Supertyphoon Mangkhut; massive damage forecast for Philippines, China

While Americans are understandably focused on Hurricane Florence, a massive supertyphoon is causing tremendous damage to the Philippines, and is on track to cause absolutely enormous economic damage in China.  As part of research Enki is conducting to support Global Parametrics (GP, an initiative started by German KfW and the UK Government’s DFID), we are working on refining and expanding global disaster models.  Here are some results using a “next generation” open source model sponsored by GP of the possible impacts of this storm.  First, let’s take a look at the storm itself: it is massive …

Currently the Joint Typhoon Warning Center is showing the storm, which is currently over the northern island of Luzon, tracking across the south China Sea and making landfall just south of Hong Kong:

This forecast is horrific.  $20 to$25 billion in direct impacts to the Philippines, which is 6.5 to 7% of GDP, a massive hit.  But the potential damage in China is eye watering.  On the JTWC official track, damage to Hong Kong alone would be that much, with China overall seeing $120 Billion in impacts.  But that’s not the worst reasonable scenario by far – the respected HWRF model, the track only a trivial 30 miles north of the JTWC forecast,  shows a direct hit on Hong Kong, putting the eye wall of the storm directly across the city, and 6 meters (20 feet) of storm surge in to VIctoria Harbour.  The damage for that scenario is truly phenomenal, over $135 Billion in economic impacts to Hong Kong alone, $230 to China overall … wow.  The humanitarian, economic, and political implications of such a disaster are hard to imagine.

Florence Afternoon Update

We’re now getting significant impacts on the outer banks, the core of Florence is fully within the range of radar and rain bands are coming ashore with gusty winds.  Remember that I do research in to natural (and anthropogenic) hazards, and data is key.  We have an amazing amount of data now to study landfalling hurricanes in real time.  Here’s a cool overlay using the AWIPS II workstation, showing radar, high resolution satellite, and wind data …

Conditions on the coast are rapidly deteriorating.  Unfortunately, given the slow motion of barely 5 mph if that, it will take its time coming ashore, so it will have plenty of time to batter North Carolina with waves, storm surge, wind, and rain.  These conditions are likely to continue for the next 48 hours or so, 3 or 4 times the usual duration of adverse conditions in a hurricane.

A big concern with any landfalling storm that slows down and stops is rain. Hard to forget Harvey!  Here is the current precipitation total map from HPC.  Totals in that white blob are amazing – upwards of 30 inches! That will potentially cause some epic inland flooding …

Here is the latest NHC forecast track and damage swath.  Not that different from previous forecasts …

Damage is likely to total over $ 10 Billion dollars for the “hurricane” part of the storm.  The inland flooding is another story.  It could easily add another $5 to $10 Billion, depending on what floods and how bad.

There are still wobbles and people are still wildly speculating on the future of this storm, especially for points south of the northern SC coast.  As usual, I urge folks to keep cool and follow the official NHC forecasts.  The storm may skirt the coast a bit before final landfall, and the next 24 hours may be nerve wracking, but by far the most likely scenario is the storm will enter over far southern NC, cross northern SC, and break up this weekend …

Florence Update for Thursday, 13 September 2018

Florence continues to “barrel” (funny how some words get stuck in the media) towards the southern North Carolina shoreline.  The short version is that for the immediate landfall area, not too much has changed.  Watches and warnings are the same, expected “worst case” impacts haven’t changed, etc.  There is some good news in the sense that the storm may not be as intense as forecast when it makes landfall, it’s down to Saffir Simpson category 2 levels.  That DOES NOT mean those in the warning areas should fail to prepare for a Cat 3 – there could be a bump in intensity before landfall, but signs are for a weaker storm than expected.  Still, will be bad enough, so if you haven’t gotten out, do it this morning before conditions deteriorate rapidly later today.  Winds along the coast are light so far (looks like 10-15 knots peak right now), but that is going to change rapidly.  Here’s a composite radar view from the two closest sites (Wilmington, range indicated in white, and Morehead City, range indicated in green).  Classic eye and banding structures becoming visible …

