A Change Point Distance-Based Multivariate Control Chart Tool for Ecological and Environmental Monitoring

• Activity Number: 615 - Statistical Process Control
• Type: Contributed
• Date/Time: Thursday, August 1, 2019 : 8:30 AM to 10:20 AM
• Sponsor: Quality and Productivity Section
• Author(s): Nurudeen A. Adegoke* and Marti J. Anderson and Adam N. H. Smith and Matthew D. M. Pawley
• Companies: School of Natural and Computational Sciences, Massey University, New Zealand and New Zealand Institute for Advanced Study, Massey University, New Zealand and School of Natural and Computational Sciences, Massey University, New Zealand and School of Natural and Computational Sciences, Massey University, New Zealand
• Keywords: Change point; control chart; dissimilarity matrix; ecological monitoring; ecological statistics; permutation

Abstract

Environmental monitoring programmes have been established across the globe to monitor whole communities of species through time. It is vitally important that such programmes be empowered to detect significant changes in sustainable ecosystems quickly, so that deeper investigations and remedial restoration measures, if necessary, can be implemented directly. In industrial monitoring, a commonly used tool for this purpose is the control chart. Analogous to the industrial monitoring, in this work, we propose a multivariate control chart for monitoring ecological data sets. The monitoring aims to discover when (and where) an impact may have occurred in the ecosystem. Our proposed charting method is based on a change point method and does not require prior knowledge of the ecosystem's behavior before the monitoring can begin. Also, the method does not require any parametric model assumptions, as it can be based on any dissimilarity measure of choice. A permutation procedure is employed to obtain the control-chart limits for a suitable distance-based model through time. We provide examples to show the applications of the procedure in detecting shifts in ecological community structure.