Hello,  

Wildfires, hurricanes, and heat waves—oh my! Extreme weather seems to be happening more frequently.

But what if we could see it coming sooner and plan smarter?

That’s where AI comes in.

Artificial intelligence systems learn from vast amounts of data and continuously improve their performance through algorithms and models. Machine learning is a specific AI technique that uses algorithms to learn patterns from data and uses those patterns to make predictions.¹

As people who like to plan for the expected and unexpected, we wanted to look deeper into the application of AI in weather forecasting, especially as wildfire and hurricane season approaches.

Weather can be unpredictable…Or is it?

We often rely on weather apps or trained meteorologists to tell us what kind of weather we can expect today and for the next 5-10 days.

But what tools are they using?

AI has quickly become one of those key tools. These systems excel at processing and analyzing large volumes of data more efficiently from various sources, such as weather satellites, weather stations, ocean buoys, and historical data.¹

Meteorologists continue to use NWP models, which utilize physics to understand atmospheric processes, but the addition of AI is further enhancing their weather forecasting capabilities.²

The Society for Industrial and Applied Mathematics (SIAM) is a community of researchers that’s been developing other models to improve our ability to simulate weather events with greater accuracy and forecast wildfire risks.³

At the 2024 SIAM Conference, Jatan Buch of Columbia University highlighted how machine learning can help predict and manage wildfires with the introduction of two models: SMLFire and SEASFire.⁴ 

1. SMLFire simulates wildfire behavior
   • Includes 51 predictors classified into categories:
     • Hot
     • Dry
     • Indices
     • Other
   • Incorporates measures of vegetation, topography, and human-related activities
2. SEASFire simulates fire frequency and size
   • Incorporates prior observations, the SEAS5 ensemble (a European seasonal forecasting system), and SMLFire

These tools aim to improve fire risk planning by accounting for uncertainty. Buch also mentioned a study with Kyleen Liao of Stanford University that used AI to predict wildfire smoke emissions, offering insights into when prescribed burns would have the least impact on air quality.⁴

Together, these models support smarter, data-driven wildfire management. Satellite technology is also being used in conjunction with AI to help with wildfire detection and mapping on the continental west coast and in Hawaii. 

Check out this 2 minute 18 second video from ABC7 News Bay Area about this application.

Another AI-based weather forecasting model, FourCastNet, is known to make weather predictions at a spatial resolution eight times higher than previous AI models.⁵ Its forecasts offer more detailed and timely information that have proved beneficial in early detection of hurricanes.

For example, the image below shows the forecast from FourCastNet of wind speeds 96 hours in advance of what turned out to be Typhoon Mangkhut (Inset 1) and Hurricanes Florence, Isaac, and Helene (Inset 2) in 2018.⁶

WWLTV in New Orleans, Louisiana—an area considered high-risk for hurricanes—reported on how the use of AI is revolutionizing their weather forecasting.

Watch the 2 minute 46 second report here.

From predicting wildfire spread to issuing early hurricane warnings, artificial intelligence is not just a tool—it's a partner in our effort to mitigate risks and plan for better protection.

The storms may come, but thanks to AI, we are more prepared than ever to weather them together.

Sincerely,

Your Eagle Wealth Team

The Road to True Artificial Intelligence

The next milestone in the fast-moving AI landscape is likely to be artificial general intelligence (AGI), sometimes defined as the point at which machines reach superior levels of intelligence across a wide array of subjects. Unlike today’s AI that is designed for specific tasks, AGI has cognitive capabilities of reasoning and problem-solving that can rival that of a human.

There’s an ongoing “arms race” among tech giants and governments, pouring massive investments into the infrastructure needed to power these increasingly intelligent models. Think of data centers packed with GPUs and processors faster than anything we’ve seen before.

The good news?

AI is becoming more affordable. Model costs are dropping rapidly—sometimes by 99% in just a couple of years—thanks to advancements in computing and smarter ways to run these systems. This cost decline means broader access, enabling AI to spread across industries and everyday life.

Want to learn more?

Read this week’s Partner Insights from TCW about The Road to True Artificial Intelligence.