The various forecast track models and other simulations are in pretty good agreement that the storm will make landfall, then stall out for at least two days near the SC/NC border, over the Grand Strand. Here is the impact estimate based on the official NHC forecast as of 5am this morning:

This is the tricky part of the forecast – if it’s on the shore, or near the short, it can maintain intensity, and pump more rain inland.  If it’s even 50 miles further inland, the intensity and rain rates will drop off pretty quickly.  On the current track, the worst wind damage will stay in North Carolina.  Direct impacts should be around $10 Billion.  The big question is rain and flooding.  The current estimates are for about $5 Billion in additional impacts from rain related flooding.  But a longer pause could double that number; likewise, if the storm is even 50, much less 100 miles inland, that could easily drop to a Billion or so.  It’s a great unknown …

For folks in Savannah GA and the Beaufort/Hilton Head area, things look pretty good on this track.  We’re on the weak, dry side of the storm.  Winds will be offshore, so storm surge flooding will be nothing to worry about.  Right on the coast, waves and wave setup might cause some issues, and rip currents will remain dangerous.  Depending on wobbles, we may get rain, or just as easily get a clear if breezy day in the subsidence zone.  But dangerous conditions (other than on the beach) just don’t seem to be in the cards right now.

Florence 11am ET Update: No changes to official track, models shifting again …

For more detail on model tracks and why you shouldn’t focus on them, see my post from earlier today.  But if you need an example, let’s look at the changes over the last 6 hours in just one global model, the NCEP GFS model, in comparison to the official forecast track (which didn’t change):

The official track didn’t change very much.  Still looking at the same folks who should be evacuating by now, same watch and warning areas, same damage estimate (around $15 Billion).  Here’s the map:

So for people in southern South Carolina, and especially those in Georgia, I again say don’t eat poorly cooked pasta.  It will give you indigestion …

Hurricane Florence: a deeper look at track models (Wed. Morning, 12 Sept)

This is another long read, with some SCIENCE!, but the details of this matter – I hope you will be patient and read on.  While I often touch on hurricane track forecast models, I normally stick with the official NHC forecast track when discussing the impacts of storms.  There are a lot of good reasons for doing this.  First, track forecasting is a very specialized profession, with a lot of complex factors involved.  NHC has specialists to do that, and while I sometimes disagree with details, they are very good at it. Interpreting the multitude of track models is not an easy task.   Second, given it is their responsibility to do official watches and warning, diluting their message to the general public is not a good idea.  Some may say this is a bit hypocritical on my part, given my criticisms of Emergency Managers and evacuations, but there is a key difference – NHC are professionals who consistently do a good job.  Governors issuing evacuations orders have a much spottier track record, in my clearly not so humble opinion 🙂 , so I feel free to trash them.

Unfortunately, there is a trend in the media to talk about specific track models at the expense of the official forecast, and to publicly second guess or anticipate the next NHC forecast.  Part of this is because beautiful, dynamic animations are available for individual track models.  These graphics are great for TV, and the temptation to show a graphic that passes over or near your viewing areas is overwhelming.  But it can also be very misleading.  So let’s take a closer look at track model outputs for Florence, and what they mean.  Here’s the track model map available at about 4am this morning:

