Use of invertebrate predictive models, the reference condition and causal criteria for ecological assessment of river condition

Abstract

This thesis presents my most significant contributions to the science of ecological assessment of river condition. The thesis traces the development of ecological assessment and shows where my work has made a significant contribution to knowledge of ecological assessment. I demonstrate the value of bioassessment and the ‘reference condition approach’ by describing applications and evaluation of the Australian River Assessment System (AUSRIVAS),which has been the national standard method of biological assessing river health for over a decade. AUSRIVAS includes a standardized invertebrate sampling method, the reference condition approach, predictive models, and software for assessing river health. However, new methods to aid the synthesis of ecological studies are imperative if the increasing body of scientific research is to improve management and outcomes for freshwater systems. My most recent work has contributed to establishing a new causal–criteria analysis method, ‘Eco Evidence’, for assessing evidence for and against environmental cause–effect hypotheses. This thesis reviews bioassessment and AUSRIVAS predictive modelling, the reference condition approach, and the origins of Eco Evidence to provide background and context for my research. I have arranged the nine research articles that comprise the body of this thesis in three categories: 1) AUSRIVAS sampling method evaluation; 2) applications of AUSRIVAS; and 3) the synthesis of multiple studies for environmental causal assessment using Eco Evidence. In addition, the final chapter outlines problems encountered and future directions for the work. A major contribution of my research has been to demonstrate the utility of the reference condition approach for (i) predicting reference (that is pre-impoundment) biota in the Cotter River (ACT); (ii) establishing reference biota within Kosciuszko National Park (Australia); and (iii) using the reference condition approach to assess the condition of Portuguese streams. This body of work is highly relevant to river managers wanting to apply the reference condition approach and (a) understand the consequences of sample variability on bioassessment results; (b) allocate resources appropriately for the level of replication required to detect an ecological response; and (c) avoid method-related bias where studies cross multiple jurisdictions that use different sampling methods. This research highlights the significance of standardized sampling of fixed sites (both test and reference) over long periods and demonstrates the value of the reference condition approach when assessing the biological response to flow regulation. When applied within a robust study design and an adaptive management framework, the bioassessment program coped with changing questions and unforeseen events, such as extended drought. Application of AUSRIVAS has shown that management actions maintained the ecological resilience of the Cotter River, enabling it to recover when higher river flows returned after the drought. This thesis also describes the recently published Eco Evidence method for systematic review of environmental science literature and draws together some lessons learned about the application of causal analysis to define ecosystem response to flow. The Eco Evidence method was adapted from epidemiological techniques for attributing causation. Such causal assessment can be necessary to inform management actions aiming to improve environmental condition. This work is highly relevant to researchers and environmental practitioners that require a method for quantifying and combining scientific evidence from multiple studies. The Eco Evidence weighting system for individual studies is a major advancement in environmental causal assessment. This research effort is part of a worldwide trend towards facilitating greater use of evidence-based methods in environmental assessment and management. My research has contributed to advancing the understanding of ecological assessment that uses invertebrate predictive models, the reference condition approach and causal criteria analysis. Rigorous bioassessment studies and the reference condition approach when applied within the context of adaptive management, long-term assessment, and a framework for causal assessment, can provide the ecological evidence to inform current and future river management.

Date of Award	2012
Original language	English
Awarding Institution	University of Canberra
Supervisor	Stephen Sarre (Supervisor), Ian Prosser (Supervisor) & Gary Jones (Supervisor)