New Zealand Statistical Association 2024 Conference

Justin Murphy

Stats NZ

A tool to evaluate environmental trend estimation methods

This is joint work with Tiana Whitehead, Pubudu Senanayake

Accurately measuring environmental trends is crucial for informing the effectiveness of environmental management policies and identifying where resources need to be allocated.

StatsNZ were commissioned by the Ministry for the Environment (MfE) to assess current and alternative methods for environmental trend estimation. To do this, we created a Shiny tool which enables the user to create realistic synthetic data by introducing a known trend to a dataset that retains the other properties of the data, such as noise profiles. This makes it possible to determine how well trend estimation methods can recover trends present in the data.

The tool works by decomposing real environmental time series into trend, seasonal, and noise components, and allows users to replace the estimated trend component with a customizable trend, choosing from multiple patterns. The customizable trend is then recombined with the original seasonal and noise components, producing a synthetic time series. In addition, users can scale the magnitude of the seasonal and noise components and can set the time windows in which trends are to be estimated.

The Shiny tool facilitates a comparison of the accuracy of trend estimation methods under different conditions (e.g., when the true underlying trend is linear vs non-linear), and the sensitivity of the methods to the magnitude of the seasonal and noise components, and to the time-window chosen for analysis.