The Global models (exemplified by the dark blue GFS line show the storm pulling up short of the NC shoreline before diving south.  The dedicated hurricane models such as HMON (the replacement for the GFDL, shown in light green), the HWRF (shown in light blue) and the official track (red lines) don’t have the sharp right turn.  There are a lot of reasons for this.  For the objective models, it is important to realize that while they are often good at forecasting hurricanes, and are used to provide “boundary” conditions for specialized models, they are not specialized, high resolution track forecast models. They often lack the detailed coupling to ocean models, and have less sophisticated model physics than the dedicated models.  This gap has closed in recent years as global models have gotten better, but the gap remains.  In their forecast package development, NHC takes in to account factors like this using their involvement in, and detailed knowledge of, these models.  They also realize that models can shift radically from one iteration to the next.  Every 6 hours, data from weather stations and satellites is collected globally and used to create a picture of what the atmosphere looks like at that time.  These are the initial conditions.  Those initial conditions are then used to spin up the global models, as well as the dedicated tropical cyclone models.  That’s the simple version, but the details are very complex.  Not all initial conditions are created equal.  One small example: the 00Z (8pm ET) and 12Z (8am ET) initial conditions are often “better” than the 06Z (2am) and 18z (2pm) data sets.  This is because most upper air observation sited only launch two balloons per day, at 00 and 12z.  Sometimes you get a satellite pass over a storm at the right time, sometimes its 3 or 4 hours earlier.  Sometimes you have airplane data, sometimes not.  So when I fuss at meteorologists, weathercasters, and storm watchers for latching on to a model shift, that’s one reason why.  It’s also why the NHC sometimes seems slow to shift tracks to follow what seems to be shifting model data – they realize that it can just as easily shift back 6 or 12 hours later.  There is much more to it than this, but I hope that gives you at least a glimpse into the fact this isn’t simple, or for amateurs.

Another factor is that the storm isn’t just some line on the map: it’s a broad, complex system.  Let’s look at the impact from two very different model simulations.  Here’s the HMON run, and for comparison, the CTCX run, which shows the storm hitting Savannah, GA …

Note especially the impacts onshore.  If you compare the areas of severe impacts, despite the fact that the track line is scary for people in Savannah and southern SC, the dangerous, life threatening impacts are all contained within the warning areas NHC has set.  Now here is the impact estimate based on the official track:

What about economic impacts?  On the official track we are still looking at about $12 Billion in direct impacts.  The HMON track would product “only” $5 or $6 Billion, while the CTCX track would likely cause $3 or $4Billion, because the stronger winds would stay offshore!

And what about the infamous “cone of uncertainty?”  I see people making much of this or that city being “touched by the cone.”  Well, to be honest, who cares.  I hate that graphic.  What you should care about is how strong the storm will be if and when it gets to you. As NHC notes, bad impacts can exist outside the cone.  But the opposite is also true.   If the death cone touches you on day 5, when the storm is inland and barely even a tropical depression, and you probably wouldn’t even know it was a storm if somebody didn’t tell you, so what?

So despite all the rending of garments and gnashing of teeth, because of the intensity of the storms on the scary scenarios showing a turn along the coast, the storm would weaken fairly radically, and the impacts would be less.  So the conservative path for NHC is to continue to slowly shift the track south (assuming the forecasts continue that trend), adjust the intensity accordingly.  And, as you can see, the present watch and warning areas are in fact perfectly fine for this storm.

Bottom line: yes, the forecast tracks are shifting. Yes, forecasting is not an exact science. But it’s not really as big a deal as some would like to make it out to be.  Stay cool …

Florence Evening Update for Tuesday 11 September 2018

First some advice.  Don’t freak out over any one computer model forecast, and be skeptical about folks who show you dramatic graphics and talk about tracks while not discussing the official National Hurricane Center forecast all that much.  Tropical Cyclone Forecasting is specialized, and understanding all of the nuances of track models takes a lot of experience.  Most of what you read here is either an interpretation of or derived from NHC forecasts using my active research work on natural hazard damage models.  While I work on forecast models of various kinds, and have a deep understanding of numerical weather models, for the most part I leave that end of the business to the folks who do it every day.  Why are my discussions often less dramatic than other sources?  Probably because my income doesn’t depend on hits, viewers, or selling you stuff 🙂  But let’s start with the obligatory dramatic satellite picture anyway …

