From data.gov.au
JSON format icon JSON RDF format icon RDF ISO19115-3 XML

An investigation of hypothesis testing and power analysis in impact assessment, using case studies of marine infauna.

Created 25/06/2017

Updated 11/08/2023

JSON format icon JSON RDF format icon RDF ISO19115-3 XML

This thesis used several BACI-style statistical tests for impact (MBACI and BACI-style planned comparisons) and examined the statistical power of the tests using infaunal data from 2 long-term monitoring programs. The monitoring programs were conducted between 1982 and 1994 to assess the impact of 2 coastal waste discharge points off Ninety Mile Beach in eastern Victoria, Australia. The sampling programs are outlined in the attached resource and the datasets generated from these programs are outlined in the related child metadata records.

Estimates of variance of infaunal abundance calculated from the statistical tests were extremely variable. Estimates from the same taxon, from studies in the same region, using identical sampling methods, differed by an order of magnitude or more in 25% of cases. The worst estimates of variance were usually obtained from single surveys, which had no component of large-scale temporal variation. This suggests that estimates of variances based on single sampling times can be unreliable and it may be desirable to initially include more replicates to allow for a larger than expected error variance.

The effect sizes observed in the infaunal data were very variable, with the most extreme being in excess of 10000% change. Many of the observed effects appeared to be natural events rather than as a result of a disturbance. Effect sizes for power calculations should be made in a meaningful ecological context to ensure impact assessments are both meaningful and statistically powerful.

Gradient analysis is the main alternative to the BACI-style designs in impact assessments and, in this study, provided an indication of the spatial extent of an impact however comparison with a more conventional BACIP design suggested it might be more costly to undertake. Compositing and subsampling infaunal samples was shown to be one way to reduce the costs of assessing impacts in soft sediments.

The patterns of variability seen in these data sets resulted in some large and unpredictable error variances and observed effect sizes. To reduce such problems it may be necessary to include multiple sampling times in long-term monitoring programs.

Files and APIs

Tags

Additional Info

Field Value
Title An investigation of hypothesis testing and power analysis in impact assessment, using case studies of marine infauna.
Language English
Licence Creative Commons Attribution 2.5 Australia
Landing Page https://data.gov.au/data/en/dataset/db37c658-432a-4797-a245-49c0985b1b65
Contact Point
School of BioSciences, The University of Melbourne
janetmc@unimelb.edu.au
Reference Period 25/06/2017
Geospatial Coverage
Map data © OpenStreetMap contributors
{
  "coordinates": [
    [
      [
        146.971,
        -38.551
      ],
      [
        147.363,
        -38.551
      ],
      [
        147.363,
        -38.244
      ],
      [
        146.971,
        -38.244
      ],
      [
        146.971,
        -38.551
      ]
    ]
  ],
  "type": "Polygon"
}
Data Portal data.gov.au

Data Source

This dataset was originally found on data.gov.au "An investigation of hypothesis testing and power analysis in impact assessment, using case studies of marine infauna.". Please visit the source to access the original metadata of the dataset:
https://data.gov.au/data/dataset/an-investigation-of-hypothesis-testing-and-power-analysis-in-impact-assessment-using-case-studi