Skip to main content

Application of the Relative Risk Model (RRM) to Investigate Multiple Risks to the Miranda Ramsar Site

Report: TR 2007/22
Author: Ibrahim Elmetri (Cawthron Institute), Malene Felsing (Environment Waikato)


This report presents the application of the Relative Risk Model (RRM) to investigate multiple risks to the ecology of the Ramsar site in the southern Firth of Thames. The work reported here is part of the multi-agency collaborative 'Muddy Feet Phase II: the Firth of Thames Ramsar site' project, which is co-ordinated by Environment Waikato and Franklin, Hauraki and Thames Coromandel District Councils, and sponsored by these agencies as well as the Auckland Regional Council and the Department of Conservation. Collaborators on the project include sponsors, plus Matamata Piako District Council and the Ministry of Fisheries.

The primary objective of the application of the RRM to the Ramsar site was to assess the relative threat posed by different risk sources, and their stressors, to selected ecological values of the Ramsar site and surrounds. Representatives from collaborating agencies participated in workshops where ecological values of the Ramsar site and surrounds were identified, as were stressors and sources of stressors that may threaten the site. A workshop was also held for members of the community, where values of the Ramsar site and perceived threats to these were recorded. Iwi values were incorporated through an analysis of submissions to Environment Waikato policy over the last six years, and feedback on these was sought in a hui.

Where limited information was available, or uncertainty about impacts of some sources in an area was high (for urban and industrial land use, marine farms, mangrove expansion, and climate change), two different scenarios were modelled. These scenarios represent predicted risk and worst case scenario, respectively.

The predictions of the RRM for the Ramsar site and its catchment found the greatest stressor on the biological endpoints to be from terrestrial drivers. Agricultural land use (dairy farming) contributed by far the largest risk to the site. Climate change, Firth of Thames sediments, point sources, fishing, urban and industrial land use, and mangrove expansion posed lesser but important risks. Sedimentation was found to be the biggest stressor, and the largest sources of sediments were agricultural land use and sediments already in the Firth of Thames. The greatest habitat risk was to the tidal flats of the Ramsar site, and lower, but relatively important risks existed for the water column, the sub-tidal seabed, and stilt ponds. At highest risk among the biological values were shellfish beds, some fish species and marine worms. The second highest risk was found to be to some shorebirds, the area important for creating the Chenier bank, saltmarsh, and other fish species. Vegetation only marginally exposed to salt water influences (bachelor's button, sedges, Maori musk, burmedic short grassland, and ryegrass), and the coastal birds (shags, heron and banded rail) and mud crabs were at the lowest risk.

The high risk score predicted from agricultural land use is not surprising considering the extensive dairy farming that dominates the catchment. Whilst sediment supply rates from land are thought to be relatively low at present, the sediment reservoir in the Firth (built up from decades of high sedimentation rates from the catchment) is thought to be fuelling the recent dramatic expansion of the mangrove forests. Thus, even if sediment runoff from land is reduced, trends of increasing sedimentation could continue to occur for a while because of the storage of sediments in the Firth.

Overall the application of the RRM to the Ramsar site was useful. The work confirmed that the RRM is a rapid, powerful, flexible and cost effective tool that can provide an overview of the relative risks to a site from multiple sources, generating outputs that resource managers can likely use to aid decision-making. The study established the RRM as a tool for the Ramsar site management, which can be used, and expanded upon, by management agencies and other stakeholders. Stressors and environmental values can be modified easily with new input. As such, the RRM results presented here are considered a starting point for discussion and a guide to prioritise management actions.

The largest uncertainty source found in this risk assessment was the lack of data or scientific knowledge and understanding. As future studies fill these data and information gaps, uncertainty will be reduced. Based on our review of the available information for the site as well as the Monte Carlo uncertainty analysis, a list of the most pressing information needs is presented.

Following on from the risk assessment described in this report, the Muddy Feet Phase II project proceeded to identify priority actions to reduce key risks to the Ramsar site. This was achieved through identifying what can be done to minimise each of the risks included in the model, and who should do it, and comparing this to what management agencies are currently doing or have plans to do in the near future. Actions not covered by existing work programmes were prioritised, as were any information needs identified, and recommendations provided to all agencies involved. The results from this part of the project are reported in a separate report.

Application of the Relative Risk Model (RRM) to Investigate Multiple Risks to the Miranda Ramsar Site
(550 kb, 78 seconds to download, 56k modem)

Table of contents

  Acknowledgement i
  Executive Summary v
1 Introduction 1
1.1 The Muddy Feet Phase II Project 1
1.2 Objectives of the Current Study 1
1.3 Structure of this report 1
2 Background to study area 2
3 The Relative Risk Model 5
3.1 Project Approach 5
3.2 Outline of the Relative Risk Model (RRM) 5
4 Relative Risk Model methodology 7
4.1 Identification of model parameters 8
4.2 Study Area 11
4.3 Assessment Endpoints 11
4.4 Habitats 14
4.5 Sources of Stressors 14
4.6 Conceptual models 21
4.7 Source and Habitat Scores 21
4.8 Exposure and Effects Filters 26
4.9 Monte Carlo Uncertainty Analysis 27
5 RRM Results 28
5.1 Risk Characterisation 28
5.2 Monte Carlo Uncertainty Analysis 34
6 Conclusions 37
References 42  
Appendix I. Exposure Filters 46
Appendix II. Effects Filters 54