Temporal trend monitoring: Robust method for analysing contaminant trend monitoring data.

Abstract

This document describes a new method of assessing contaminant trends in fish muscle, fish liver, and shellfish.

Previous methods of assessing trend data have often been complicated by the need to respond to unusual features of the data, either in the way they were collected, processed, or in their distribution. These complications are discussed in detail to show the reasoning behind the method described here, which aims to provide a simpler, robust and more complete method of analysing and presenting trends.

Essentially, the method summarizes trends using a smoother, a specific class of smooth curves fitted to median log-concentrations. The theory and methodology of fitting smoothers is new and our knowledge of the performance of the fitted smoothers, particularly with small sample sizes, is only approximate. Although a preliminary application of the new method to the 1993 assessment of the data from the Joint Monitoring Programme of the Oslo and Paris Commissions was promising, more assessment and development of the method will be necessary. A very simple smoother is used here to make the computations and theory easy to follow.

A detailed worked example IS provided. Statistical theory and formulae are included as annexes.