Environmental water needs of Western Australia's Fitzroy River: Final report

This project was funded through the National Environmental Science Program’s Northern Australia Environmental Resources Hub with the aim of determining environmental water requirements for the Fitzroy River in the Kimberley region of Western Australia. The project was part of a larger program which used a transdisciplinary research approach that included researchers from a range of disciplines, decision-makers from government agencies, Traditional Owners and Indigenous rangers.
The project was designed in three phases. The first phase used desktop knowledge to describe hydro-socio-ecological relationships to encourage more integrated and inclusive water allocation planning. The results from this first phase, namely key considerations for water planning, fed into the development of initial water-planning documents by the Western Australian Government. The first phase also sought to identify key ecological knowledge gaps, revealing there was very little ecological data on the relationships between flow and biota or habitats for the Fitzroy River. The knowledge gaps identified by the conceptual model were used to guide data collection in the second phase of the project.
The second phase involved the implementation of targeted on-ground ecological studies, with Traditional Owners and Indigenous rangers. On-ground ecological research was conducted over three and a half years (mid-2017 to late 2020), with a focus on aquatic biota and riparian vegetation. The research outputs included data supporting five flow-biota relationships and 14 habitat-biota relationships. Flow-biota relationships are particularly useful for environmental water planning because they provide a direct link between river flow and an ecological outcome – indicating the potential implications of water extraction. Outputs from this research included new evidence relating to the following.
• Links were identified between flow velocity and algal biomass, which is important for riverine productivity.
• Local food sources support fish in the main channel at the end of the dry season, as demonstrated in a food web study.
• The body condition and intramuscular fat of fork-tailed catfish were greater in years following moderate to high wet-season flow, and these reserves decreased as the dry season progressed.
• Larger wet-season flows increase the cherabin population.
• The duration of inundation from flood flows was a strong predictor of the occurrence of woody riparian plant species, which may be used to predict species response to water-take scenarios.
• There is a zonation in the composition and structure of riparian tree species from the edge of the river and across the floodplain.
• The water sources used by riparian trees reflect the local hydrology, with trees using regional groundwater sources where it is available.
• There is a relationship between some physiological traits of riparian trees and their distribution along a hydrological gradient.
The third phase of the project reviewed new evidence that had emerged since the start of the research program, either from this project or other published sources. New scientific
Environmental water needs of Western Australia’s Fitzroy River | 2
evidence gathered during the course of this project was in broad alignment with our general theoretical understanding of how the river would function. This meant that the general structure of the conceptual model, which detailed relationships between key flow components and the habitats and biota they support, changed little between Phase 1 and Phase 3. The main change was that the model transitioned from being largely reliant on knowledge transferred from elsewhere in northern Australia and further afield to one that was populated largely by knowledge generated from the study system. This model is more defendable and creates greater certainty for management. The trade-off is that it now does not span the same ecological breadth as the original version. Therefore, some important relationships between flow and biota or flow and ecological processes such as dispersal are now no longer captured in the model. Those using the model need to be aware of this limitation. We recommend that both models be used to guide management decision-making and policy development for the river. We conclude with identified knowledge gaps that could direct future research, as well as provide considerations for the development of an effective monitoring program.