a) there’s no interference on a single well
b) we barely had MWD at all back in the stone age, let alone look ahead 
Tapioca Dextrin and Oredigger77, you guys are a perfect example of why people love the Dope so much. We have experts in EVERYTHING on this board!
(I don’t know if you consider yourselves experts, but you’re experts to the rest of us…)
One of the companies that provides services for well collison (but more usually well avoidance in congested top hole situations) process the data from the 3 axis magnetometers in the MWD tool and takes multiple rotational surveys and then compares the magnetometer readings with expected field to determine the direction of the interfering well.
It works well, obviously you need to get close with conventional surveys.
Incidentally people have been considering using azimuthal resitivity tools (the HAL azi res tool, Periscope from the smurfs and BHI has a somethingTrak) to detect wells instead of adjacent bed boundaries. It is in the ‘possible but not certain’ catergory and needs more work.
That’s the Azitrak. POS IME.
I think the tool physics are good, they are pretty much the same for the 3 providers, build quality is another issue - but data rate is key for those types of tools, down hole processing just can’t do the inversion, shitty RT data rate = useless log. That and if the res contrast isn’t sufficient the results are also crappy.
The whole falling off the end of the drill string was bit of let down for me.
Sorry for the hijack, but looking at this diagram got me wondering: how much oil is actually in the reservoir they drilled into? How extensive is the deposit? Do they have a detailed map of exactly how thick the reservoir is and how far it extends?
And how accurate is their cross-section graphic in that PDF? Are the layers shown just some graphic artist’s concept, or do they actually represent data that BP has on the composition of the seafloor in that area?
That would tend to lead to a negative customer exprience, did they buy a nice lunch?
I believe 50MM bbls is the estimated recoverable reserve, so there would be more actually in the reservoir which would not be produced. Also it should be noted that all the recoverable reserves would not be recoverable from a single well. The reservoir is not a big void filled with oil. The oil is contained in the pore speace between grains of sand.
The reservoir reserve estimates will be made from measurements of the porosity of the rock, analysis of the area of extent of the reservoir based on sesimic images and this would be calibrated by actual mesurements (logs) of the reservoir.
So they have a pretty good picture of what is going on, where the formations start and stop and what the approximate extendt and shape of the reservoir is, there are unxcertainties associated with those measurements.
As for the accuracy of the PDF - I would say ‘for illustrative purposes only’