IRI-Danger: Probability of Natural Ignition (PNI) [ET] [Version 1]

The Fire Forecasting Model based Natural Ignition Probability product predicting the wildfires ignition probability by natural causes through basic meteorological parameters provided by Numerical Weather Predicting models. The product was developed within the Activity “A1.1.2 Natural fire ignitions: Lightning”. FFM employs a multi-step machine learning procedure to construct a statistical model that predicts Fire Radiative Power (FRP) based on global ERA5 reanalysis (https://www.ecmwf.int/en/forecasts/dataset/ecmwf-reanalysis-v5), ERA5-Land reanalysis (https://www.ecmwf.int/en/era5-land), IFS operational forecast (https://www.ecmwf.int/en/forecasts/documentation-and-support/changes-ecmwf-model), Terra's Moderate Resolution Imaging Spectroradiometer (MODIS, MOD14/MYD14, collection 6, https://www.earthdata.nasa.gov/learn/find-data/near-real-time/firms). The model utilises various meteorological parameters, including Cloud-to-Ground Lightning Flash Density and Fire Danger Indices, as predictors for training and making predictions by calculating their respective contributions to total FRP. The contribution of cloud-to-ground lightning flash density to FRP is then used as a proxy for natural ignition probability. Contact details of the developer: evgeny.kadantsev@fmi.fi