A typical scene for drilling a groundwater monitoring well. The big boom arm on the back of the truck runs the drilling augers. The drill bits used are hollow stem augers and allow for a hole to be drilled in the soil and the well to be constructed inside the augers. This is really entertaining stuff here. The cows could barely break away and only lost interest when the rancher brought them their lunch.
Water is extremely important to life on this planet. Finding freshwater is a challenge to many of the inhabitants on earth, particularly in poorer regions. Even though the surface of the earth is roughly 70 percent covered by water, the vast majority of this water is salty ocean water. Freshwater amounts to just three percent of the Earth’s total water, and about a third of freshwater is groundwater. Groundwater is an important part of our tiny usable water supply, and we need to monitor it.
The remainder of the soil boring with the PVC well casing inside the hole.
In mid-February, while many other people were wooing their significant others for Valentine’s Day, we were hard at work establishing the first portion of NEON’s groundwater observation well network. The first wells are part of the Central Plains region (Domain 10) and are situated around the Arikaree River in eastern Colorado. The site is a pristine grasslands area which has a gentle topography of rolling hills with a small portion of the Ogallala Formation present along the southern boundary.
NEON collaborated with the U.S. Geological Survey to install the wells at the site. Initial characterization of the site, performed by the USGS, used a combined geophysics approach to attempt to define the anticipated depth that we would need to drill the wells. In other words, we used electricity to look for water. For those of you who haven’t used electrical resistivity (a geophysics method) to study aquifers before, here’s how it works: We apply a small amount of electrical current to the soil though a pair of electrodes and measure the resulting electric potential at a separate set of electrodes. Several electrodes are positioned in a line and the current and electrode pairs are automatically swapped around many times through all the electrodes. If you do this enough, swapping the current pair of electrodes and potential pair, what you get are a set of measurements that yield a map of the soil’s ability to conduct electricity. We use a fancy numerical modeling technique called an “inversion” which takes the raw data that we collect from the electrodes and computes what the subsurface architecture must look like to yield the electrical signals we collected.
The reason this works so well is that water conducts electricity and we can “image” underground soils and soil moisture this using this technique. More accurately, we can use this technique to look for differences in the electrical conductance of the soil layers and make inferences as to where there is water or at least a shift in the underlying layers. We also make some eye-catching images.
The results of the inversion model for the site. The upper red and pink layers are the unsaturated dry soils and the top of the yellow band is where the water table is. The transition between the green and light blue layers is the thick clay layer before you reach bedrock. This image was the start of defining how deep the observation wells would need to be drilled to reach water.
Based on the geophysics analysis, we estimated that the thickness of the soil layer was about 30 feet. While drilling the wells we reached a thick clay layer at approximately 30 feet depth, marking the end of our drilling and also confirming the interpretations from the geophysics. At this site all the wells were drilled to approximately 30 feet deep as we encountered the clay layer at the same depth. The plan was to have a screened section of well (where the groundwater can enter and exit) long enough to capture the full seasonal and long-term oscillation of the groundwater table (fully water saturated soils).
Here is a quick conceptual drawing of the well.
We opted to install a protective steel casing around the PVC casing. A barbed wire fence was installed around the casing to keep the cattle away when they are being ranched in the research site area of the property.
In the coming months the NEON aquatic team will be installing sensors in the wells and in the stream to begin prototyping activities of the sensors and their data streams. Once we install water level sensors in the wells we’ll be able to track the oscillation of the groundwater table and will begin to establish the historical groundwater record in this region. The data collected through this network will establish a database that will allow local farmers to track the generalized effects of irrigation on groundwater levels in the region. This data will also be useful to the greater scientific research community as a part of the puzzle piece for examining continental scale ecological drivers and responses, among many other potential studies.
Stay tuned for more developments!
Loading tweets...
4 comments
Allen Macfarlane says:
May 2, 2012 at 5:52 pm (UTC -6 )
Why not tap into the network maintained by the Kansas Geological Survey and the Division of Water Resources, Kansas Department of Agriculture. They have been monitoring changes in water levels in the High Plains (Ogallala) aquifer for decades?
Michael Fitzgerald says:
May 3, 2012 at 10:02 am (UTC -6 )
Hi Allen, thanks for reading the post! The well network at this site serves a few purposes, part of which is looking at changes in water levels in the region, which we could tap into the KGS network to do some of. The main focus however for this network is to be able to observe hydologic responses between groundwater and the stream at the site, so we needed the wells to be local to the stream.
Drilling Equipment Man says:
May 10, 2012 at 10:49 am (UTC -6 )
Really like the mapping with resistivity, that’s a fantastically clever idea. Would I be right in thinking the diagram there is just a cross section of a 3D map or do you just plot a linear course from say a river?
Michael Fitzgerald says:
May 11, 2012 at 4:22 pm (UTC -6 )
Hi Drilling Equipment Man,
Thanks for your interest in the article! And yes I agree, resistivity mapping of the subsurface is a great idea, and I would ultimately like to do this at all NEON aquatic sites. To answer your question, the image you see is called a psuedo-3D or 2.5D tomogram and is plotted by putting in a line of electrodes that transects the stream perpendicularly. Though the transect could be orientated in any direction. The reason this is called 2.5D is that the electric field generated at the electrodes does transmit in 3D but we compress it into a 2D image. With enough transects positioned parallel to each other and with cross-transect measurements we can construct an image of the subsurface in full 3D. To give proper credit, this resistivity image was made by the USGS staff (Jared Abraham, Lyndsay Ball, and Burke Minsley). As a postdoc at Penn State University I used resistivity coupled with a saline tracer to examine hyporheic exchange in a small stream in Oregon. If you’re interested more info can be found here. If you don’t have access to the journal article and are interested I’d be happy to email it to you.