Skip to content

Code to support prospective cost-effectiveness analysis of farmed finfish stunning intervention in Europe.

License

Notifications You must be signed in to change notification settings

rethinkpriorities/farmfish_stunning_cea

Repository files navigation

Github repository: Farm fish stunning in Europe CEA

Sagar Shah, Rethink Priorities, February 2024

About

This is the README file for the Github Repository containing data and code used in the publication: Prospective cost-effectiveness of farmed fish stunning corporate commitments in Europe.

Analysis was conducted in R and code was written in a mixture of R scripts (.R) and Quarto markdown (.qmd).

In order to edit and run the code, I suggest you download the entire repository to your local drive. Be sure to include the .Rproj file to ensure file references operate correctly.

View rendered versions of script

For anyone interested in looking at the code or results in more detail (but not necessarily interested in downloading, editing and running the script) the rendered .html files are probably the best starting place. You can either download them from the respository and view them locally, or see rendered versions via the links below.

Note that most of these scripts rely heavily on functions and variables defined in three scripts (a_functions, b_fish_assumptions, and c_model_assumptions) in the 2_code folder.

Description of contents each folder in this repository

Folder name Contents
1_input_data Data that inputs into the analysis
2_code Contains the code for the analysis (written in R, either as base R script or as a Quarto markdown file.
3_intermediate_data Contains processed data (in.rds format) that is an output of one script file but an input into another.
4_charts Charts and tables that are used in the final report.

Input_data folder contents

File Contents
EUFOMA_consumption_data.xlsx Per country apparent consumption estimates for various farmed fish species, mostly from European Market Observatory for fisheries and aquaculture (EUMOFA) market study reports:
- Seabass (2016)
- Seabream (2019)
- Carp(2018)
- Large trout in the EU (2020) - with small rainbow trout consumption estimated using the difference between all trout and large large trout consumption.

Also contains salmon consumption estimates from the chart from page 42 of the Mowi (2023) industry report for France, Germany, Italy, Spain and Sweden only.

And contains EU country codes (from Eurostat
chickens_per_dollar.csv Raw samples from the Guesstimate model estimating the historical cost-effectiveness of chicken campaigns from Šimčikas (2019). Each column contains 5000 rows, with each row representing an individual monte carlo simulation result:
-- Total chickens affected per dollar
- Broilers affected per dollar
- Layer hens affected per dollar
- Also contains the total number of hens affected by US and non-US cage free commitments.
country populations.xlsx Population data for EU27 countries to estimate salmon consumption (from Eurostat) in each EU27 country.
fishcount_living.xlsx Spreadsheet from fishcount.org, providing estimates of the mean slaughter weight and mean lifespan of the 24 most commonly farmed finfish species.|
welfarefootprintdatachickens.xlsx Datasets from the Welfare Footprint Project, estimating time in pain per system for conventional and reformed layer hens and broiler chickens respectively.

Code folder contents

Folder name Contents
a_functions.R Contains functions used throughout the analysis, including:
- setting number of simulations
- seed for random number generation
- allowing statistical distributions to be defined using mean and standard deviation or 90% CI.
- Plot functions to quickly visualize distributions or quantiles of a distribution.
- Functions to save charts or tables.
- Functions to call vectors
b_fish_assumptions.R Assumptions regarding of seabream, seabass and small rainbow trout, including:
- Aggregate consumption in France, Italy and Spain
- Typical weight (live weight equivalent) at slaughter
- Slaughter duration
- Share stunned
- Typical lifespan
c_model_assumptions.R Defines modelling assumptions used in the cost-effectiveness analyses, including:
- Implementation discount
- Number of years of credit
- Share of consumption going through grocery sector
- Market share of supermarkets who make commitments
- Probability of success
- Program costs
- Duration of intervention impact (relative to slaughter duration or lifespan)
- Impact of intervention (as a share of fish welfare range)
1_prep_consumption_data.qmd Produces estimates and charts of total consumption of 5 farmed finfish species by country across EU27.

Note that most of the code in this file estimates consumption of Salmon across the EU27, but this analysis did mostly not make it into the final publication.
2_intro_charts.qmd Produces charts used in the introduction section of the report.
3_main_results.qmd Main cost-effectiveness analysis file. Produces charts and tables used in the Results section of the report.
4_annex_results.qmd Cost-effectiveness analysis of non-slaughter intervention contained within Annex 2 of the report.
5_time_pain_exploration.qmd Exploratory analysis of extent to which time in different pain intensities might compare for fish stunning intervention to historical chicken interventions.
d_render.R Script that renders all quarto codes into html format.

About

Code to support prospective cost-effectiveness analysis of farmed finfish stunning intervention in Europe.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published