Drinking water reservoir resiliency in a changing climate
This project will look to create a 3D physical model of the Welsh Water reservoir Llandegfedd to improve processes and determine impacts of interventions.
The Welsh Water reservoir Llandegfedd is important to the company since in an extreme drought scenario it will be one of the last remaining reservoirs and expected to supply the entire south of Wales. As such, its quality must be assured. Water quality issues, such as discolouration, taste and odour problems, can arise from a number of process from summer stratification and the influx of nutrients. Due to this it can be important to monitor and study these aspects of the reservoir. Furthermore steps like catchment management and installation of mechanical mixer to destratify reservoirs can be used to reduce these issues, with Llandegfedd already having a ResMix system installed. However there is a fear that climate change could further exacerbate such issues and that modern practice might not be able to ensure water quality moving forward. If these processes are fully captured a risk assessment or early monitoring system can be developed and predict any issues giving the water companies time to prepare necessary steps for prevention or treatment. This project will look to create a 3D physical model of the reservoir, using existing reservoir models to characterise biogeochemistry and hydrodynamics, to support this endeavour with the following aims:
For the model to help diagnose when and where potential processes that lead to poor water quality may occur so they can be better planned for.
Determining the impact of engineering interventions, such as mixers, on surface water quality, as well as the role of forced versus natural mixing in dictating nutrient and metal transport and transformation within Llandegfedd.
Help in the longer term management of the reservoir by looking at the resiliency in different climate scenarios since increased summer heating will lead to stronger thermal stratification and diminished transport of oxygen in the water body. This will include looking at how nutrient and metal transport changes under stronger stratification conditions. The ability of current interventions to maintain quality in the long term can then be assessed.
The models in consideration are AEM3D and Delft3D. AEM3D is a coupled 3D model of hydrodynamics and ecology within a number of hydro- environments. This model has been employed in a number of reservoir studies. This model works by coupling ELCOM and CAEDYM routines in order to achieve its aims. The solver for the hydrodynamics, ELCOM, uses unsteady, viscous Naiver-Stokes equations and can calculate a huge number of processes. While the biogeochemical element, CAEDYM, includes an array of algorithms to incorporate various processes like production and cycling. AEM3D has a module that can simulate surface mixers meaning it is capable of dealing with the ResMix placed in Llandegfedd. Delft3D also has a history of being used within reservoir problems and has been used for a wide variety of issues with morphology changes and temperature highlighting the model’s versatility. Delft3D has a number of different modules that cover a wide range of topics from water quality to flow simulation and is known to represent hydrodynamics measurements well. Delft3D lacks a module for mechanical mixers but has a more open code in general so could be tailored for our needs. Both models have their benefits and a decision should be reached early within the project.
The Llandegfedd Reservoir has data stretching back 20 years and over this time has been monitored by Welsh Water. This mean that there is a large supply of data that can be used for verification and calibration of the model. This includes both before and after the ResMix was operational, since it was installed in 2016, meaning that data will be available for both states the model will be run with. The reservoir is still visited by both Welsh Water and other academics for collection of data, which I will also contribute towards. This means up-to-date data will be captured. This data collection will include water samples, dissolved oxygen and temperature profiles. This will offer numerous variables from which the model can be calibrated against. The project will base the climate change scenarios will be based on the ones presented by the IPCC based on greenhouse gas emission and economic growth among other factors.
Once an appropriate model has been selected reservoir data will be used to calibrate the model until it reflects reality to an acceptable degree. The model can then be forced with different climate scenarios which would affect rainfall, summer temperature among other factors. The results from this model can then be used to diagnose how this changed forcing will affect the physics of the reservoir.