growR - R implementation of the grassland model ModVege[^1]

This R implementation of the grassland model ModVege by Jouven et al.[^1] is based off an R implementation created by Pierluigi Calanca[^2].

The implementation in this package contains a few additions to the above cited version of ModVege, such as simulations of management decisions, and influences of snow cover. As such, the model is fit to simulate grass growth in mountainous regions, such as the Swiss Alps.

The package also contains routines for calibrating the model and helpful tools for analysing model outputs and performance.

Contents

1. What is this for?
2. Installation
3. Getting Started
4. Contributing
5. Contact
6. Glossary
7. Footnotes and References

What is this for?

This R package allows simulation of grass growth.

Why simulate grass growth?

Grasslands constitute one of Earth's most widespread terrestrial ecosystems[^3] and a core element in global agriculture, providing roughly half the feed inputs for global livestock systems [^4]. Beside their contribution to global food production, they provide a catalogue of other ecosystem services, such as water flow and erosion regulation, pollination service, carbon sequestration and climate regulation [^3]. The latter have become particularly important in light of anthropogenic climate change [^5].

Understanding the functioning of grassland ecosystems and their responses to external changes is therefore of significant interest. Vegetation models provide a powerful platform for such studies.

How does this compare to other grass and vegetation models?

The number of grassland models is large and ever-growing. We can therefore not give a comprehensive list, but will try to make a couple of representative comparisons to illustrate where growR has its niche. For the most part, an advantage of growR over other, similar models and their implementations is its distribution as R package via CRAN.

• The Hurley Pasture Model [^6] is a detailed mechanistic model for managed pastures. It is implemented in the Advanced continuous simulation language (ACSL) and the source code is available on request.
• BASGRA [^7] and its descendant BASGRA_N [^8] are multi-year grassland models which include tiller dynamics. They are also implemented in R with the source code freely available. However, they do not come packaged, as growR does.
• PROGRASS [^9] was developed to capture the interactions in grass/clover mixtures. As of this writing, no accessible implementation was found.
• The focus of PaSim [^10] is the investigation of livestock production, which is not directly covered in growR, under climate change conditions.

Installation

From CRAN

This is the preferred installation route for most users.

This R package can be installed as usual from CRAN by issuing the following at the prompt of an R session:

install.packages("growR")

From source

Installing from source might make sense if...

• you intend on making changes to the model[^3],
• you want to contribute to package development and maintenance,
• you want to get access to the cutting edge version, which may have changes not yet available on the CRAN version but is also likely less stable,
• for some reason installation from CRAN is not an option for you.

In this case, start by cloning this repository

$ git clone git@github.com:kuadrat/growR.git

or via https:

$ git clone https://github.com/kuadrat/growR.git

This will create a directory growR in your file system.

If you don't have or don't want to use git, you could alternatively copy the source code as a .zip file from github. Unzip the contents into a directory growR.

Alternative A

You can now install your local version of growR by issuing the following at the prompt of an R session:

install.packages("/full/path/to/growR", repos = NULL)

You should replace "/full/path/to/" with the actual path to the growR directory on your computer. Also, replace slashes (/) with backslashe (\) if you're on Windows.

growR should now be installed and available in R through library(growR).
If you make changes to the source files in the growR directory, just uninstall the current version (issue remove.packages("growR") in R) and repeat this step.

Alternative B

If you make frequent changes to the code, it might be unpractical to uninstall and reinstall the changed version each time. In that case, devtools comes in very handy (if needed, install it with install.packages("devtools")). It allows you to load a package into an active R session without the need of it being properly installed. The following has practically the equivalent result as the method described in alternative A:

library(devtools)
devtools::load_all("/full/path/to/growR")

The notes about "/full/path/to" as in Alternative A apply here as well.

Non-package version

If you just want to focus on using and adjusting the ModVege model and feel that the structure of an R package is more of a hindrance than a help to your cause, there is a third option. Simply use the pre-R-package version of growR, called rmodvege, which is essentially a collection of R scripts. Some users might be more familiar or comfortable working in this manner instead of working with package code.

Go to https://github.com/kuadrat/rmodvege-scripts to access the script-based implementation of ModVege. Note, however, that the script based version is not maintained and might therefore lack some functionality which is provided by the growR package.

Getting Started

The package documentation is hosted on github pages: https://kuadrat.github.io/growR/. Have a look to find an introductory tutorial, further information as well as the complete package reference.

Alternatively (in case github pages are down or you prefer an offline solution), you can find the same information under Reference manual and Vignettes on the CRAN package homepage: https://cran.r-project.org/web/packages/growR/index.html

Finally, it's also possible to directly access the package documentation and vignettes from an R interpreter, using the ? and vignette() tools, e.g.

