Skip to contents

This packages combines data collected as part of an MSc. Thesis Project and an MSc. Semester Project conducted in Durban, South Africa. The projects were supported by the Global Health Engineering group at ETH Zurich, Switzerland.

Installation

You can install the development version of durbanplasticwaste from GitHub with:

# install.packages("devtools")
devtools::install_github("Global-Health-Engineering/durbanplasticwaste")

Alternatively, you can download the individual datasets as a CSV or XLSX file from the table below.

dataset CSV XLSX
litterboom_counts Download CSV Download XLSX
litterboom_weights Download CSV Download XLSX
locations Download CSV Download XLSX
plastic_types Download CSV Download XLSX

Projects

MSc. Thesis Project

Evaluating the potential of Extended Producer Responsibility returns for a small local waste collection company through a brand audit of riverine plastic waste in Durban, South Africa.

Description

This Master’s Thesis Project focuses on determining the growth opportunities for a small-sized plastic recycling enterprise in light of the shift from a voluntary to a mandatory Extended Producer Responsibility (EPR) policy in South Africa.

To achieve this goal in the context of a small start-up in Durban, South Africa , a brand audit is conducted to identify the top brands that can be targeted for financing or partnership opportunities. The company, called TRI ECO Tours, is a small tourism and waste collection startup in Durban operated by Siphiwe Rakgabale.

Research Question

What is the characterization by type, application, and brand of plastic waste collected in the uMngeni River system in Durban, South Africa?

Data

The data was collected throughout two months in Durban, South Africa right before the rainy season. The collection took place in 6 different litterboom locations throughout Durban. The data gathered was the audit of the occurence of the brands washed into the litterbooms.

The package provides access to three data sets.

The litterboom_counts data set has 7 variables and 2784 observations. For an overview of the variable names, see the following table.

litterboom_counts
variable_name variable_type description
date date Date of the collected litterboom sample.
location character Descriptive name of the sample location. See [locations] for longitude and latitude.
brand character Brand name of the collected item (e.g. Coca Cola).
group character Group name that owns the brand (e.g. Coca Cola Beverages South Africa).
plastic character Type of plastic of the item (identified plastic types are PET; HDPE; and PP. HDPE and PP were categorised together as HDPE/PP.
category character Categorisation of waste into 15 product type categories (e.g. Alcohol; Milk; Tobacco; Water).
count numeric Number of counted items.

The litterboom_weights data set has 4 variables and 14 observations. For an overview of the variable names, see the following table.

variable_name variable_type description
date date Date of the collected litterboom sample.
location character Descriptive name of the sample location.
pet numeric Weight (in kg) of PET items.
hpde_pp numeric Weight (in kg) of PET items.

The locations data set has 3 variables and 6 observations. For an overview of the variable names, see the following table.

variable_name variable_type description
location character Descriptive name of the sample location.
latitude numeric Latitude coordinate.
longitude numeric Longitude coordinate.

Locations data as a map illustrating the six litterboom sampling locations in Durban, South Africa. For an interactive map and other code examples, see vignette("examples").

MSc. Semester Project

Examination of non-recycled marine plastic litter in order to identify recycling and beneficiation pathways in Durban, South Africa

Description

This Semester Thesis Project focuses on determining the distribution of plastic litter on the Durban beachfront in order to identify key targets for policy and financial support through the South African EPR policy to reduce plastic spills into the environment and promote higher recycling rates.

Research Question

What types and amounts of plastic are found along the beachfront in the mangroves of Durban-North, South Africa?

Data

The data was collected in collaboration with a local team that provided the waste for the study. Two different approaches were used to clean up the waste. One to collect the waste floating on the river, preventing it from reaching the sea and one to collect waste washed ashore at the beaches. Three locations along the rivers close to settlements and one at the beach in Durban North.

The package provides access to one data set of this project.

