One of the after action reports I saw after Isaias made landfall was that Google’s AI based weather model performed better than most of the traditional numerical models that scientists use. After a moment of reflection on this news, I had one reaction.

Duh.

Discussions of AI are everywhere and are sure to enflame emotions the more you think about it. So let’s talk about what AI is, or at least what it comes from: advances in computer hardware, which in turn leads to faster computing. From a societal or economic standpoint, I’m not an expert on what that would mean. From a meteorology perspective, though, it’s a boon.

Numerical models have existed as long as meteorology has been around. They weren’t particularly useful until computers reduced the time it took to compute the equations that were plugged into the model. The first models were calculated by hand and took 6 weeks to put together. They were more diagnostic than prognostic.

Models have continued to get better and better as computing has gotten better and better. It makes logical sense that the Google AI model, using the best computing software available, should be even better. There are a few caveats to that, however.

First, meteorology will forever be held back by A) a lack of available observations. There simply aren’t enough observations at the surface and certainly not enough aloft to perfectly encompass the atmosphere and calculate it successfully. Even if we did, B) there aren’t enough known equations to solve for all the variables that go into a forecast. There will always be a bit of mystery.

Caveat number two, is that Google AI wasn’t the only one to accurately forecast a more eastward track than most other outlets indicated. For example, there was me. I’m not trying to brag (yes, I am) I am just trying to note that there will always be room for common sense assessments of the weather. There always needs to be someone to communicate, especially in dangerous situations, or situations that aren’t as dangerous as the unfamiliar might believe.

Having the ability to run many more calculations in a much faster time is the essence of building better weather model. I’m an AI skeptic about a lot of things, however meteorology is a perfect use case for it. The ability to run more equations in a shorter time will allow for better model resolution, which in turn can mean better forecasts and increased safety. The compute power of AI is a positive development.

I’m still not getting in a self driving car, though.

After a lull in the late summer, The Weather Channel came storming back in September. They were the top outlet in 60% of the forecasts for the month, and manhandled the competition. They have a commanding lead in the forecast wins, but when they’ve been bad, they were really bad, so there is a lot to play for this year. If TWC keeps forecasting like they did in September, they will be hard to catch!

Outlet	Forecast Wins (year)
The Weather Channel	8.5
Accuweather	5.66
Victoria-Weather	4.33
WeatherNation	3.66
Clime	3.66
National Weather Service	2.83
Weatherbug	0.33

Well, it’s here. Hurricane Isaias is on the doorstep of the Florida Panhandle, and will make landfall this afternoon to early evening.

Presently, Isaias is a Category 3 hurricane, which is what I anticipated, but it is also taking a bit deeper channel in towards landfall, which at this point appears to be just east of Fort Walton Beach.

While surge is going to be an immediate problem along the coast, within the next few hours, the threat for high wind and a few tornadoes will be pronounced in places like Panama City and Tallahassee. Fortunately, Isaias is moving steadily, and while there is a threat for flash flooding, it is not as severe as we have seen with storms in recent years.

Isaias will march through Alabama tomorrow, which has brought some speculation about the viability of the college football game in Tuscaloosa. It should start drying out fairly early, and football is important at the University of Alabama, so expect it will be a miserable game on.

Isaias is major hurricane and will certainly cause problems in the Panhandle, but it is not as strong as Matthew, the most recent strong hurricane to hit the area, and is not going to dawdle in the high terrain, like Helene. Any landfalling hurricane is notable, and it is particularly notable this year, as it is the first hurricane of any kind in the Atlantic. It only takes one!

Be safe out there.

10:08PM. Check out the SPC outlook on the right panel. That is a VERY late summer look, with scattered showers and storms over the southeast and the Rockies. These are the typical spots for air mass thunderstorms, and the weather right now is…typical.

1:29PM: There are two stories this weekend on the weather front, and they are both part of the same story: There is a whole lot of rain in the middle of the country, and it is very warm in the Southeast. That rain is along a slow moving boundary, behind which lies early autumn.

I just posted about the unusually quiet Atlantic hurricane season. All good there, even a day later. We don’t often talk about the tropical season in the Pacific, because the gyre of the ocean usually keeps storms further to the south, and the general flow of the tropics is east to west, away from the North American mainland. There can be storms that veer poleward, striking the Mexican coast, but the Mexican Pacific is typically quieter than the Mexican Atlantic, particularly the Yucatan Peninsula.

There is one other territory out there, however. Hawai’i sits all by it’s geographic lonesome in the middle of the Pacific Ocean. Typically, hurricanes remain to the south of Hawaii, or never can sustain themselves far enough to the west to afflict the Aloha State. This year, the state has been impacted by two.

