Workflow-for-identifying-potential-water-quality-events-using-time-series-data

A fully functional water quality project written in R showing a simple step-by-step workflow for processing time series data from a water quality sensor and how to use that data for the identification of potential water quality events.

Description

This project contains a concise, reproducible, open-source workflow detailing the use of R, R markdown and EPA CANARY event detection software and was designed for catchment scientists, researchers and those working in on stream hydrochemistry, diffuse pollution or any EU Water Framework Directive water quality projects who require a simple, detailed workflow that can be easily replicated requiring minimum expertise in R. One of the primary objectives of this project was to provide simple code that non-coders can follow and to produce outputs at each step of the workflow so the user can validate the code for their data. This project includes a sample data file and a detailed markdown so you can take the sample file provided through each step of the workflow, prior to applying the code to your own data.

Each part of this project is simple sample code which details how to do the following: * import time series data, * clean the data, * visualize the data, * maintain data integrity, * manually identify potential water quality events, * match those events to the output of EPA CANARY event detection software, * identify the most suitable configuration of EPA CANARY for that particular water quality monitoring station.

Getting Started

Dependencies

• Needs recent install of R and Rstudio.
• Dependencies are managed through the renv package which was created by RStudio developers specifically for this purpose. The "renv" package will create a package library which is specific to this project. Thus you can ensure that you have the correct version of all required packages (those used in the original build). Outside of the project, you can use or install whatever package versions you prefer, those inside the project space will not be affected.

Installing

• Install R and Rstudio latest versions.

• Install the "renv" package by typing install.packages("renv") in the R console.

• You need to have a github account and a local install of git on your machine.

• Open a git BASH (start the "git BASH app") and introduce yourself to git. This will allow Rstudio to connect successfully to github.

• Open RStudio and create a New Rstudio Project.

• Select the option to install from "Version Control".

  

• Select "git".

• In the dialog box that follows, fill the first two boxes in as shown below:

  

• In the third box, browse to the folder on your local machine where you want to install the project.

• Select "Create Project"

• Ensure that "renv" is enabled for the project as follows:

  • In Rstudio click on "Tools" menu, followed by "Project Options".
  • Select "Environments" from the list of icons on the left of the dialog box.
  • Ensure the the box beside "Use renv with this project" is ticked as shown:

• Now check that the correct versions of all required packages are installed by running the command renv::status(), from the R console. You will see a list of any dependency issues.

• To resolve any dependency issues, type the command renv::restore(). The correct version of every required dependent package will now install.

• At this point you should be able to execute any of the scripts or RMd documents within any directory.

Executing program (from Rstudio)

• Follow the step by step instructions, the code and the output of the code in the 'Workflow for Hydrodynamic Event Detection' markdown file.
• The scripts are also available but the markdown file is the place to start for users who are new to R.

Executing program (from Docker)

In the event that you want to run the package from docker, follow the setup steps described above in order to create a copy of the repository on your local machine.

Then follow the steps outline here.

The Workflow

Open the Project file, Project Workflow for detecting Hydrodynamic Water Quality Events (extension R Project) followed by the Workflow-for-Hydrodynamic-Event-Detection ( extension RMD file).
There is also a Workflow-for-Hydrodynamic-Event-Detection HTML file which can be a helpful read along user guide for those new to R. It is too large to open in GitHub so must be downloaded to your web browser.

Help

More information on reusing other people's code in your project

How to tweak this project for your own use

We would encourage you to clone and rename this project to use for your own data.

Find a bug

If you find an issue or would like to submit an inmprovement to this project please contact me.

Known issues (work in progress)

This workflow is still ongoing. The streamlining of importing datafiles has not been completed yet. This is coming soon!

Authors

Lisa Cronin

Cian Taylor

Version History

• 0.1
  • Initial Release April 2025

License

This project is licensed under the [MIT] License - see the LICENSE.md file for details

Acknowledgments

The authors would like to thank Jonathan Burkhardt, Ph.D., Office of Research & Development, U.S. EPA for providing further information on the matching of manually identified events to CANARY events which enabled the development and refinement of the code.

Citation

Lisa Cronin and Cian Taylor (2025) “LCroninATU/Workflow-for-identifying-water-quality-hydrodynamic-events-in-time-series-data: V1.0.0”.