OSVAS

LAI Estimation - Quick Reference

TL;DR - Enable LAI Estimation

1. Configure Your Station

Edit your station YAML (e.g., config_files/Stations/Cabauw/Cabauw.yml):

Station_metadata:
  ...
  estimate_lai: true

2. Run OSVAS Workflow

export STATION_NAME=Cabauw
export OSVAS=$HOME/OSVASgh
export HARPSCRIPTS=$OSVAS/HARPSCRIPTS

python3 scripts/python_scripts/surfex_OSVAS_run_linux.py

The workflow automatically:

1. Downloads validation data (Step 2)
2. Estimates surface albedos from observations (Step 2b)
3. Estimates LAI from CGLS satellite data (Step 2c) ✨
4. Updates namelists with estimated LAI values
5. Runs SURFEX simulations with new LAI parameters (Step 3)

3. First Run: Authenticate

On first run, OIDC authentication is required:

1. Script prints a URL and device code
2. Visit the URL in a browser and enter the code
3. Credentials are cached automatically

Subsequent runs use cached credentials automatically.

What Gets Computed?

• Data source: CGLS LAI from Sentinel-3 OLCI + PROBA-V via openEO BIOPAR process
• Spatial resolution: 300 m
• Temporal resolution: 10-day composites
• Processing: All valid 10-day observations per month → monthly average
• One parameter:
  • XUNIF_LAI - Vegetation Leaf Area Index [m²/m²]

Output Files

File	Purpose
namelists/{station}/lai_estimates.nam	Raw LAI namelist blocks
namelists/{station}/OPTIONS.nam_{expname}.backup_lai	Backup of original namelist
namelists/{station}/OPTIONS.nam_{expname}	Updated with new LAI

Disable LAI Estimation

Set estimate_lai: false in Station_metadata to skip this step (not yet implemented in launcher—edit manually).

Key Parameters

Parameter	Value	Notes
Data source	CGLS LAI 300m (BIOPAR)	Via openEO, Copernicus Data Space
Fetch method	10-day composites	Physical LAI units (m²/m²)
Monthly aggregation	Mean of 10-day values	Simple average, gap-filled if needed
Typical grass LAI	0.2-0.5 (winter), 2-4 (summer)	Depends on phenology
Typical crop LAI	0.1-0.5 (harvest), 1.5-3.5 (growing)	Seasonal vegetation cycle
Saturation limit	~6-8 m²/m²	Dense forest max (CGLS may saturate)

Troubleshooting

Problem	Solution
“No LAI data found”	Extend period: --run-period 2017-01-01 2018-12-31 or use multi-year: --start-year 2017 --end-year 2022
“Authentication failed”	Check internet; ensure Copernicus account active (free at https://dataspace.copernicus.eu/)
“LAI values unrealistic”	Verify vegetation type matches site; check seasonal patterns plotted
Namelist update fails	Verify vegtype is consistent between LAI file and namelist

Manual Commands

Generate LAI estimates only

python3 scripts/python_scripts/estimate_lai.py Cabauw $OSVAS --run-period 2017-11-01 2018-01-31

Update namelists only

python3 scripts/python_scripts/update_namelist_lais.py Cabauw $OSVAS

Use pre-computed LAI values (no openEO connection needed)

python3 scripts/python_scripts/estimate_lai.py Cabauw $OSVAS \
    --print-only \
    --monthly-lai '{"1":0.5,"2":0.4,"3":0.6,"4":0.8,"5":1.2,"6":1.5,"7":1.6,"8":1.4,"9":1.0,"10":0.7,"11":0.5,"12":0.4}'

Full Documentation

See docs/step2c_lai_estimation.md for detailed information.