Underground weather, rain vs well levels

[SLOweather]

In another thread, DanS said:

Should be not much of a problem with the water table this spring (in the lower 48) between CA's multiple drenchers/floods and the East Coast's snows/floods.

I'm president of, and operate our 58 meter neighborhood mutual water company (for 15 years now). We have 6 hard rock wells. One of them went dry a couple of years ago, so I embarked on a routine of weekly static water level monitoring. It took me a while to make all of the wells probe-able. I got the $900 tape probe stuck a few times. That was not fun.

Anyway, after I started gathering readings into a spreadsheet, it took a while to figure out how to reduce all the data so it would plot on one graph. All of the wells but 2 are drilled into different formations, and all have differing static water levels.

I finally decided to use the first readings as an arbitrary "zero" and plot the change from that on the graph. Coincidentally, the rainfall in inches and well change in feet plot nicely.

"A" and "B" are in the same formation, so they track each other. Any "V" shaped drop in a well line is because I read the well level too soon after it was used and it hadn't fully recovered yet back to static level.

Well "4" is the one that went dry. It's still-off line and recovering nicely. It's static level when drilled in the 80s was 44 feet down. When I started measuring in Dec 2008, it was down to 338' from the top of the casing. Now it's up to 280', and rising so fast with all our rains that I'm going to need to figure out how to rescale the graph so the other well lines don't get squished down by Excel's autoscaling.

I reset the rain (dark blue line) to zero on July 1st, which is when the precip year starts here.

[graculus]

Have you looked into other automatic/electronic methods of measuring the water depth?

[casa manana]

SLO,

This is really neat data.  Can't wait to see an update over a longer time line.  Looks like some really complex coupling between when the rain falls and when the water table gets impacted (positively or negatively).  Sounds like a lot of work to actually make the measurements.

Being in North-Central Arizona, I am obviously very concerned about similar water table variability, but there the water table is probably best correlated with how much snow the Rocky Mountains got 15,000 years ago since we are fed by the deep Colorado aquifer.  We get our water at about 680' deep.

[SLOweather]

Have you looked into other automatic/electronic methods of measuring the water depth?

Yeah, I've already upgraded one well and have the gear to do another. We have a Rugid RTU system to control the wells based on tank level, so I add an analog input card to the RUG9 at the well, and then hook it to a Wika 4-20 ma pressure transducer I've dropped in the well. (And then edit the programming to add the new input, scale it and such, and send it to the master unit.)

That's a bit of a balancing act, since the transmitters are 50 PSI or 155 feet of water column. Since the pumps can be more than 200' below static water level, I can't just set the transmitter at pump level. If the static level drops too much over time, I'll need to lower the transmitter more.

The instrumentation has already paid off. I thought the well in which I installed the first transmitter was going dry, but watching the level drop very slowly as it pumped and then rise very rapidly after shutdown told me that the pump was just pumping about 5GPM more than the well was recharging. I throttled the discharge valve a bit and it's been fine ever since. 

[SLOweather]

SLO,

This is really neat data.  Can't wait to see an update over a longer time line.  Looks like some really complex coupling between when the rain falls and when the water table gets impacted (positively or negatively).  Sounds like a lot of work to actually make the measurements.

Thanks!

I use a Solinst 101 Water Level Meter to make the readings.



Now that all the wells are set up for it, it takes only about 20 minutes to read 7 wells, and another 10 to put it in the spreadsheet and generate a new graph.

[graculus]

Thanks for the info, very helpful, I've been thinking of doing this (just for one well) for a while. Many years ago someone told me that they monitored their well by monitoring the pump current so I thought I might log that through the year and see if there is a correlation. How do you sanitize the tape, or more likely I guess your system has chlorination.

[tomwxman]

Many years ago someone told me that they monitored their well by monitoring the pump current so I thought I might log that through the year and see if there is a correlation.

That's easy enough to do I suppose (I'm thinking "clamp meter" though you may have more sophistication in mind), though I wonder how much accuracy you would get out of it i.e. whether you'd actually see differences in current draw under "normal" conditions.

You probably know you can buy automatic current monitors for your well pump that shut it down in an underload (going dry) or overload (e.g. mud clogging condition). I have a Pumptec Plus for this; cheap protection for a $3K deep well pump.

But the easiest thing to do if your concern is running dry is to periodically run water into a bucket and ID your gallons-per-minute, comparing that to the original well report or at least a known good set of conditions. Then when the water table starts dropping to where your gpm goes down noticeably, you can expect problems.

Our well when dug produced about 10gpm, under drought conditions dropped to 5 or 6 gpm, but when it went below that I knew I was in trouble and sho nuff it stopped producing altogether within a week.

[SLOweather]

SLO,
This is really neat data.  Can't wait to see an update over a longer time line. 

Wow, it's just over a year since I started this thread!

Here's the latest graph. As Well 4 (which is still off line), recovers more, it forces all of the other traces flatter on the chart. More than half of them are above where they were when they were first drilled.

We've had 5.77" of rain in the last 4 days here. It will be interesting to see what impact that has on them this Friday and the next couple of weeks.

[SLOweather]

A couple of months ago, I got a wild hare and decided to dig out all of the meter records and such I have beginning from when we moved up here and activated the water company.

I started entering everything from the last 8 or 9 years into a big Excel sheet (dates, well meters, well electrical usage, and total shareholder use) and then played with different ways to crunch it.

So, now I know interesting things like:

How many cubic feet we produced each year, and how much electricity we consumed.

How many cubic feet per kilowatt hour each well pumps. If this started to slip, there's a problem with the pump.

What percentage of the total production each well does. This depends to some extent on the automated well rotation schedule.

And, how much loss we have in each billing period. That is, what percentage of water was metered to the customer vs: how much we pumped out of the ground. In our case "Loss" is something of a misnomer, as some unmetered water is used to backwash the sand filter, and some was sold from hydrants to trucks or on-site for construction water. Anything else is cumulative meter inaccuracies, leaks, or the time a truck knocked off a hydrant.

It's interesting as well, to see how the production and use rise in the dry season, and drop in the wet season.