> library(growR)
# Get help on a function or object
> ?growR_run_loop
# some output...

# List available vignettes
> vignette(package = "growR")
Vignettes in package ‘growR’:

parameter_descriptions  
                        Parameter Descriptions (source, html)
growR                   Tutorial (source, html)

# Inspect a vignette
> vignette("growR")

Contributing

All forms of contributions to this project are warmly welcome. You are invited to:

• provide direct feedback over e-mail.
• submit bug reports and feature requests via github issues.
• make changes and additions to the code and submit pull requests to let your contributions become part of future versions.
• suggest improvements for or write documentation and tutorials.
• reference work that made use of growR here.

If you intend to collaborate in a regular and ongoing manner, best get in touch with Kevin Kramer.

Contact

Kevin Kramer: kevin.pasqual.kramer@protonmail.ch

Glossary

Terms used in this project

• growR

Name of this project and the corresponding R package. The shown capitalization is adhered to even when used in function or object names in the code base.

• ModVege

The basis for the underlying grassland model implemented here. The naming convention of objects overrides the capitalization shown here, when the model name is referred to in function and object names.

• rmodvege

Early name of this project and still the name of a legacy project that was not factored as an R package, but rather as a collection of R scripts. Still available, though unmaintained at https://github.com/kuadrat/rmodvege-scripts/.

Footnotes and References

[^1]: Jouven, M., P. Carrère, und R. Baumont. „Model Predicting Dynamics of Biomass, Structure and Digestibility of Herbage in Managed Permanent Pastures. 1. Model Description“. Grass and Forage Science 61, Nr. 2 (2006): 112–24. doi:10.1111/j.1365-2494.2006.00515.x.

[^2]: Calanca, Pierluigi, Claire Deléglise, Raphaël Martin, Pascal Carrère, und Eric Mosimann. „Testing the Ability of a Simple Grassland Model to Simulate the Seasonal Effects of Drought on Herbage Growth“. Field Crops Research 187 (Februar 2016): 12–23. doi:10.1016/j.fcr.2015.12.008.

[^3]: Zhao, Yuanyuan, Zhifeng Liu, and Jianguo Wu. “Grassland Ecosystem Services: A Systematic Review of Research Advances and Future Directions.” Landscape Ecology 35, no. 4 (April 1, 2020): 793–814. doi:10.1007/s10980-020-00980-3.

[^4]: Herrero, Mario, Petr Havlík, Hugo Valin, An Notenbaert, Mariana C. Rufino, Philip K. Thornton, Michael Blümmel, Franz Weiss, Delia Grace, and Michael Obersteiner. “Biomass Use, Production, Feed Efficiencies, and Greenhouse Gas Emissions from Global Livestock Systems.” Proceedings of the National Academy of Sciences 110, no. 52 (December 24, 2013): 20888–93. doi:10.1073/pnas.1308149110.

[^5]: IPCC Report 2022, Chapter 5.

[^6]: Thornley, J. H. M. Grassland Dynamics: An Ecosystem Simulation Model. CAB International, 1998.

[^7]: Van Oijen, M., M. Höglind, D.R. Cameron, and S.M. Thorsen. “BASGRA_2014.” Zenodo, August 13, 2015. https://doi.org/10.5281/zenodo.27867.

[^8]: Höglind, Mats, David Cameron, Tomas Persson, Xiao Huang, and Marcel van Oijen. “BASGRA_N: A Model for Grassland Productivity, Quality and Greenhouse Gas Balance.” Ecological Modelling 417 (February 1, 2020): 108925. doi:10.1016/j.ecolmodel.2019.108925.

[^9]: Lazzarotto, P., P. Calanca, and J. Fuhrer. “Dynamics of Grass–Clover Mixtures—An Analysis of the Response to Management with the PROductive GRASsland Simulator (PROGRASS).” Ecological Modelling 220, no. 5 (March 10, 2009): 703–24. doi:10.1016/j.ecolmodel.2008.11.023.

[^10]: Graux, A. -I., M. Gaurut, J. Agabriel, R. Baumont, R. Delagarde, L. Delaby, and J. -F. Soussana. “Development of the Pasture Simulation Model for Assessing Livestock Production under Climate Change.” Agriculture, Ecosystems & Environment 144, no. 1 (November 1, 2011): 69–91. doi:10.1016/j.agee.2011.07.001.

[^3]: If you make changes that generally improve growR, it would be great if you could share them to make them available to all future users. See Contributing.