A computer-intensive approach for hypothesis testing and estimation of the Morisita aggregation index

• Grant number: 17/02479-2
• Support Opportunities: Regular Research Grants
• Start date: May 01, 2017
• End date: April 30, 2018
• Field of knowledge: Biological Sciences - Ecology - Theoretical Ecology
• Principal Investigator: Miguel Petrelli Junior
• Grantee: Miguel Petrelli Junior
• Host Institution: Pró-Reitoria Acadêmica. Universidade Santa Cecília (UNISANTA). Santos , SP, Brazil
• Associated researchers: Davi Butturi Gomes

Abstract

The subjective perception that living organisms of different taxa are aggregated (or clumped, in clusters or in patches), both in time and in space, is a notable pattern even in early studies in Population Ecology. The quantification of such aggregation level -- an interesting issue in itself but with important potential applications to fisheries and agroecosystems management, for instance -- is frequently achieved by applying methods as the patchiness and the mean crowding indices; even though the investigation of the statistical properties of these measures is a pending study. The lack of substantial knowledge in this context may be due to the fact that combining contagious distributions (either mixtures or compositions) and point process (which, in turn, are responsible for the observable spatial patterns in ecological populations) is not straightforward from a mathematical standpoint. In fact, in the Statistical literature there little, if any, information relating more sophisticated stochastic processes (than a simple random sample) and contagious distributions, which play a fundamental role in developing useful models for discrete data. Thus, the objectives of this project are: (i) to obtain both, point and interval, estimators for population aggregation measures, under different contagious distributions; (ii) to develop hypotheses tests alike; (iii) to evaluate the empirical distributions of the aggregation indices and, if possible, their exact counterparts; and (iv) to study the behavior and performance of the statistical inferences in realistic scenarios in Population Ecology, considering different shapes for area (fragments), edge effect influence, point processes intensity and type (population density and homogeneity), sampling effort and sampling method. In order to fulfil (i)-(iv), several procedures of classical and computational statistics will be applied (maximum likelihood, delta method, nonparametric bootstrap, Monte Carlo simulations), providing results that may eventually be confronted with real-world data. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

BUTTURI-GOMES, DAVI; PETRERE JR, MIGUEL. Edge influence and population aggregation: On point and interval statistical performances of Morisita patchiness index estimators in different sampling schemes. ECOLOGICAL INDICATORS, v. 108, JAN 2020. (17/02479-2)