Inconsistencies between theory and methodology: a recurrent problem in ordination studies

摘要: Context A historical review of ordination studies is presented with particular reference to the pioneering contributions late Prof. I. Noy-Meir and their continuing relevance. Issues (1) Comparisons methods are often confounded by differences in methodological algorithm, dissimilarity measure used data standardization employed. (2) Artificial where ‘truth’ known offer a means evaluating approaches but highly sensitive ecological model assumed. (3) Data frequently its influence on poorly understood lacks theoretical justification. Review Historically, above issues have been continually raised since first use artificial Swan 1970 demonstrate ‘horseshoe distortion’. Three distinct conceptual models generate data, yet no consensus suitability has emerged they were mid-1970s. Comparative had, 1980s, shown that some recovered ‘ecological truth’ better than others. Differences between comparative models, nature matrices used, different measures, algorithms evaluations limited acceptance this conclusion. alters properties vegetation matrix. Yet little regarding results collective stand abundance, dominance or species richness, which altered standardization. Recent developments Knowledge individual measures increased; few new emerged. Pragmatism type ‘this method gives me useful answers so I do not need method’ common. Tests now occurring based distribution modelling. Conclusions A emerging non-metric multidimensional scaling such as Bray–Curtis coefficient should be preference correspondence analysis χ2 measure. Absence comprehensive composition limiting community analysis. Inconsistencies recognized 1970s remain today.