Monday, February 17, 2014

field names

One of my big pet peeves when starting work at a site that's been around for a while is erratic/random nomenclature. I start looking at figures and old reports, and they're just a big mess of different names with no rhyme or reason. So it takes ages (and keeping some cheat sheets close at hand) to figure out which wells are located where and at what depths - critical data to figure out what the heck is going on out there.

Here's what often happens. Contractor #1 calls the monitoring wells BCS-1, BCS-2, BCS-3 ("Big Complicated Site", first well, second well, third well installed). Meanwhile, a local government agency or institution gets a grant and adds a few more for scientific purposes and those wells get labelled as GA-1, GA-2, GA-3 (government agency 1, 2, 3). Time passes and a couple of the wells get destroyed by local roadwork or are filled with something unspeakable by the locals and are decommissioned. Depending on the budget and the needs of the project, a well may be replaced (add R1, R2 etc to the designation, depending on how many times you replace the bugger) or may not, leaving a gap in the names. Then someone else does a study of the bedrock, specifically, and now you have BR-1, BR-2, BR-3 in addition to the other wells, which may or may not be bedrock.

Worse, perhaps there are a couple of properties that have been investigated separately and the contamination has been determined to be from one source. All those parcels get grouped together as one site. So you have multiple MW-1s, and so they get an additional ID: MW-1-AS (for place Across the Street) and MW-1-OT (for place Over There).

Then it gets more complicated. We often have well clusters or nested wells, so that each designated location has different depths associated with it. So Contractor #1 has BCS-1S for the shallow well in the cluster and BCS-1D for the deep well. But the government agency always uses A/B designations, so you also have GA-1A, GA-1B, GA-1C going from shallowest to deepest. And then someone else comes around and in addition to BR-1, BR-2 and BR-3, they also add some other bedrock wells to existing locations, so at BCS-1, you have the original well (BCS-1) and then you have BCS-1BR. I am being generous by assuming that each organization is using depth as a descriptor, and not calling the first well installed in a cluster, say, MW-1A and the second one MW-1B so that you have a mass of well IDs that make no sense whatsoever a month after the wells were installed.

You can't just go through and reorganize and rename all the wells at your convenience. Once those wells have an ID, they're in a database (likely filed with the state) and you lose continuity with previous reports at your peril. But you can try to pick IDs going forward that are consistent and follow whatever convention appears to be the most common at your site.


Dan McShane said...

I've gone through a fair bit of scrap paper trying to figure out multiple generation sample sites and wells.
Then there is the trick of writing a report that spans the entire history of site sampling in a manner that makes sense.

C W Magee said...

In exploration, management was always keen to make sure all logging and sampling designation was a opaque as possible, so that subcontractors couldn't speculate based on a logical system reconstruction and get us into insider trading trouble.

In theory the same could happen in environment, if a testing subbie started guessing whose site was contaminated and tried to dump shares based on risk exposure. Do y'all have to worry about that sort of thing in practice?