does not change with surface age. The growth and coalescence of patches of vegetation may be counterbalanced by their senescence and breakup, and by continued colonization.