Hurricane Lala danced around the the southern coasts of the islands, causing it’s greatest impact to the Big Island, where 4 deaths were attributed to the storm. The greatest issue here was the heavy rain and flooding, though the high winds certainly caused issues of their own, including down power lines in Oahu. On the north side of Hawai’i (the island) one site reported 43″ of rain! While this site was on the north shore, LaLa passed on the south side of the the chain, which is the leeward side of the islands, and does not receive anywhere close to the rainfall seen on the windward side.

Lala was the first hurricane to impact the Big Island since 1871, if that give you an idea of how rare that is. So now that Hurricane Lowell is bearing down on Kauai, this hurricane season is truly remarkable in Hawai’i. Lowell, like Lala, won’t make landfall on the island, but it is forcing the National Weather Service to do something that I don’t know they have ever done: They issued a tornado watch for Kauai and the surrounding waters.

Lowell, as charted from the image pulled from his Wikipedia page, has taken a hard turn to the north towards the islands, and will likely be a hurricane when he makes his pass. Hurricanes often produce tornadoes, but Lala, for example, didn’t have the same rotation and pressure depth Lowell currently does. Given the threat of tornadoes from hurricanes, a watch absolutely makes sense, but Hawai’i’s place in the tropics, and it’s previous ability to avoid tropical storms makes this an incredible rarity.

Also, it’s a lesson for me, even. The Storm Prediction Center only covers the Continental US. Alaska has never had a tornado watch, and until now, neither has Hawai’i, and it was issued by the forecast office in Honolulu.

Lowell will continue to shift north and west. Tornadoes are scary, especially for a part of the world that doesn’t have basements or shelters, however the primary threat, like with Lala, will be rain and flash flooding. Hawai’i contains islands of volcanic origin, so they are filled with rugged terrain and narrow valleys. The leeward side will collect the brunt of the storm, and may sustain significant flood related damage. There will be some surge, but Hawai’i being isolated in the ocean means that water will have other places to go than inland.

Lowell will be off to bother the fish and birds by Tuesday afternoon, and subsequently weaken before hooking back to the east to batter Midway. This is an historic event for the Islands, and a real test for residents there.

I don’t want to jinx anything. I’ve been around long enough to know that there is no such thing as a jinx, but I also have seen enough that I can’t fully be sure. All that said, it is 4 days until the traditional peak of the north Atlantic Hurricane season. So far, we have seen no hurricanes. The strongest storm was Tropical Storm Bertha, and even I don’t remember that one.

The best part? The outlook remains clear, at least for the next 7 days.

I post this with my typical caveat. Even though the hurricane season has been very quiet thus far, it only takes one storm in the wrong spot to make it much more memorable.

We have a timely, and fairly surprising forecaster of the month. The surprise is because out top point earner didn’t actually win any of the forecasts this month, but was consistently near the top, while the others were all over the map. Clime was imprecise, but only a little bit every time, and are our forecasters of the month.

Outlet	Forecast Wins (year)
The Weather Channel	5.5
Accuweather	4.66
Victoria-Weather	4.33
WeatherNation	3.66
National Weather Service	2.83
Clime	2.66
Weatherbug	0.33

This time of year is one of the more active. There is the most heat load in the oceans, so we see the rise of hurricane season, which will peak, on average, around September 10th. Also this time of year, the heat allows more moisture at higher levels of the atmosphere, which allows it to drift inland and participate in the season monsoon across the Rockies and the Southwest.

Because of the scope, ferocity and impact on high population areas, hurricanes tend to get the most attention, not only this time of year, but even when they are out of season. This season is offering a different perspective.

Tropical Depression Five is lingering off the coast of Louisiana, but otherwise, we are going to get to September without an Atlantic hurricane this year. The NHC and Colorado State predicted that we would see a below normal season this winter, but this is particularly quiet, a welcome change over the last several years.

While the Caribbean and Gulf have been quiet, the Monsoon has been working on all cylinders. Monsoonal storms tend to crop up and hang out in place, thanks to a lack of motivational flow. Since they pop up, typically over a slope, and don’t move, there is a primary threat of flash flooding from these storms, as we have seen in the Grand Canyon.

Otherwise, Monsoonal storms fall on dry regions, and are famous for producing the haboobs and dust storms as the collapse under the weight of their own moisture content. Sandstorms are visually arresting, however this year looks to already be memorable for the torrential downpours. Flash flood watches and warnings are issued by the National Weather Service. Warnings in particular can be specific to individual valleys and riverways in their text, following the paths which floodwaters may take.

I’m overdue on these verifications, so here are the top outfits from June and July. You may get the first one right, if you were to guess, since you are here, at Victoria-Weather.com. It was indeed us. In July, an unlikely champion emerged. WeatherNation was the top forecaster for the second month of summer. Congrats!

Outlet	Forecast Wins (year)
The Weather Channel	4.5
Accuweather	3.66
WeatherNation	3.66
Victoria-Weather	3.33
National Weather Service	2.83
Clime	2.66
Weatherbug	0.33