Detection and Monitoring of Intense Pyroconvection using Remote Sensing and Numerical Weather Prediction

Fire-triggered thunderstorms, known as pyrocumulonimbus (pyroCbs), produce a high injection altitude of smoke particles, influence smoke plume trajectory, and are detrimental to fire suppression efforts. PyroCbs may also inject smoke into the upper troposphere or even lower stratosphere, where wildfires are a key source of particulate intrusions. However, the frequency of pyroCb occurrence and the meteorological conditions driving their development are still uncertain. The Naval Research Laboratory is developing a near-real-time regional pyroCb detection system using satellite observations and numerical weather prediction (NWP). PyroCb detection begins with the characterization of observed fire activity via a normalized time series of fire radiative power (FRP) obtained from geostationary satellites. A high-altitude cloud, defined as having a thermal infrared brightness temperature suitable for ice formation, must also be anchored to the region of high FRP. The height of the pyroCb can be inferred from an NWP-derived atmospheric profile at the location of the fire. Pyroconvection will have a different spectral signature than traditional convection because of a large quantity of smoke-induced small particles within the cloud. Observations from the Geostationary Operational Environmental Satellites (GOES) show that pyroCb development can occur at any point in a fire's lifetime, but the fire must be energetic enough to maintain a robust convective column. While GOES pyroCb detection is possible over much of the Western Hemisphere, its accuracy is reduced in high latitude boreal regions (e.g. north of 50°N latitude) where viewing angles are large. The combination of the pyroCb detection system and NWP data show that the majority of pyroCbs in North America occur with conditions very similar to those that produce dry thunderstorms, which is a critical step toward producing a regional pyroCb prediction system.