From an investment perspective, the article suggests more near-term potential in the “AI Enablers”—the companies building the infrastructure—than in those simply adopting AI tech. But long-term, as costs keep falling and capabilities grow, the AI ecosystem is predicted to create opportunities across the board.

^{1. Climavision [https://climavision.com/blog/ai-and-weather-data-revolutionizing-accurate-forecasting/#:~:text=In%20Summary:,rapid%20response%20and%20decision%2Dmaking]}

^{2. Climavision [https://climavision.com/resources/ai-weather-forecasting-guide/]}

^{3. Society for Industrial and Applied Mathematics, October 4, 2024 [https://www.siam.org/publications/siam-news/articles/environmental-impacts-of-artificial-intelligence-peril-promise-and-policy/?gad_source=1&gad_campaignid=21862187165&gclid=CjwKCAjw_pDBBhBMEiwAmY02Ns2GrcgrJCM1orYFkK0ibgZSbP8xflSz6rsK9q9qqlNUEMveeqw5nBoCg2IQAvD_BwE]}

^{4. Society for Industrial and Applied Mathematics, June 11, 2024 [https://www.siam.org/publications/siam-news/articles/machine-learning-forecasts-wildfire-risk-on-subseasonal-to-seasonal-timescales/]}

^{5. Society for Industrial and Applied Mathematics, May 27, 2022 [https://www.siam.org/publications/siam-news/articles/convergence-of-artificial-intelligence-high-performance-computing-and-simulations/]}

<sup>6. FourCastNet image [Pathak, J., Subramanian, S., Harrington, P., Raja, S., Chattopadhyay, A., Mardani, M., … Anandkumar, A. (2022). FourCastNet: A global data-driven high-resolution weather model using adaptive Fourier neural operators. Preprint, arXiv:2202.11214.]

Investing involves risks, and investment decisions should be based on your own goals, time horizon, and tolerance for risk. The return and principal value of investments will fluctuate as market conditions change. When sold, investments may be worth more or less than their original cost.</sup>

^{The forecasts or forward-looking statements are based on assumptions, may not materialize, and are subject to revision without notice.}

^{The market indexes discussed are unmanaged, and generally, considered representative of their respective markets. Index performance is not indicative of the past performance of a particular investment. Indexes do not incur management fees, costs, and expenses. Individuals cannot directly invest in unmanaged indexes. Past performance does not guarantee future results.}

^{The Dow Jones Industrial Average is an unmanaged index that is generally considered representative of large-capitalization companies on the U.S. stock market. Nasdaq Composite is an index of the common stocks and similar securities listed on the NASDAQ stock market and is considered a broad indicator of the performance of technology and growth companies. The MSCI EAFE Index was created by Morgan Stanley Capital International (MSCI) and serves as a benchmark of the performance of major international equity markets, as represented by 21 major MSCI indexes from Europe, Australia, and Southeast Asia. The S&P 500 Composite Index is an unmanaged group of securities that are considered to be representative of the stock market in general.}

^{U.S. Treasury Notes are guaranteed by the federal government as to the timely payment of principal and interest. However, if you sell a Treasury Note prior to maturity, it may be worth more or less than the original price paid. Fixed income investments are subject to various risks including changes in interest rates, credit quality, inflation risk, market valuations, prepayments, corporate events, tax ramifications and other factors.}

^{International investments carry additional risks, which include differences in financial reporting standards, currency exchange rates, political risks unique to a specific country, foreign taxes and regulations, and the potential for illiquid markets. These factors may result in greater share price volatility.}

^{Please consult your financial professional for additional information.}

^{This content is developed from sources believed to be providing accurate information. The information in this material is not intended as tax or legal advice. Please consult legal or tax professionals for specific information regarding your individual situation. This material was developed and produced by FMG Suite to provide information on a topic that may be of interest. FMG is not affiliated with the named representative, financial professional, Registered Investment Advisor, Broker-Dealer, nor state- or SEC-registered investment advisory firm. The opinions expressed and material provided are for general information, and they should not be considered a solicitation for the purchase or sale of any security.  Copyright 2025 FMG Suite.}