The potential destruction posed by climate change has found an unlikely savior in artificial intelligence (AI). The California Department of Forestry and Fire Protection (Cal Fire) recently unveiled a groundbreaking initiative — the Alert California AI program — developed in collaboration with the University of California San Diego (UCSD). This pioneering initiative harnesses the power of AI to spot wildfires in their early stages.
- Technology in Action: Using live feeds from a network of 1,032 360-degree rotating cameras, the AI program identifies discrepancies and anomalies within these feeds.
- Quick Response: On detecting potential threats, emergency services and relevant authorities are promptly notified. This program commenced in July and promptly acted to halt a potential wildfire in the Cleveland National Forest.
- Underlying Technology: The Alert California technology website highlights the use of LiDAR scans, sourced from airplanes and drones, to generate precise, 3D information about the terrain. Combined with data about tree species, this gives valuable insight into California’s forest biomass and carbon content.
- Funding and Support: With an investment surpassing $20 million over the past four years and upcoming funding of an additional $3,516,000, Cal Fire solidly backs this program.
Microsoft’s AI Intervention Post Lahaina Wildfire
The devastating aftermath of the Lahaina wildfire witnessed the deployment of Microsoft’s AI for Good Research Lab’s advanced AI tools to gauge the damage inflicted by this catastrophe.
- AI-powered Damage Assessment: In tandem with the American Red Cross and Planet, Microsoft utilized advanced AI techniques to provide a preliminary damage assessment. Detailed maps showcased the status of buildings in the affected regions were created based on sophisticated AI models and satellite images.
- Damage Breakdown: Of the 2,810 buildings surveyed: 1,088 buildings experienced 0%-20% damage. 110 buildings sustained 20%-40% damage. 169 buildings faced 40%-60% damage. 238 buildings suffered 60%-80% damage. 1,205 buildings endured 80%-100% damage.
- Limitations and Further Validation: Satellite data has its limitations in providing a comprehensive damage assessment. Thus, these AI-generated evaluations serve as initial guidelines, necessitating on-site verification for complete accuracy.
AI’s Rapid Response and Future Prospects
The early detection capabilities of AI have proven critical in combating wildfires. Its ability to notify authorities before they become extensive threats has been demonstrated multiple times since the program’s initiation in July.
- Cal Fire’s Vision: Cal Fire envisions the technology being emulated by other countries and states, addressing the growing concern of wildfires on a global scale, as witnessed recently in Hawaii, Canada, and the Mediterranean.
- AI’s Learning Curve: Suzann Leininger, a Cal Fire intelligence specialist, highlighted the continuous learning process of the AI system. Regular reviews of the camera feeds by specialists statewide have led to the AI’s increased accuracy over a short period.
- Extensive Data Collection: Significant data collection efforts are underway from the camera network. These include aerial surveys to analyze vegetation, atmospheric rivers, snowpack measurements, and assessment of post-burn scars. This data, available for public and private entities, has the potential to refine AI applications for environmental studies.
Neal Driscoll, geology and geophysics professor at UCSD and the program’s principal investigator, emphasized the current climatic challenges. He asserted, “We’re in an extreme climate right now. So we give them the data because this problem is bigger than all of us. We need to use technology to help move the needle, even if it’s a little bit.”
In conclusion, the ongoing developments in AI offer a beacon of hope in the face of growing climatic adversities. These technological advancements, in partnership with dedicated efforts from various organizations, promise a safer and more responsive future in addressing environmental catastrophes.