From an evacuation standpoint, note that there are only two ways off Galveston island at this end: the causeway bridge (which is only open until the winds get too high, yellow stripe on this map) and the ferryboat across to the bolivar penninsula, which wont' do much good since a) it's slow, b) they have to shut it down when wind/wave action gets high, and c) it puts you on a less sheltered barrier island.  If you click over to satellite you'll see that these islands are really just tall sandbars.  Pelican Island (the largish island between Galveston and the Ship Channel) was basically created by dredging the channel.

The refinery workers are all further inland (Texas City, La Marque, etc) but FEMA says a Cat III can cause flooding there, so if the storm becomes a Cat III they'll be told to evacuate.

So a landfall of the eye just south of Galveston could be the worst-case scenario for Galveston, Galveston Bay, and the refineries just north of Galveston Bay.

http://www.boatus.com/hurricanes/hurricane_field2.asp

And for those of you interested in Sea Surface Temperatures (SST) there is a better model for tropical cyclone (hurricane) intensity based upon Tropical Cyclone Heat Potential (TCHP).  As most know, a hurricane is powered by transferring heat from the ocean surface to the upper atmosphere.

From the article http://www.aoml.noaa.gov/phod/cyclone/data/pubs/EOS_FINAL.pdf

The thickness of the upper ocean layer from
the sea surface to the depth of the 26°C
isotherm is only a few tens of meters in most of
the tropical regions.A passing tropical cyclone
draws energy from these warm waters and
mixes them with the cooler waters below.
This creates upwelling; the depth of the base
of the mixed layer is raised,and the temperature
of the surface waters is subsequently lowered.
These cooler waters now provide less energy
to the tropical cyclones, most likely slowing
the rate of intensification.On the other hand,
the depth of the 26°C isotherm in the core of
warm currents (such as the Loop Current in
the Gulf of Mexico) and warm anticyclonic
rings may reach more than a hundred meters.
This type of condition has values of thermal
energy that are usually several times larger
than those associated with the genesis and
sustainability of a tropical cyclone, and it is
found in most regions where tropical cyclones
occur.
Understanding the role of these warm
features in the intensification of hurricanes
in the tropical North Atlantic is an ongoing
research topic that is still at an early stage.
Preliminary results have shown their importance
in the sudden intensification of hurricanes in the Gulf of Mexico [Shay et al., 2000].
For example, in September 1995,Hurricane
Opal suddenly intensified when passing over
a warm ring that had gone undetected by the
sea surface temperature derived from the
Advanced Very High Resolution Radiometer
(AVHRR) imagery. In August 1999,Hurricane
Bret also traveled over two warm features in
the Gulf of Mexico, and it intensified each
time. Since then, the monitoring of the upper
ocean thermal structure has become a key
element in understanding and predicting
sudden tropical cyclone intensification.

TCHP is measured in heat energy per sea water surface area (kilo Joules per square centimeter).  A passing hurricane will subtract heat measuring 60-100 kJ/cm2, so THCP values above 80 kJ/cm2 are especially dangerous.  You will remember that Katrina rapidly intensified from a Category 2 to a Category 5 hurricane, and that occured just as it passed over an area of high TCHP (80+ kJ/cm2) in the Gulf of Mexico.

Here is the link to the dynamically updated graphic of TCHP for the Gulf of Mexico: http://www.aoml.noaa.gov/phod/cyclone/data/go.html

Click on the thumbnail for a larger image.  There is also a historical link on this page for the TCHP data and track for Katrina.

-geological risks: how big are the reserves, will we find more, are the reservoirs being overproduced? ...

--geopolitical risks: a higher share of world output coming from less stable places, pressure on big consuming nations ...

--socio-economic risks: issues of public policy, demand elasticity, human perception and behaviour, environmental damage ...

--Now, weather risk. We're implicitly saying that natural events, which may be growing more severe but certainly aren't new, are a risk that can throw the whole world system off track.

Put it all together: we're on a knife-edge of supply versus demand, with no safety margin. All types of risk are increasing, at a time when we can't afford any real risk. Sounds like peak oil is here. Keep watching the signals of change.

PEAK OIL  2004-2007  

Updated periodically; I would hope more frequently as time goes on.

"Rita is developing into our worst-case scenario," said John Kilduff, vice president of risk management at Fimat USA in New York. "This is headed right into our other major refining center just after all the damage done to facilities in Louisiana. From an energy perspective it doesn't get any worse than this."

I live in the Austin, TX area.  I've been to Houston several times, and this is my general impression of what worst-case Cat 5, northeastern quadrant strike on metro Houston would do, in a social sense.  The worst situation is simply in terms of populations.  Greater Houston is inhabited by about 5.4 million people, where New Orleans housed roughly 1.2 million.  This is an area with roughly four times the population.  The kind of disasters that could happen in the same effect as the Superdome are numerous, and could be on a grander scale.  This is a decent prediction, as Houston is probably the most stratified city you'll ever go to.  The suburb of Sugarland houses the vast majority of oil engineers, executives, professionals, etc, while Houston is mainly inhabited by - how is this for dejavu - poor, black families, many of which don't have access out of the city.

