shahinmg / icebergs_allie

iceberg data

allie icebergs

I have saved Sid's iceberg data as geoparquet files due to their speed and storage efficiency. These can be read with geopandas and in QGIS. Though to use with QGIS, you need to have it compiled with a version of GDAL > 3.6 and the libgdal-arrow-parquet plugin. I created two yml files that will create two seperate conda environments. One for QGIS and one for geopandas. I try not to mess with my conda environments where QGIS is installed which is why we are installing two environments. To make these environemts, you first need to install anaconda if you don't already have it installed.

1) Clone the repo

Open a terminal or anaconda prompt if on Windows then follow the commands below

git clone https://github.com/shahinmg/icebergs_allie.git

and cd to the directory

cd icebergs_allie

Create QGIS environment

Note: Be patient. These may take some time

conda env create --file qgis_environment.yml

to run the new qgis install first activate the environment

conda activate qgis

then run the qgis command

qgis

Create geopandas environment

Now we need to make a seperate environment to work with the data in python. Run the command below to creat the new environment

conda env create --file icebergs.yml

conda activate icebergs

Visualize the data

Since the data are quite large, we should use the library polars. We will use the scan_parquet method to read our data, use grouby to sort the data by date and then take the sum of the area. Use the polars_plot.py to make the figure in the figs directory. Place the parquet files in the geo_parquets directory to run the script. The polars_plot.py will also save the very small (1 kb) time series csv into the time_series_csvs directory. Area units are in meters^2.