As usual, lets will look at the track, the likely impacts on that track, followed by the implications of those impacts in both economic and human terms.  This is a bit of a tough forecast situation.  There is a high pressure ridge to the north of the storm that is currently guiding it towards North Carolina.  That should control the motion for the next two days (until Thursday).  But then things get a bit less clear.  This morning, most indications were that a “weakness” in that ridge would cause the winds steering the storm to slow down or even stop as it makes landfall over North Carolina, drifting for a couple of days, then begin to move northestward, dumping a bunch of rain in the process.  Today, however, the dramatic ECMWF model runs making the rounds in social media show the storm “bouncing” off the high and moving SW,  showing Florence heading south and perhaps even hitting again as far south as Georgia.  Most models, however, just show some kind of jog  offshore rather than just inshore as was the case this morning. So the big picture is more or less the same Here’s some of the track models as of 4pm this afternoon (ECMWF isn’t shown for licensing reasons) …

So what does all that mean?  Well, let’s see what the experts have to say.  Here is the 5pm NHC forecast track.  I added in the 5am track from this morning in yellow for comparison.  Not so different, huh?  From their forecast discussion, “there remains no significant change to the previous forecast track or reasoning.” 

So what’s the bottom line? No major developments. The areas under watches/warnings/evacuations are not changed. The damage estimates are about the same – somewhere around $14 Billion in direct damages, flood damages highly dependent on exactly where Florence stalls out and how much rain gets dumped.  Folks in the immediate path of this thing should get out of the way.  IF the storm does do the stop and turn shown above, that’s good: it will be weaker, and cause less damage.  If it follows the ECMWF scenario, it will almost certainly not be in very good shape as storms don’t like to run in to walls of air and get shove at a 90 degree angle … so there is still no need to panic, just keep watching the official forecasts, smile (and don’t make any sudden moves) at folks who are freaking out and latching on to every weird variation in this complex situation, overreacting to pretty pictures, and stay cool …

Florence: NC most likely landfall as Category 4 on Thursday/Friday

Today is a major day for decision and action as Florence is clearly headed towards North Carolina.  Here is the current NHC forecast track (read with hurricane symbols), with the estimated damage zones based on my Haetta/TC model.  Also shown, in yellow, is the official forecast track from yesterday morning.  For all of the talk of uncertainty and track models among the chattering classes, the official forecast has been remarkably consistent:

Notice the area of catastrophic, severe, and widespread damage. Seems small, given the talk, doesn’t it?  The thing to keep in mind with hurricanes is that the swath of actual severe damage most times, even for a major hurricane, isn’t all that wide.  Media folks love to talk about how big the cloud shield is, or the extent of the over water tropical storm force winds, but for Florence, severe effects will extend about 120 miles to the right (north) of landfall, and 80 miles to the south.  Of course the problem is we don’t know exactly where that 200 mile wide swath of damage will be in 3-4 days when it hits, thus the 600 mile wide watch and warning zones.  Don’t be mislead by fluctuations in intensity, or brief north or south wobbles.  For example, this morning the storm is going through an eye replacement cycle.  It may even be a cat 5 at times.  Along the coast the damage will be epic if the storm stays as strong as it is. 25, maybe even 30 foot storm surges in places where the water can pile up.  Even inland, damage will be severe, with trees down and structures damaged.  But especially scary is a detail in that map: notice that landfall is expected in 3 days – then the storm moves only 50 or 60 miles over the next two days.  While the intensity will drop quickly after landfall, enough of the circulation is likely to remain over water to pump immense amounts of moisture over NC and VA.  That means the potential for tremendous inland flooding.  If you live near a creek, stream, or other flood feature, be aware that the waters may rise quickly.  Local knowledge is everything in this kind of situation, but with so much water, and so much development (which changes radically the hydrology of an area) areas that have never flooded will probably flood.  Comparisons are being made with Harvey.  That may not be far-fetched.

My damage models are still showing approximately $20 Billion in damage, with as much as $15 Billion in additional flood damage possible depending on variables like how long the storm sits and now much rain is dumped into Virginia.  Florence is definitely on track to enter the top ten most damaging storms.

