Rapid intensification – Vongfong update

This is the down side of trying to blog an extreme weather event in real time while holding down a full time job as an academic. Just can’t keep up.


When I wrote my previous post on Typhoon Vongfong (shown above in a recent visible satellite closeup, from this great site) last night, Vongfong’s winds were estimated at 90 knots, and the Joint Typhoon Warning Center (JTWC) had forecast that it would intensify to 130 knots over the next 48 hours. When I woke up this morning it was estimated at 120 knots already; it’s now at a screaming 155 knots, or 178 mph (per JTWC).

The NHC defines “rapid intensification” as an increase of 30 kt in 24 hours, so Vongfong has definitely undergone rapid intensification. It was not forecast, but it rarely is. While tropical cyclone track forecasts have improved steadily for decades, intensity forecasts haven’t, and the difficulty of forecasting jumps like Vongfong just made are a big part of the reason.

It’s not just the forecast of the storm’s future that’s the issue, though. Even estimating its intensity at the present moment poses challenges. While JTWC is calling it 155 knots – the very upper end of category 4, pushing category 5, the top of the scale – the most recent estimate from the Japan Meteorological Agency (JMA) puts it just at 115 knots, low category 3. Some of this difference is attributable to the different standards; JTWC uses 1-minute averaged sustained winds while JMA uses a 10-minute average, which tends to be a little lower. But that isn’t the whole story.

In fact, no one is measuring the winds in Vongfong. There is no routine aircraft reconnaissance in the Western Pacific to make direct observations. The estimates all come from analysis of satellite observations, and satellites don’t measure the winds directly. The satellite images used to estimate intensity see some combination of clouds and water vapor, and either human analysts or automated pattern recognition algorithms use those to infer the wind speed. (To get some idea how it’s done, try this.)

There’s no doubt that Vongfong is a very powerful storm. The forecast models are mostly predicting that it will start to weaken by a day or so from now, perhaps after holding steady or even intensifying a bit more before then (see for example a collection of current forecasts on Kerry Emanuel’s page). The track forecasts are still taking it over Japan early next week. Assuming that pans out, as seems likely, the question is just how rapidly Vongfong loses its currently formidable head of steam.


Vongfong – like mother like daughter?

Fortunately, Phanfone turned out not to be a major disaster for Japan. There have been some power outages, some disruption to air travel and other things, and a handful of people killed (mainly due to being in or near the waves at at a bad time, whether the Americans in Okinawa or the surfer in Kanagawa). Nothing to make light of, but not nearly as bad as it could have been.

But it’s not yet time for Japan to relax. The next Typhoon, Vongfong, is spinning its wheels in the tropical western Pacific, and is likely to take almost the same path as Phanfone did, down to the landfall on Honshu, Japan’s big island.

Vongfong is currently estimated to be a category 2 storm, with peak winds of 90 knots, or over 100 mph. It has developed the appearance to go with that intensity, including a proper eye, seen here in the latest visible satellite image:



Track and intensity forecasts for this storm, as for all tropical cyclones in the western Pacific, are issued by the Japan Meteorological Agency (JMA) and the US Joint Typhoon Warning Center. Like the US National Hurricane Center (which produces all our forecasts here in the US for the Atlantic and Eastern Pacific) and other forecast centers worldwide, both agencies issue forecasts out to 5 days. The current track forecasts look quite similar to what Phanfone’s did at this stage. This includes especially the northward turn, scheduled for around 48 hours from now, shown here in the current forecast map from JMA:


While these forecasts have no landfall in them yet, the numerical models seem to be reaching some agreement that Japan is likely to get another one in about a week. Here is a forecast map from NOAA’s GFS model for 00 UTC next Tuesday, October 14. (The recent runs of the ECMWF model look pretty similar.) The map shows surface pressure (contours) together with 500 hPa geopotential height (color shading). Vongfong is the bullseye riding up the east coast of Honshu, interacting with an upper level trough (the southward dip in the ribbons of color) to undergo extratropical transition into a midlatitude cyclone:



Even the intensity forecast (from JTWC) is eerily similar to Phanfone’s. At 90 kt now, Vongfong is forecast to make it up to a 130 kt super typhoon, in the low end of category 4, over the next 48 hours, before starting to weaken again as it moves over cooler waters and higher wind shear. Just how much it weakens is the key question – the phasing with the upper trough in the model run above is just of the right kind to add an extra jolt of energy from the jet stream at landfall. (That’s what happened to Sandy just about two years ago, and in fact the map above would look quite similar to the equivalent one at Sandy’s landfall, if one were to swap North America for Asia.)

