CASE STUDY - CFD MODELLING OF MILDURA FISHWAY
The Mildura Weir (or Lock 11) Denil Fishway was established under the Murray-Darling Basin Authority's 'Sea to Hume Dam' program around 2013 to facilitate both upstream and downstream passage of medium to large-sized fish species (like Murray Cod or Gold and Silver Perch) in this region.
Between 2020 to 2023 investigatory works were conducted with concept designs for short-term improvements to fish passage at the site.
This case study will examine the hydraulics at the entrance of this fishway, focusing on the base case (2013 design) and short-term concept plans (design case). Computational fluid dynamics (CFD) with the software package FLOW3D was undertaken for this. CFD modelling of the fishway was undertaken in accordance with practical best practices.
Special thanks to Steven Slarke of Jacobs, Ivor Stuart of CSU and the Commonwealth Environmental Water Holder (CEWH) for the site footage provided below and support of this case study.
SETTING THE SCENE
The fishway, positioned at the left abutment of the weir, is built on a sloped concrete apron and flat apron slab. The fishway is constructed separately from the weir allowing for complete removal from the river during flood events or sectional dismantling during different operational periods.
This flexibility in design allows for cost-effective interventions to increase fish migrations with the denil design selected due to site constraints, ease of dismantlement and ability to pass larger fish species in the catchment.
The base case (2013 design) denil fishway included an entrance shield to deflect high velocities and turbulence from the weir, and aid in fish attraction. The short term concept plan that was modelled for this CFD study included increasing the depth of the shield to provide a seal with the apron slab, and additional denil panels at the entrance.
FINDINGS & OUTCOMES
Initial coarse modelling of the entrance ramp and weir for entrance attraction found the updated design performed significantly better than the 2013 design. This improvement was measured by the following:
- A reduction in velocities at the entrance during operation of the weir
- A reduction in eddies at the entrance, most notably the vertical roller that passed under the shield
- A reduction in air entrainment at the entrance
Finer modelling of the entrance ramp in isolation found the additional denil plates reduced velocities on average through the ramp. Interestingly, the high velocity (streaming flow) at the top of the ramp remained for both designs. This high velocity may be a function of the slope of the ramp but more modelling would be required to confirm this.
For a shallower tailwater condition, velocities were found to increase for both designs with the short-term concept plans performing better than the previous (2013) design.
A comparison of model results against site footage for both designs was conducted with the CFD model replicating observations quite well. As shown in the image, the following was observed (from top to bottom):
- A substantial vertical roller was found upstream of the shield and low velocity zone downstream for the updated design.
- "Bubbling" directly downstream of the entrance was observed on site from the 2013 design, air entrainment modelling seemed to predict this effect.
- Streaming and plunging zones from the model seem consistent with site observations.
Regardless of what the modelling is showing, success is measured in practice. How these new velocities and eddies translate to fish passage through the structure is still to be determined, we look forward to seeing monitoring results for the next migratory season.