At this point it is fairly clear who needs to take clear, decisive action.  NHC has issued a hurricane watch for the coastline from Edisto Beach SC to the NC/VA border.  I expect a watch or warning of some kind for parts of Virginia today are possible depending on minor track shifts and timing.  Especially for those in Myrtle Beach and the northeast corner of SC, and all of eastern North Carolina, the hyperbole isn’t really that wrong here.  If on the coast, or in low lying areas, get out.  If within 50 miles of the coast in anything other than a substantial structure that you know can stand up to high winds, get out.  If you are in the NHC watch areas you really need to take action and be gone before Thursday morning.

That said, I again want to point out that for far southern South Carolina, south of Edisto Island, the risks are decreasing rapidly other than for rip currents and waves right on the beach in the surf zone.  For Georgia, the risk at this point is vanishingly small.  It will probably be a nice day as we (I live in Savannah GA) will be in the subsidence zone at the fringes of the storm, but don’t plan on going to the beach and swimming, as again waves and rip currents will be dangerous.

Florence update, Monday evening, 10 Sept 2018

No doubt about it, Florence is truly a monster category 4 storm now, here the eye seen at 5pm  …

So the questions are, as usual, where is it going, how bad when it gets there, and what to do?  Where is more or less clear.  The track guidance is fairly well clustered on the lower NC coast, and has been shifting a bit northward during the day.  Note the blue track of the NOAA GFS model doing a loop offshore.  That’s not a glitch – that could really happen, although it’s not likely.  Here’s the latest map using tracks available as of 4pm or so, with the latest NHC forecast track in bright red (the previous track is in darker red) …

How bad?  Pretty bad.  Here’s the forecast damage swath, using the official NHC forecast track.  The track didn’t shift that much during the day today, a little bit north, but the broad swath of impacts still covers much of the northeastern corner of South Carolina and all of Eastern North Carolina.  Storm surges could easily be in the 25 foot range in places.  Damage along the coast will be catastrophic if the intensity holds, with severe damage as far inland as the Research Triangle area.  

On this track it’s a $20 Billion dollar storm at least, with many estimates coming in at $25 to $30 Billion.

As for what to do, this is a seriously bad storm.  If you are in on the coast of South Carolina in or north of the Charleston area, or North Carolina, you need to evacuate.  It’s a no brainer.  Get out, and if inland in a secure structure, button up.  If in mobile homes get out even if inland.  You really need to listen to local emergency management on this one, hyperbole aside, it is a very dangerous hurricane.


For the Georgia and far southern SC coast (Beaufort/Hilton Head), the bad impacts will likely all be to the north, and the risk decreases with every hour.  This raises an interesting (and tough political) question, given the state boundary between SC and Georgia.  The Governor of SC has ordered mandatory evacuations for all zones for the entire coastline.  So we have the odd situation where people on Tybee are under no instructions to do anything, while folks across the river who they can literally see are being told to flee for their lives.  One of my complaints about the way evacuation decisions are made is that they are often set up as canned responses with little room for flexibility.  There are a variety of legal and operational reasons for this beyond the scope of this post.  While there is some coordination between the states, most of these decisions are stovepiped, and given the different political considerations in each state, along with different skill levels and experience of personnel, things like this happen.  To folks in Georgia, don’t freak out.  While you should carefully listen out for a possible change of direction, again the risk is decreasing.  For people in the Hilton Head area, you have a choice to make – follow the directions of your emergency managers, or take a chance, given the odds are likely very much in your favor since it would take a pretty radical change in track to bring life threatening condition to your area?  In general I support following the advice of local EMA’s.  Here, I personally think a more targeted approach would have been appropriate.  I can absolutely see getting ready to evacuate special needs populations, etc. I can see putting the communities on notice.   But a general evacuation carries risks – evacuating Sun City and HHI will very possibly kill some people due to stress, traffic accidents, etc.  It risks clogging up the roads and  taking up hotel and shelter space for people who absolutely must get out of the way.  Unfortunately, this is a discussion that tends not to happen until a storm is “barreling” into the coastline, and it devolves in to a “follow evacuation orders or die” kind of thing.  We need to have a nuanced discussion about this, and some better planning and coordination at the *national* level.  That said, the bottom line is that EMA’s are often a bit conservative, and sometimes have operational, political, and legal considerations that aren’t always in the best interests of every individual, but for the most part they are trying their best and deserve your respect and attention.  In this case in the far southern counties of SC (Beaufort, Jasper) those other factors might be in play.