Why would two storms develop so similarly, one right after the other?

First, the Western North Pacific has more tropical cyclones per year than any other region on earth. A typical year sees around 90 in the whole earth; the WNP typically accounts for 25 or 30 of those. For our North Atlantic 10 would be about average.

But there’s more to it than just a lot of storms. They do form in bunches sometimes, and it may not be an accident. One typhoon can help generate the next one, when its own spin interacts with the earth’s much larger-scale rotation to produce a “Rossby wave” wake in the atmosphere to its east or southeast – a high pressure, clear weather ridge, followed by another low-pressure trough, typically about 2000-3000 km away, which then is a promising site for another typhoon to form. I haven’t analyzed the early biographies of Phanfone and Vongfong yet to see if they might have been such a “mother-daughter” pair at Vongfong’s genesis, but if they did it wouldn’t be very unusual.

As for the tracks being the same after genesis, that may be just a consequence of the large-scale steering winds not having changed much during the interval between the two storms. It’s fall, the Pacific jet stream should be starting to make its way south, but isn’t far enough south yet for its westerlies to blow the storms back out to sea before they get as far north as Japan.

While those scientific explanations are backed by some research – the links above are to my own research papers, with my former Ph.D. student Kyle Krouse – I don’t think we understand everything we could about the clustering of tropical cyclones in time and space.

Perhaps the most extreme case in (relatively) recent memory was the Atlantic season of 2005, when the hurricanes just kept coming, one after another, long after they had any right to. That was mind-boggling at the time. This current pair in the Pacific may be less so, but it still feels a little spooky.

Typhoon Phanfone – a post that will be immediately out of date

Typhoon Phanfone has been closing in on Japan for days, and is in the process of making landfall as I write.

The storm earlier had made it to the bottom end of category four on the Saffir-Simpson scale, with peak sustained winds of 130 knots and a very small “pinhole” eye. The pinhole eye is almost always a sign of a scary storm, because the way tropical cyclones develop really strong winds is by contracting their eyewalls inward. As with a skater who pulls her arms in while she does a spin on the ice, conservation of angular momentum increases the rate of rotation because the distance from the axis has shrunk.

By now the storm is considerably weaker, with peak sustained winds estimated at 80 knots, category 1. According to the cyclone phase space diagnostic of Prof. Robert Hart at Florida State University, Phanfone is in the process of extratropical transition, turning into an extratropical or “winter” type storm. It has lost its eye and most of its circular symmetry, as one can see in the current infrared satellite image, and is forecast soon to lose its warm core. In a current weather map with colors showing temperature at 850 hPa (about 1.5 km above the surface; this is actually a plot from the most recent 12-hour forecast made by NOAA’s GFS model about 12 hours ago, image courtesy weather.unisys.com), one can see the strong contrasts associated with fronts, typical of extratropical storms, cutting through the center. The strongest rains will be on the left side of the track – over land – as the moist tropical air rides counterclockwise up and over the colder air mass to the west.



Despite its weakened status, Phanfone is still no joke. The rains will likely be in the hundreds of mm (100 mm = about 4 inches), including likely on Mount Ontake, the volcano where over 50 people died in the recent eruption and the recovery of bodies is still ongoing.

The storm surge forecasts appear to be in the 1-2 meter range, or 3-6 feet. The warnings appear to be concentrated in Fukushima prefecture, site of the 2011 tsunami-induced nuclear disaster. While storm surges and tsunamis have completely different causes, they are otherwise very similar phenomena – either way something pushes the sea onto the land. The 1 or 2 meters this typhoon is forecast to produce are nothing compared to the 40 meters that the earthquake of three and a half years ago produced, so that’s good. But still… I have no detailed knowledge of the current state of the Fukushima Daiichi nuclear complex, but my understanding is that it is still not in particularly great shape. I hope that 1-2 meters of surge plus combined heavy rains and wind aren’t enough to do any more serious harm to this already awfully blighted place.

In news coming in now: images of flooding, and speculation that Tokyo, almost directly in the path of the center, might see an all-time record for strongest wind.