Darius A. Görgen 2024-10-16
This repository contains the codes used to calculate local indicators of
conflict exposure. Conflict event data are taken from
ACLED and indicators are calculated based on
administrative units of level 2. Popluation data is based on
WorldPop. Project informations represents
data internal to KfW. Since most of the used data cannot be shared
publicly, this repository only containes the code in order to reproduce
the results. The conflict indicators, however, can be calculated using
{mapme.biodiversity}
and a valid ACLED account with API access.
As indicators of local conflict exposure we calculate:
- total number of aggregated fatalities per unit and year
- number of fatalities per 100,000 inhabitants per unit and year
- total number of population exposed to certain types of events per unit and year
- number of people exposed to these events per 100,000 inhabitants per unit and year
based on the following configuration file which is used as input to
{mapme.pipelines}
input: data/adm2_polygons.gpkg
output: output/adm2_polygons_conflict_indicators.gpkg
datadir: /home/rstudio/mapme/data
batchsize: 5000
options:
overwrite: true
maxcores: 10
progress: true
chunksize: NULL
resources:
get_acled:
args:
years: [2019, 2020, 2021, 2022, 2023]
accept_terms: TRUE
get_worldpop:
args:
years: [2019]
indicators:
calc_exposed_population_acled:
args:
distance: [5000, 2000, 5000, 3000]
years: [2019, 2020, 2021, 2022, 2023]
filter_category: event_type
filter_types: [battles, riots, explosions/remote_violence, violence_against_civilians]
precision_location: 1
precision_time: 1
calc_fatalities_acled:
args:
years: [2019, 2020, 2021, 2022, 2023]
stratum: event_type
precision_location: 1
precision_time: 1
calc_population_count:
args:
engine: exactextract
stats: sumTo fetch ACLED data successfully, the code expects to find a file called
.env in the top-level with the following valid content to be set:
ACLED_ACCESS_EMAIL=<your-email>
ACLED_ACCESS_KEY=<your-key>