Catchment Resilience Exploration Modeller (CREM)

Overview:

This repository produces two key applications:

• The CREMExplorer
• The CREMEngine

Both applications are configured via TOML files, based on a 'convention over configuration' approach.

CREMExplorer:

The CREMExplorer is a highly configurable modelling platform that wraps a river catchment model in either a single or multi-objective simulated annealer to explore stakeholder objectives around river catchment resilience.

The river catchment model tracks the following stakeholder objectives:

• Sediment Produced
• Particulate Nitrogen Produced
• Dissolved Nitrogen Produced
• Management Action Implementation Cost
• Management Action Opportunity Cost

This river catchment model allows management actions to be applied in order to mitigate pollutants entering a river system. The following management actions have been implemented:

• Riparian revegetation
  • predominately targeting sediment and particulate nitrogen
• Gully Restoration
  • predominately targeting sediment
• Hillslope revegetation
  • predominately targeting sediment and particulate nitrogen
• Wetland establishment
  • predominately targeting dissolved nitrogen

Single-objective simulated annealing is used to find optimised solutions to minimising/maximising a particular stakeholder objective, optionally limited by a 2nd objective. For instance, a scenario can be configured to answer a question like "Find a near-optimal minimised sediment producted for a budget of $10M in implementation costs."

Multi-objective simulated annealing is used to explore trade-offs between all the supplied stakeholder objectives, optionally limited by one of those objectives. For instance, a scenario can be configured to answer a question like "Find me a set of tradeoffs between Sediment and Dissolved Nitrogen produced for a budget of $10M in implementation costs"

CREMEngine:

The CREMEngine wraps the river catchment model described above in a web-server interface, allowing the following:

• The river catchment model can be deployed independent of the annealing, and manipulated it in real-time for visualiation purposes.
• Ability to configure the model to run exactly as per scenario definitions supplied to the CREMExplorer.
• Ability to set the model's state to match solutions produced by the CREMExplorer, to showcase optimised solutions, or individual tradeoff solutions of interest.

Getting Started:

CREM makes use of a number of 3rd-party libraries that are not included in this source repository. Go's built-in 3rd-party module support is used to track and integrate needed 3rd-party libraries. Once you've git-cloned this repository, run:

> cd <new CREM repository folder>
> go mod vendor

to download compatible versions of the libraries CREM depends on as vendor libraries.

From there a go build from within cmd/cremexplorer should produce a cremexplorer.exe executable. Then run your new executable from the command-line, specifying a scenario config file like this:

> cremexplorer.exe --ScenarioFile <someScenarioFile>

You'll find a simple test scenario configuration here. Further detail on configuring a scenario can be found in the wiki.

General Usage Notes:

• This software was constructed and tested on a 64-bit Windows 10 platform using GoLang 1.16.

• Continuous integration via travis-ci is also employed.

Contact Information:

• This software is produced on behalf of Griffith University within the Australian Rivers Institute, and originally authored by Dr Lindsay Bradford.

• E-Mail: ari@griffith.edu.au, or l.bradford@griffith.edu.au

• Voice: +61 7 3735 7402, or +61 7 3735 6598

Copyright:

The Catchment Resilience Exploration Modeller (CREM) software is licensed under a BSD 3-clause "New" or "Revised" licence, detailed in LICENCE.md.

Dependencies:

The following 3rd-party libraries are required for this code-base:

• Gomega for a Fluent-API based approach to test assertions
• go-ole for I/O via Excel files
• BurntSushi/toml for TOML config file support
• pkg/errors For error wrapping
• nu7hatch/gouuid for UUID generation