Skip to content

Analyse-kode bak «Slik endres Norge», VGs oversikt over værendringer i Norge

License

Notifications You must be signed in to change notification settings

vgno/senorge-downloader

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

senorge-downloader

This code downloads NetCDF files from MET and converts them to aggregated CSVs, then builds VG outputs.

Setup the project

pip install -U poetry
poetry shell
poetry install

System requirements

The dataset is about 150 GB and some of the steps are memory intensive. We recommend at least 32 GB of RAM.

Usage

The process has four steps:

1. Download data

We currently fetch three both seNorge and seNorge_snow datasets. We are interested in the following variables:

  • tg: Gridded average temperature by day
  • rr: Gridded sum daily precipitation amount
  • sd: Gridded snow depth by day (used to calculate snow and ski days)
poetry run senorge download -output ./data/raw

You can also download a single variable:

poetry run senorge download -output ./data/raw -variable tg

2. Convert NetCDF to CSV

We convert the downloaded files to CSVs and output one per year.

We resample the daily data to yearly, so that we end up with the following data for each year and grid cell:

  • mean daily temperature
  • sum daily precipitation
  • mean snow depth
  • count of ski days (depth > 25 cm)
  • count snow days (depth > 5 cm)
poetry run senorge csv -input ./data/raw -output ./data/csv

3. Find the largest grid cell in each municipality

This step currently requires PostgreSQL with PostGIS support:

scripts/find-largest-pop-grid.sh

4. Build VG-formatted outputs

NB. You need PostgreSQL with n50 data with kommuner_uten_hav from our election maps to complete the vg-maps and step.

We create the following outputs:

  1. Municipality timelines:
    • Yearly timelines for all five variables for the most densly populated grid cell
    • Only absolute values / no diffs
  2. National timelines:
    • Created by averaging the municipalities in the previous step
  3. Grid diffs
    • the absolute and relative diffs of 1960-1969 compared to the last ten years (currently 2013-2022) for all grid cells
    • Used for 3D maps
  4. Downloaded world comparisons and Norway projections
  5. Top lists of changes
poetry run senorge vg-timeseries -input ./data/csv -output ./data/vg
poetry run senorge vg-maps -input ./data/csv -output ./data/vg
poetry run senorge vg-comparisons -output ./data/vg
poetry run senorge vg-top-lists -output ./data/vg