Atlantic update (Florence, Isaac, Helene)

Of course most of the attention is on Florence, but Isaac is out there headed towards the Caribbean.  Fortunately it’s not well organized, and expected to fade out over the mid Caribbean sea.  Helene is way out there and likely to head out into the ocean.

Florence, on the other hand, is rapidly intensifying as expected.  Here’s the view at 11am, classic hurricane structure …

The forecast is still pretty grim for the Wilmington NC area, and didn’t change much from this morning.  Evacuations have been ordered for the Outer Banks as of noon, and expect further evacuation orders to be announced in North and South Carolina this afternoon.  NHC plans on issuing watches tomorrow (a watch is issued 48 hours before tropical storm or hurricane conditions are expected).  Here’s the 11am NHC track and Haetta/TC estimated impacts …

On this track and intensity, the model gives economic impacts of $25 to $30 Billion dollars, which would place Florence in the top ten for damage.  You’re going to hear a lot of hyperbole about this storm, but that doesn’t mean it isn’t a very dangerous system.  If you are in the potential impact area, listen to the guidance of your local emergency managers and take appropriate action.

Florence Update for Monday Morning, 10 September 2018

Short version: The SC coast north of Charleston, all of North Carolina, and Virginia/DelMarVa  Peninsula really need to take this storm seriously and be ready to act today. Georgia should keep an eye on it, but things are looking ok at this point. Watches and warnings, and the attendant evacuations, etc. are just about inevitable for SC/NC and maybe VA at this point. A reminder, be very careful when using social media like Facebook that uses algorithms to display information, since it does not always work chronologically, and you may be seeing old data!

The steering environment is still somewhat fuzzy, but as expected has come in to much better focus over the last 24 hours.  Starting with the track guidance, it is fairly stable pointing to a landfall along the southern North Carolina coast (Wilmington area).  Here’s the current track forecast maps:

Florence is intensifying and now looks like a serious hurricane as it moves over warmer waters, and more favorable wind environment.  It may well be a strong category 3 or even category 4 before landfall.  Here’s the satellite view as the sun comes up (just before 7 am ET):

So, what might happen?  It’s not good.  Here’s the impact graphic using the official NHC track and intensity as estimated by my Haetta/TC model.

Damage right at the eyewall at landfall and inland could be epic if this intensity forecast holds up.  Storm surges of 5 to 6 meters (upwards of 20 feet) are likely to the right of the landfall, and depending on the exact storm configurations, 25 feet are not out of the question.  Using the 5am NHC forecast track/intensity, Florence is shaping up to be about a $15 to $20 Billion storm. A BIG question on this track is how intense the storm remains over the Research Triangle area of NC (Raleigh-Durham-Chapel Hill). The NHC forecast has it decaying to a tropical storm before crossing the triangle, but storms like Hugo stayed as hurricanes to that point, and if so the damage racks up pretty quickly. The tops of pine trees shear off on average at 70mph or so – and NC has lots of pine trees to end up in the living rooms and business in the triangle. Subtle changes in how fast the storm decays, a 20 or 30 mile wobble as it moves inland, or a tornado or two in a bad place and this *easily* turns into a $30 Billion storm. Wobbles or shifts towards Virginia and the “target rich” environment of the Norfolk and DelMarVa penensula (and the Washington DC area, even as a decaying tropical storm) could also up the totals.

Bottom line: if you are in the swath of this storm, take is seriously.  Don’t panic, but pay attention to National Hurricane Center advisories and your local emergency management, and take appropriate action.  Along the rest of the coast, beware high waves and rip currents.  Surf might be good for another day or so, but be aware of your limitations and don’t get in to trouble!