Use, Misuse, and Limitations of Threshold Indicator Taxa Analysis (TITAN) for Natural Resource Management

Abstract

Detection of ecological thresholds has broad relevance to management of ecosystems. However, ecological community data present a distinct problem because current statistical methods used for identifying thresholds were not developed for analysis of multiple, individual species abundances. We developed a new method, Threshold Indicator Taxa ANalysis (TITAN), specifically to deal with some of the limitations of existing methods for estimating community thresholds. Our objectives in this chapter are to (1) summarize the theoretical basis for the method and related methods, (2) provide a brief overview of how it works, (3) use a real data set to illustrate an application of the method, and (4) conclude the chapter by addressing several issues related to the appropriate use of the method, misconceptions about how it works or what the results mean, and limitations that could lead to erroneous conclusions. We explain that step-function conceptualizations of community thresholds are not sufficiently inclusive of all the response forms that satisfy threshold criteria, how gradual responses of univariate community metrics do not rule out community thresholds, and that linear regression techniques do not provide an adequate test for the absence of thresholds, especially in the presence of long environmental gradients. We note substantial misunderstanding in the recent literature regarding appropriate use and interpretation of statistical change points identified by taxon-specific analysis in TITAN, that univariate community metrics are inappropriate response variables for such analyses, and that extreme variation in the density of the sample distribution can affect results of any method, including TITAN. We end by reminding users that despite the additional insight it brings to community analysis, TITAN is neither a causal analysis nor a black box for developing regulatory criteria. Instead, we intend TITAN to complement current analytical approaches, while highlighting assumptions and flaws in the broader paradigms in which they are often applied.