The plastic_types data set has 16 variables and 13 observations. 11 observations were made on the beach which were added together in the row that has “Beach Total” as the value for the variable “place”. One observation was done at a litterboom. For an overview of the variable names, see the following table.

plastic_types
variable_name variable_type description
date date Collection date of the data.
place character Place where the collection was done.
total double Total amount of waste chategorized in kg.
pet double Amount of polyethylene terephthalate (PET) in kg.
pe double Amount of polyethylene (PE) in kg. During the sorting process HDPE and LDPE plastics were not distinguished.
pp double Amount of polypropylene (PP) in kg.
ps double Amount of polystyrene (PS) in kg.
hdpe_pp double Amount of high-density polyethylene (HDPE) and polypropylene (PP) combined weight in kg. During the sorting process these two plastics were not distinguished.
other_plastics double Amount of other plastics in kg. The category ‘Other plastics’ included plastics that couldn’t be identified by the scanner or plastics that are in such small amounts the weighing scale could not distinguish it.
other_waste double Amount of other waste in kg remaining at the end of the categorization.
glass double Amount of glass in kg.
shoes double Amount of shoes in kg.
shoes_quantity double Number of shoes found at the location. Surprisingly a lot of shoes were found. As they often consisted of a combination of different plastics then a separate category was made for shoes.
bag_quantity double Number of bags that where collected.
beach logical Categorical variable showing whether the collection was done at the beach (TRUE) or at the a litterboom (FALSE).
details character Additional information about the sampled data.

The following plot is showing the distribution of the waste weight for different plastic and waste types at the Beachwood Mangrove across the sampling period (September 19th 2022 - October 17th 2022) in percent.

Examples

The litterboom_counts data identifies 40 unique groups that own the identified brands. The top 10 brands are shown in the following table. All other brands are lumped together as OTHER.

library(durbanplasticwaste)
library(dplyr)
library(forcats)

litterboom_counts |> 
  mutate(group = factor(group)) |> 
  mutate(group = fct_lump(group, n = 10, other_level = "OTHER")) |> 
  group_by(group) |> 
  summarise(
    count = sum(count)
  ) |> 
  arrange(desc(count)) |> 
  mutate(percent = count / sum(count) * 100) |> 
  knitr::kable(digits = 0)
group count percent
OTHER 8086 52
Coca Cola Beverages South Africa 4030 26
unidentifiable 1202 8
Clover Industries LTD 737 5
Unilever 442 3
Tiger Brands 232 2
danone 183 1
Siqolo Foods 144 1
Willowton Group 139 1
Amka Products 132 1
RCL Foods 95 1

License

Data are available as CC-BY.

Citation

To cite this package, please use:

citation("durbanplasticwaste")
#> To cite package 'durbanplasticwaste' in publications use:
#> 
#>   Bergen R, Schöbitz L, Meyer-Piening C, Boynton L, Tilley E, Kalina M,
#>   Rakgabale S, Luvuno S, Loos S (2023). "durbanplasticwaste. Durban
#>   (South Aftica) Plastic Waste Data." doi:10.5281/zenodo.7845779
#>   <https://doi.org/10.5281/zenodo.7845779>,
#>   <https://global-health-engineering.github.io/durbanplasticwaste/>.
#> 
#> A BibTeX entry for LaTeX users is
#> 
#>   @Misc{R-durbanplasticwaste,
#>     title = {durbanplasticwaste. Durban (South Aftica) Plastic Waste Data},
#>     author = {Raúl Bergen and Lars Schöbitz and Chiara Meyer-Piening and Lin Boynton and Elizabeth Tilley and Marc Kalina and Siphiwe Rakgabale and Sfiso Nduduzo Luvuno and Sebastian Camilo Loos},
#>     doi = {10.5281/zenodo.7845779},
#>     url = {https://global-health-engineering.github.io/durbanplasticwaste/},
#>     abstract = {Waste characterisation data for extended producer responsibility.
#> The plastic waste was collected in the uMngeni River system in Durban,
#> South Africa, and categorised by brand, corporate group,
#> application, and type of plastic.},
#>     version = {0.1.1},
#>     year = {2023},
#>   }