Machine Learning/AI Capabilities: using satellite- and sensor-based wildfire data to improve impact tracking

For decades, satellite systems and other remote sensing technologies have provided critical data points and analytics in the detection of wildfires and real-time tracking of fire spread. Since satellite-based systems are extremely important pre-, during-, and post-wildland fire incidents, there remains a need to improve data interpretation and combine with other datasets to produce actionable information about risks and impacts to local populations.

How can we use ML/AI with satellite and/or other wildfire-specific data sources to identify and assess second and third level impacts to further detail populations at risk? What new solutions can you develop?
What machine learning, AI, or geofencing technologies can be leveraged to track downstream wildfire impacts?

Can you develop an alert or triggering system which informs local communities about certain downstream hazards or potential impacts from a wildfire incident?

Can you develop an algorithm, application or tool that provides advanced indicators of 2nd or 3rd level impacts from wildland fires? Utilize a variety of datasets (air quality, soils, vegetation types, population, etc.) and satellite remote sensing technologies (fire spread, burnt areas, debris flow, etc.).