No problems. It's a scary disease,and I've already lost access to several colleagues because of it. "Run in circles, scream and shout" isn't helpful. Neither is overconfidence.

30 seconds after typing that, the stumps of the 1880 - odd Tay Bridge Disaster came into view. In theory, the bridge was strong enough. In practice. .. not.

In theory, there is no difference between theory and practice, and in practice. . .

Trite, but true.

I spend a significant part of my working life teaching this to people fresh out of university. I don't see me running out of this sort of work.

The argument could equally well be put as, for a decade or so, companies have been handing out free samples of their drug (information ) and have now started to call in the debts owed by millions of new addicts.

I lived in an apartment under a crack cocaine dealer when I got my first 2400bps modem. I recognised the deal being offered then. The deal is the same now, but the free gear is being stopped.

Next article?

Hippos? The whales that did not go to sea?

I had heard of Hantavirus before - typically a SW-desert area disease, if my memory serves well - but I wasn't aware that it was a haemorrhagic virus.

So ... how many nurses are there trained and experienced in handling these diseases in the USA? Or, is it a case that essentially everyone who does treat the disease is going to be 30 minutes out of their first training course / briefing? It does take time to change a training course into a practised, skilful performance of the required tasks. That's why we, err, practice things.

It's also why the NE of England had a full dress rehearsal for an Ebola case collapsing in a public place a couple of days ago.

I wonder how long until they go on strike over this?

The unions (RMT, ASLEF) have been pointing out that without drivers, there is going to be no-one to react to suicides (which close tube lines and stations for around a day each time, causing huge disruption, and it happens frequently). The train drivers are also trained to operate the heart attack machines on most trains and platforms, and routinely save the lives of passengers by being present to react.

You're right - there are plans in place for strikes to resist the de-drivering of trains.

Have you any idea what percentage of the UK's GDP might be affected by such a breach in the underground transport system?

I do have an idea that it was really stupid to build in such a place.

I blame the Romans myself for being so inconsiderate as to build a city at a fording point of a major river. If only they'd forded the river up on top of a hill somewhere, London wouldn't have any problems with flooding.

Still, you can avoid it by choosing where to live in the city. My city has a harbour that is inconveniently at sea level, and the harbour often floods in heavy rain or high tides. But my house is 25m above the harbour level and has a site slope of about 2 degrees, so rain water runs away down hill. It's not rocket science.

Many tests for hydrocarbons are cross sensitive, such as a sensor for Propane will detect gasoline, natural gas, butane, etc.

Very true. I've put together and worked gas detection machines for three decades now. Different sensor technologies have different cross sensitivities. It sounds as if your experience is with IR absorption sensors, probably on one wavelength (more wavelengths reduce cross-alkane sensitivity, but increase cost ; single wavelength sensors are fit for purpose for flammability/ explosion risk monitoring, but not for interpretive/ analytical work).

What sensor is used, what is the sample time, what else is it sensitive to, and were there any significant accidents or releases in the area recently?

FTFA ; sorry, the second FA

In the study published online today in Geophysical Research Letters, a journal of the American Geophysical Union, researchers used observations made by the European Space Agencyâ(TM)s Scanning Imaging Absorption Spectrometer for Atmospheric Chartography (SCIAMACHY)

FTFIWP (From The Fucking Instrument's Web Page, http://www.sciamachy.org/

is a passive remote sensing spectrometer observing backscattered, reflected, transmitted or emitted radiation from the atmosphere and Earth's surface, in the wavelength range between 240 and 2380 nm.

Well, that tells me enough - medium UV to medium IR, plenty of appropriate absorption bands there. If they say "methane" they mean "methane."

If it was from the soil, soil based sampling should have seen this concentration long ago in gas exploration.

Looking for actual figures ... the absolute values they're recording are around the 1740-1760ppbv level. Every contractor I've worked with (apart from our own in-house machines) claims to be able to detect at this level in drilling gas results ; none of the contractors I've investigated in depth have been able to prove these claims. When I ran our own machines in the field, I could get them down to this level of sensitivity, but it would take several hours a day of adjustment to keep them there - which is not something you can really spare the time for from your other duties. These days, I set up machines to just detect at this level, then leave it at that level. We're looking for aditions of methane (and other hydrocarbons) to our gas stream from drilling a well ; we're not interested in the atmosphere except as a source of noise to be accounted for.

A quarter is the weight of a tuppenny piece and should cost no more than thirty quid.

Cows release methane by burping, not farting.

With such extensive knowledge of bovine anatomy, can you tell me if Frank Zappa was right? With a tongue like a cow, can she really make me go 'Wow!"?

some estimates put it on the order of being 80 times more powerful.

I don't see that quote in the AGU article, so I'm going to guess that comes from "Vice" ; quite why you'd take a figure from a website about prostitution over one from a geophysical union in a discussion over a science question, I don't understand.

It looks as if Vice's cut'n'pasters (I can't call them journalists) grabbed several figures from various sources and just said "up to [the highest]" without doing any checking on why they were getting a large range. From other sources, I'd have said that methane was about 23 times as powerful a greenhouse gas as CO2 ; upthread hyades1 says 25x ; meh to that difference. Where an 80-fold ratio comes from, I've no idea. That could be a figure for some CFCs (they are also potent greenhouse gasses as well as ozone-fuckers) ; it might be true for ethane or higher alkane gasses (minor components of some natural gas deposits) ; or it might just be a figure made up from whole cloth. but the 80x figure is clearly wrong.

The half life of methane in air (against oxidation to CO2) is around a decade ; there are probably considerable temperature effects in there, making an average a bit dodgy to measure. It's not centuries (we'd see it in the records of previous major releases of methane to the atmosphere, such as my bread and butter Palaeocene-Eocene Thermal Maximum ; it's not days (otherwise this plume would have oxidised away ; it's somewhere in between.

OTOH, because the leaks haven't been accurately measured (Could they be?),

Which part of the original article (first post in the thread) did you not read? Was it the bit that says

Frankenberg noted that the study demonstrates the unique role space-based measurements can play in monitoring greenhouse gases.

âoeSatellite data cannot be as accurate as ground-based estimates, but from space, there are no hiding places,â Frankenberg said.

Surely everyone knowns that, for small values of theta (the situation that is being discussed), then sin(theta) ~= theta (in radians). It's a basic trigonometry result, used all over the place where you're looking at small angles. It's probably at the root of the original articles interest in using the sin() calculation to expose the (in-)accuracy of the system's value for pi.

So, if your angle gets to low values (choose where that is for yourself) , then you switch to using that approximation and cut out a large chink of calculation. Saves a lot of hassle. It's as basic as checking for "divide by zero" situations.