Introduction to the Riparian Restoration Experiment
Faithful Creek, Victoria
A degraded riparian zone, typical of small creeks throughout southeastern Australia.
The restoration of riparian zones is being carried out across streams throughout Australia, costing millions of dollars annually. These efforts are motivated by an understanding that the overall health of our streams is intimately linked with condition of the riparian zone. However, the magnitude, rate and timing of ecosystem recovery once restoration activities have commenced is far less certain. There is also a need to better understand the specific mechanisms involved in recovery and the key factors that might indicate success. In most cases the responses of a stream to riparian rehabilitation are not monitored, and where monitoring is conducted no consistent methods are used. With the support of the Murray-Darling Basin Commission, and various Catchment Management Authorities, scientists at Monash University are addressing this knowledge gap.
The Riparian Restoration Experiment was initiated in 2000 following the recommendations of a report to the Murray Darling Basin Commission by a joint project team from the CRC for Freshwater Ecology and the CRC for Catchment Hydrology. The team was commissioned to identify:
- The type of restoration projects that would most usefully be targeted for evaluation;
- How to implement an evaluation program;
- The selection of key monitoring sites;
- A scientifically rigorous experimental design.
After a review of restoration activities (Stewardson et al. 2002), the most common form of habitat rehabilitation in Australia was found to be riparian revegetation accompanied by livestock exclusion. Consequently, this became the focus of the study. Three approaches of evaluating the effectiveness of riparian restoration were considered: 1) post project monitoring of existing sites; 2) combining management and monitoring as part of a current restoration program; 3) dedicated experimentation. Of these, the latter was deemed the most valuable.
Experimental Design & Monitoring
The focus of the study is lowland tributary streams in the southern Murray-Darling Basin with historical riparian vegetation dominated by river red gum (Eucalyptus camaldulensis). Initially, sites were established on three Victorian streams ( Faithful Creek, Joyces Creek and Middle Creek) in 2003. The project has since been expanded to include two further sites in New South Wales ( Little Billabong Creek and Narrallen Creek).
The original intention was to follow the MBACRI* approach (Downes et al. 2002 ). With this in mind, a number of criteria influenced the selection of study locations, the most important of which was the existence of suitable reference sites (Stewardson et al. 2002). Following an extensive site selection exercise during which 98 sites on 39 streams were assessed, it was evident that the inclusion of reference sites was impossible as they were unavailable within our study area (Anderson et al. 2004). Therefore, the experimental design consists of a control site and treatment site located along multiple creeks.
Pre-restoration data were collected prior to fencing and replanting using a monitoring protocol designed to identify key changes over the short (1-3yrs), medium (3-8yrs) and long-term (10+yrs). The intention is to continue monitoring these sites for at least 10 years.
Predicted responses to restoration, and the importance of context
As suitable reference sites could not be found, conceptual models were developed at the beginning of the project to set hypothesized targets in terms of the likely timing and magnitude of responses to restoration ( Reich et al. 2008). However, contextual variables such as land use, drought and other stressors are also likely to influence restoration (e.g. Harding et al. 1998, Bond & Lake 2005, Palmer et al. 2010), and generate spatial and temporal variability in ecological responses. Because sites needed to meet a number of selection criteria (e.g. catchment size, water quality, land use, vegetation type) to be included in the study initially, it is therefore uncertain the degree to which these conceptual models may be applicable to sites with different characteristics. A current focus of the project is on beginning to develop methods to test how site and landscape context might result in spatially variable responses to restoration, and to subsequently examine the degree to which the findings of the RRE might be able to inform riparian restoration more generally across the Basin.
* Multiple Before After Impact Reference Control