Basic Workflow

Basic workflow --> identify available land (no timeseries data)

This guide walks through the end-to-end workflow for running the Land Availability Analysis (LAVA) tool on a new study region. The steps below assume that you have already cloned the repository, created the lava environment and finished the data setup. Those steps are documented in the Getting Started instructions.

Overview of the basic workflow

1. Create the study-region configuration files in configs/.
2. Run spatial_data_prep.py to download and prepare necessary data.
3. Inspect the pre-processed data (optional but recommended) with data_explore.ipynb.
4. Run Exclusion.py for each technology to create available-land rasters.

Figure: Flowchart of the basic workflow

Configuration files

In the configs-folder copy the file config_template.yaml, rename it to config.yaml and fill it out. This is your main configuration file for the data download. Under the headline #--exclusions-- in the configuration the variables scenario and technology are used to control the available land output.

For the exclusion criterias copy the files onshorewind_template.yaml and solar_template.yaml. Rename them to onshorewind.yaml and solar.yaml respectively. Fill these files out in order to set the exclusion parameters.

Note

An overview of possible exclusion criterias found in selected literature can be downloaded here.

Prepare data

Run the preprocessing script spatial_data_prep.py after having filled out config.yaml. It downloads data, clips the raw inputs to the study area, aligns rasters, and computes helper layers such as slope, terrain ruggedness, and proximity rasters. All outputs are written to a study region specific folder data/{RegionName}/.

Click on the play button to run the script or run it from the terminal with the following command:

python spatial_data_prep.py

When landcover_source is openeo the script will prompt for Copernicus Data Space credentials the first time it runs.

Inspect data inputs

Use data_explore.ipynb to verify the preprocessing results. The notebook loads data from the data/{RegionName}/ folder, visualises selected layers, and summarises the available land-cover codes to support tuning of exclusion thresholds.

Land eligibility exclusions

Create technology-specific available-land rasters by running Exclusion.py. The command-line flags mirror the configuration entries so that single technologies or scenarios can be processed independently. Alternatively you can specify in config.yaml under the headline #--exclusions-- the variables "scenario" and "technology". There are configuration files for the land exclusion for wind onshore and utility-scale solar PV.

python Exclusion.py --technology solar --scenario ref

python Exclusion.py --technology onshorewind --scenario ref

The script loads the prepared rasters and vector layers, applies the filters defined in the technology configuration, and writes *_available_land_*.tif files under data/{RegionName}/available_land/. A log of each scenario run is stored in the region folder for traceability.

Note

A template to document the exclusion parameters and data sources can be downloaded here.

Batch processing with Snakemake

When analysing multiple study regions, you can use a snakemake workflow to automatically execute all scripts for all regions one after another. Adjust config_snakemake.yaml and run the batch job with the following comamand in the terminal:

snakemake -s snakemake/Snakefile --cores 1 --resources openeo_req=1

You can parallelize by using multiple cores. You can use a maximum of 2 openeo_req to download the landcover data.