Of course, it is a slightly different situation.  I'd imagine that the same sort of distaster response mechanisms that failed with 'rina won't fail with Rita, as that would be a completely unacceptable response.  But, this will more than likely be a stronger storm, hitting a higher population center.  No doubt there will be heavy casualities, and devestating destruction (refer back to the previous comment predicting $50 billion in damage.)  It's my opinion that this number may even be low.  Houston is one of the best examples, outside of LA, of urban sprawl.  The area is frickin' huge, people.  Relativly low population densities create buildings most vulnerable to destruction by high winds, where things such as apartment buildings, high rises, etc generally survive.  Couple this to the regional economy, which is already straining under the weight of Katrina (I can attest to this, as can anyone in the region, that the evacuees have generally caused alot of shortages) and there's a distinct possibility of regional economic collapse, which would effect people across the entire southeast.  And this is a scenario that completely disregards possible damage to one of the largest concentrations of refineries in the country.  If we thought possible oil shortages following Katrina were bad, what about reduced production in the already strained gas refinery industry.  This isn't a market which marks delivery 6 weeks out.  It is immediate.  

I don't mean to sound alarmist, but that's just my honest assessment of what a Houston strike would do.  With any luck, the storm will head much further south, to the South Padre region.  

On a personal note, I have a question.  I already have heard people are starting to board up their homes in the Coastal region.  What I haven't heard though, is whether this should be a concern for the Austin area.  We are a good 200 miles north, but as I was looking at maps, it seems like we aren't much further north from the coast as some of the most devestated areas in Alabama were.  Since we haven't really heard anything from local officials, should evacuations, or at least, boarding up my house be something to consider?

http://www.channelcincinnati.com/news/5001862/detail.html

THERE HAS BEEN NO CHANGE IN THE STEERING PATTERN AND RITA IS MOVING WESTWARD OR 275 DEGREES AT 11 KNOTS. THE HIGH PRESSURE SYSTEM THAT HAS BEEN FORCING RITA WESTWARD IS FORECAST TO WEAKEN AND SHIFT EASTWARD. THIS WILL ALLOW THE HURRICANE TO TURN GRADUALLY TOWARD THE WEST-NORTHWEST AND NORTHWEST DURING THE NEXT DAY OR TWO. THE CORE OF RITA IS BASICALLY MOVING TOWARD THE TEXAS COAST AND THIS IS CONSISTENT WITH THE TRACK MODEL CONSENSUS.

"The latest model runs remain very tightly clustered along the coast of Texas, and have shifted a bit more
to the north ... with the consensus showing landfall near Freeport -- about 30 miles to the SW of Galveston.
The shift by the models is supported and consistent with the initialization data for the location and strength of the
strong, but now weakening High Pressure ridge over the Gulf coast states. This ridge is forecast to
weaken further and drift eastward during the next 48-72 hours, and will allow Rita to begin heading more
northwestward starting in about 24-36 hours, and then taking a NNW track for during the final 18-24 hours
prior to landfall. The storm's forward motion of 10Kts is not expected to change, and as long as it maintains
a forward motion in excess of 10Kts -- no appreciable weakening due to upwelling of cooler sub-surface waters
seems likely. What may support weakening prior to landfall is some increased shear to the northwest of the
storm as it approaches the coast -- and some weakening in the outflow channels. .

Rita could reach the coast as a CAT 5 hurricane, though a strong CAT 4 seems most probable.. But even if the
storm does weaken for any one of the above reasons -- with eye wall replacement being the most probable --
there is no doubt that a CAT 5 Storm Surge will hit the coast. With the still expanding and intensifying
circulation field -- the storm may produce a CAT 5 storm surge for a 60-80 mile stretch of coast line from near
the point of landfall and extending to the northeast."

No good!!!!

Reference - http://www.wunderground.com/blog/SteveGregory/show.html

Louisianna Border area 1120k (6.5%)
Beaumont ExxonMobil 349k
Port Arthur Motiva 285k
Port Arthur Premcor 255k
Port Arthur Total 234k

Galvaston (Houston coastal) area 1710k (10%)
Texas City BP 437k
Texas City Marathon 72k
Texas City Valero 210k
Pasadena Crown Central 100k
Deer Park Refining 334k
Baytown ExxonMobil 557k

Houston and nearby inland area 580k (3.4%)
Houston Citgo  270k
Houston Valero 83k
Sweeny ConocoPhillips 229k

Corpus Christi area 590k (3.5%)
Corpus Christi Citgo 156k
Corpus Christi Flint Hills 288k
Corpus Christi Valero 142k

 --J

picture here.

red= Hurricane force winds
yellow= tropical force winds