A Datacube for the analysis of wildfires in Greece

dataset_greece.nc

This dataset is meant to be used to develop models for next-day fire hazard forecasting in Greece. It contains data from 2009 to 2020 at a 1km x 1km x 1 daily grid.

Check our Jupyter notebook for an example showing how to access the dataset.

====================================================================================

Dynamic Variables

IMPORTANT NOTE: The Fire, Meteorological Variables and Fire Weather Index have been shifted one day back to ease the development of the models. This is to ease the development of our models, because operationally Meteorological variables and the Fire Weather Index are available as forecast and the Fire Variables are what we want our models to forecast given all the other variables.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------

It includes the following dynamic variables resampled at daily temporal resolution and 1km spatial resolution:

1. Previous day Leaf Area Index - MOD15A2H Variables (https://lpdaac.usgs.gov/products/mod15a2hv006/)

Fpar_500m
Lai_500m
FparLai_QC
FparExtra_QC
FparStdDev_500m
LaiStdDev_500m

2. Previous day MOD13A2 Variables (https://lpdaac.usgs.gov/products/mod13a2v006/)
1 km 16 days NDVI
1 km 16 days EVI
1 km 16 days VI Quality

3. Previous daty Evapotranspiration. MOD16A2 Variables (https://lpdaac.usgs.gov/products/mod16a2v006/)
ET_500m
LE_500m
PET_500m
PLE_500m
ET_QC_500m

4. Previous day Land Surface Temperature. MOD11A1 variables (https://lpdaac.usgs.gov/products/mod11a1v006/)
LST_Day_1km
QC_Day
LST_Night_1km
QC_Night

5. Meteorological data. ERA5-Land variables (https://cds.climate.copernicus.eu/cdsapp#!/dataset/reanalysis-era5-land?tab=overview)
era5_max_u10
era5_max_v10
era5_max_t2m
era5_max_tp
era5_min_u10
era5_min_v10
era5_min_t2m
era5_min_tp

6. Fire variables
ignition_points Ignition points derived from the association of burned areas product from EFFIS (effis.jrc.ec.europa.eu/) with FIRMS active fire product.
burned_areas: Burned areas from EFFIS (effis.jrc.ec.europa.eu/), associated with FIRMS active fire product to find ignition date
number_of_fires: Count of fire events for the given day.

7. Fire Weather Index (https://cds.climate.copernicus.eu/cdsapp#!/dataset/cems-fire-historical?tab=overview)
fwi

------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Static Variables

It includes the following static variables resampled at 1km spatial resolution:

1. clc_YYYY for years 2006,, 2012, 2018: Corine Land Cover. (https://land.copernicus.eu/)

2. roads_density_2020: raster derived from OpenStreetMaps polygons for 2020. (https://www.openstreetmap.org/)
3. population_density_YYYY for years 2009-2020: population density at 1km spatial resolution. Source - https://www.worldpop.org/

4. Topography layers derived from EU-DEM. (https://land.copernicus.eu/)

dem_{agg}, aspect_{agg}. slope_{agg}, where agg is mean (mean value), std (standard deviation), max (maximum value), min (minimun value) and specifies the applied aggregation for the resampling to 1km.