Table 1 Summary of the variables used in this study (and by the proposed model) together with their sources.

From: Global lightning-ignited wildfires prediction and climate change projections based on explainable machine learning models

Group

Variable

Acronym

Used in model

Source

Wildfire data

Lightning resulted in ignition (target feature)

Burned

v

31

Total burned in fire

Total_burned

x

 

Area burned in first day of fire

First_day_burned

x

 

Fire duration

Duration

x

 

Thunder data

Thunderhours

Thunderhours

x

30

Vegetation, meteorological, and anthropogenic factors

Relative humidity

RH

v

32

Temperature

t

v

 

Precipitation

Prec

v

 

Days since last precipitation

Days_since_prec

v

 

Monthly precipitation

Monthly_prec

v

 

Wind velocity

v_total

v

 

Wind direction

v_direction

v

 

Low vegetation cover

Low_veg

v

 

High vegetation cover

High_veg

v

 

Normalized Difference Vegetation Index

NDVI

v

33

Soil moisture

sm

v

32

Vegetation water content (skin reservoir)

Water

v

32

Population density

Pop

v

34

Historical fire data

Mean burned area in years 2003–2013

Historical_fires

v

35

Fire weather indices

Fire weather index

fwi

v

32

Buildup index

bui

v

 

Drought code

drought

v

 

Duff moisture code

duff

v

 

Fine fuel moisture code

ffmc

v

 

Initial spread index

isi

v

 

Spatiotemporal data

Latitude

lat

v

–

Longitude

lon

v

–

Month

Month

v

–