flowchart TD %% --- Project Overview --- A[Project Start<br>Sept 2025] --> B[GLM-AED Model Upgrade] B --> C[Model Calibration<br>with Sediment Dynamics] C --> D[Climate Scenario Development] D --> E[Alum–Phosphorus Module<br>Collaboration with UWA] E --> F["Scenario Testing<br>(Climate & Management)"] F --> G[Final Lake Model Outputs<br>Feb 2027] %% --- Parallel / External Links --- B -.-> H["SWAT-MODFLOW Catchment Pilot<br>3 Catchments"] H -.-> I[Scaling Decision<br>Early 2026] I -.-> D %% --- Outputs & Advice --- F --> J["Preliminary Advice on Lake Rotoiti<br>(Sept–Oct 2026)"] G --> K[Documentation & Reporting] %% --- Styles --- style A fill:#cce5ff,stroke:#004085,stroke-width:1px style G fill:#d4edda,stroke:#155724,stroke-width:1px style K fill:#d4edda,stroke:#155724,stroke-width:1px style H fill:#fff3cd,stroke:#856404,stroke-width:1px style J fill:#fff3cd,stroke:#856404,stroke-width:1px
Rotorua Lake Modelling – PC10 Support
The Rotorua Lake model will be updated to GLM-AED, an open-source hydrodynamic–ecological model that enables improved simulation of vertical mixing, nutrient cycling, and management interventions under variable climate conditions.
This upgrade will allow long-term scenario testing and a more flexible, modular architecture.
🔍 Overview
Key enhancements over previous Lake Rotorua models will include:
- Dynamic sediment–water feedback to better represent internal nutrient loading and delayed responses to external inputs.
- Support for climate change scenarios, enabling projections of lake responses over decadal scales.
- Mechanistic simulation of alum dosing, which is currently conducted in two inflows on a near-daily basis.
- Active and ongoing development of model software and related tools.
⚗️ Alum–Phosphorus Modelling
One important improvement will be incorporating a mechanistic representation of Al–P interactions.
Current models do not explicitly simulate alum chemistry or its interaction with phosphorus.
We are engaging with the University of Western Australia (UWA) to explore support for code modification within the AED structure.
The proposed model will simulate:
- Hydrolysis of dosed alum and speciation of aluminium (e.g. Al³⁺, Al(OH)₃)
- pH and temperature dependence of phosphate binding efficiency
- Sediment-mediated sorption and variable effectiveness depending on depth, organic content, and redox state
- Aging or decay of reactive Al over time to simulate persistence of alum effects
- Dosing through inflow boundary conditions, allowing for flexible scenario testing
If this collaboration is not confirmed, alternate pathways for implementing alum dynamics will be explored to maintain model delivery timelines.
🗓️ Timeline and Resources
- The model upgrade and calibration will begin in September 2025, using existing University of Waikato resources.
- Initial climate change scenarios will be prioritised and delivered by September 2026, with alum scenario testing dependent on code development.
- This lake modelling work will proceed independently of catchment model scaling, ensuring continuity regardless of SWAT-MODFLOW outcomes.
📅 Deliverables and Milestones
| Milestone | Target Date | Deliverable |
|---|---|---|
| SWAT-MODFLOW pilot (3 catchments) | March 2026 | Calibrated sub-catchment models and model evaluation |
| GLM-AED model upgrade | March 2026 | Lake model calibrated with sediment dynamics |
| Scaling decision – catchment model | Early 2026 | Evaluation and planning for full-catchment rollout |
| Initial climate scenarios for lake | September 2026 | Long-term projections under climate change (inclusion of catchment scenarios contingent on scaling decision) |
| Preliminary advice re Rotoiti | September–October 2026 | Advice on potential downstream impacts (if Ōhau Wall is repaired) |
| Complete lake model and scenario delivery | February 2027 | Finalised lake model outputs including full climate and management scenario set; documentation and reporting |
🧭 Other Considerations
The proposed work is strongly aligned with PC10 implementation.
Council staff have highlighted the need for early advice on potential changes in nutrient delivery to Lake Rotoiti, particularly if the Ōhau Wall is repaired.
This advice — ideally delivered by September or October 2026 — may be supported by the updated lake model and preliminary catchment scenario runs (depending on the outcome of the pilot).
🧱 Repository Structure (for development)
If this repository will include R modelling workflows (e.g. using the targets package), a recommended structure is:
🧭 Project Workflow Overview
The following diagram summarises the key phases, dependencies, and outputs of the Rotorua Lake Modelling project.