aj-sykes92 / ggmacc

R package repository for building marginal abatement cost curves with ggplot2.

ggmacc

Dr Alasdair Sykes 16/11/2020

Easy, stylish marginal abatement cost curves with ggplot2

Marginal abatement cost curves (MACCs) are a data visualisation and decision tool used in the economic analysis of pollution (typically greenhouse gas emissions) abatement. A MACC is an easy-to-read visualisation of a series of mitigation measures vs. a baseline (usually business-as-usual) contextualised by economic cost.

In pure data visualisation terms, MACCs are a little special and so most off-the-shelf graphing packages (e.g. MS Excel) struggle to produce them. This small and simple package brings together a group of functions which make use of the ggplot2 environment (and specifically geom_rect) to draw simple, clear and attractive MACCs with the minimum of fuss for the user.

This package was imagined and written by Dr Alasdair Sykes (@alasdair_sykes). If you use this package in your work, please cite it as:

Sykes, A. J. (2020) ggmacc: Easy marginal abatement cost curves with ggplot2. Version 1.0.0. Available at https://github.com/aj-sykes92/ggmacc.

Installation

This package is hosted on Github and can be installed using the devtools package:

# install.packages("devtools")
devtools::install_github("aj-sykes92/ggmacc")

Note: If this doesn’t work, update devtools, as Github recently changed the term it uses for the primary version of a source code repository.

Usage — ggmacc

This package is comprised of three functions; one (macc_prep) for preparation of a raw dataset for plotting, one (geom_macc) for adding the MACC geom to a ggplot object, and one (ggmacc) which brings these together with other functional elements for an ‘out-of-the-box’ marginal abatement cost curve plot.

The package also contains two datasets; a small, fiction dataset (small_example) for demo purposes, and a larger dataset based on real analysis (uk_agroforestry).

The ggmacc function provides a fully developed ggplot-based MACC plot with all the trimmings:

social_cost_of_carbon <- 66.1

full_macc <- uk_agroforestry %>%
  ggmacc(abatement = co2_tyear, mac = mac_gbp_tco2, fill = crop, cost_threshold = social_cost_of_carbon,
         zero_line = TRUE, threshold_line = TRUE, threshold_fade = 0.3)

full_macc

I’ve avoided replicating any ggplot2 functionality unnecessarily. However, the output is a ggplot object, so refinements can be added with the usual syntax:

full_macc +
  scale_x_continuous(labels = scales::number_format()) +
  scale_fill_manual(values = Manu::get_pal("Kea")) +
  labs(title = "Marginal abatement cost curve for UK agroforestry",
       fill = "Crop type",
       x = expression("Abatement (tonnes CO"[2]*"-eq)"),
       y = expression("Marginal abatement cost (GBP tonne CO"[2]*"-eq"^{-1}*")")
       ) +
  theme_classic()

Thanks to the very cool Manu package for providing this colour palette — find it at https://github.com/G-Thomson/Manu.

Usage — base functions

The two base functions wrapped in ggmacc have been designed to work with a typical tidyverse workflow and use tidy evaluation. They are made available to allow the user to fully manipulate the basic elements of this package. Pipe data into macc_prep and identify the variables which characterise abatement and cost; pipe into ggplot and add geom_macc to the result. geom_macc also takes a fill argument:

small_example %>%
  macc_prep(mac = mac, abatement = abatement) %>%
  ggplot() +
  geom_macc(fill = cat)

It’s worth noting this is a slight perversion of the ggplot2 grammar-of-graphics philosophy in that geom_macc does not exist separately from its aesthetics — the aes() arguments are tied up in the function call — however, for flexibility and ease of use, this seemed like the best approach. The result can be augmented and refined in the same way as any other ggplot object. This disaggregated syntax also allows the user opportunity to modify the base data following macc_prep if desired.

Acknowledgements

Thanks to Defra and the Natural Environment Research Council for providing projects and questions which led to the original development of the code in this package. The package is based around tidyverse ideas and functions, so thanks go also to Hadley Wickham and the tidyverse team for building and maintaining this incredible environment.

Contribute

If you would like to contribute to this package, please file an issue, make a pull request on GitHub, or contact the author on Twitter @alasdair_sykes.