The point of a water well is, of course, to produce water, so the objective is always to place it in a spot where the aquifer is plentiful. Drilling a well that won’t yield enough and then having to drill again somewhere else is anything but ideal. Any well driller would be pleased to have a solution that takes the guesswork out of the well placement process.
With that in mind, an aquifer mapping project is being worked out in Sioux Falls, S.D. The city, the U.S. Geological Survey (USGS) and geoscience company CGG are collaborating to paint a comprehensive picture of what lies below in the Big Sioux Aquifer.
Over the course of four days in October, CGG towed a cylindrical sensor about 30 meters above the section of the aquifer that Sioux Falls has water rights to. It collected raw data about the physical characteristics of the subsurface, which will be handed over to USGS to process further into a comprehensive groundwater model for the city to refer to.
The CGG sensor technology, named RESOLVE, has been used before on other USGS aquifer mapping projects, but this is the first time it’s being used in the state of South Dakota. The main advantage of RESOLVE is that it is airborne, which means it takes away the challenges associated with ground-level data collection.
The tried and true way to zero in on an aquifer is to drill multiple test wells and analyze the formation layers those wells punch through. “The shortcoming of that though is that that really only gives us information for one point, just that spot where you’re drilling,” says Mark Anderson, director of the USGS South Dakota Water Science Center. “So, if you went over a mile to drill again, you get a different story. Then what you have to do is infer what’s going on between the two wells. With this technology in the air, we have information for every two meters of the energy that’s put in the ground. It is not just a simple cross-section.”
There are a number of other ways to map aquifers, including a ground version of the electromagnetic sensing that RESOLVE does. It involves inserting copper spikes into the ground, stretching wires out and inducing a current into the subsurface from which resistance is measured. However, similar to the effort that goes into drilling one test well after another, a great deal of manpower is required to travel the distance an aquifer can cover, setting up and carrying around equipment.
In addition to accounting for the majority of the aquifer where test wells aren’t drilled, an airborne approach is simply easier to carry out because tough terrain and prohibited access to private property are no longer obstacles. Even better, the use of the RESOLVE sensor is economical. “We looked at the cost,” Anderson says, “and the cost would actually be cheaper than doing it from land.”
The RESOLVE technology is a frequency- domain electromagnetic system. Housed inside the cylinder are transmitter and receiver coil sets that are tuned to a wide range of frequencies. The coils generate an electromagnetic field downward into the ground and the interaction energizes the subsurface, allowing the sensor to measure conductivity and resistivity. “From there, we can interpret and start to understand what sort of changes there are down there; what structures, what conductive features, groundwater, that kind of thing,” says Adam Shales, business development manager at CGG’s Toronto branch.
The mapping technology offered by CGG completely changes the initial data collection step, increasing both speed and capability, which translates into more quickly completed models that convey more aquifer characteristics than traditional methods. “In this particular project, we’re offering the geophysical technology and the personnel with the expertise to operate it,” Shales says. “So we’re acquiring that data, and then we’re also doing the processing on the data to get it to a point where there’s products that we can deliver and hand off to USGS, that will take it and do further interpretation, and work with it from there.”
Once the data from RESOLVE is in the hands of USGS, further data processing will be conducted to more clearly interpret what the information means as far as water is concerned.
Those interpretations, Anderson says, will include where the water table is, where the most permeable zones in the aquifer are, and characteristics of the geology. “That’s interpreting the data the company gives us, but then the next step is we’ll migrate those data into this groundwater model, where we’ll numerically simulate the whole aquifer,” he says. “This allows us to run simulations into the future about what would happen if we put a well here, and so on.”
The groundwater model Anderson is referring to is called MODFLOW. Regularly used by USGS to better understand aquifers across the country, the model offers a numerical representation of what scientists believe represents the natural system. “Models can take on many different forms, depending on what questions we are trying to answer,” says Joshua Valder, USGS hydrologist at the South Dakota Water Science Center.
Valder, a North and South Dakota groundwater specialist, explains that horizontally the model is made up of rectangular cells that form the model grid. Vertically, the aquifers are represented by overlapping layers of the horizontal model grid. The grid spacing, or cell size, can be large or small, depending on the intended use of the model. If the spacing is large, individual model cells represent a larger area in reality. If the spacing is small, individual model cells represent a smaller area in reality. Smaller cells allow a lot finer representation of the aquifer and, subsequently, the results can be interpreted at a much more local scale. Larger-model cells are typically intended for regional-scale models.
Primary model outputs are hydraulic head (altitude of water level), drawdown (the lowering of the water level due to factors such as pumping) and groundwater flow, which is calculated for each cell in the model area, Valder says. Optional outputs include, but are not limited to, streamflow, spring flow and particle or contaminant transport. The hydraulic head output represents the numerically calculated value of where the water level should be in a particular cell at a given point in time, based on stresses applied to the system like groundwater withdrawal or recharge. The drawdown output represents numerically-simulated change in head values from a reference time in the model. The final maps that are created can then be used to look at changes in the model area over time and how groundwater will react to various stresses.
“MODFLOW is not a program that will allow the user to look at a visual representation of the results,” Valder says. “It only calculates results and provides the output files in text format.” Other programs are needed to process the output files to get a visual representation of the results. There are USGS and proprietary programs that can be used to process the results and visualize what was calculated by MODFLOW.
A geographic information system (GIS) is commonly used to look at model results and create maps from those results. Other software that can be used includes, but is not limited to, MODELMUSE, Model Viewer and GW Chart.
For a city like Sioux Falls, having a good sense of where groundwater stands is vital. Jeff Dunn, Sioux Falls water principal engineer, says about 40 percent of the water supply comes from a regional water supplier. The other 60 percent comes from surface water or groundwa- ter. Of that 60 percent, about 99 percent is groundwater, he says.
“Our wells aren’t getting any younger.” Dunn says. “We need to streamline and make our raw water collection systems more efficient. The city is growing quickly and we need something that we can plan, not just five or 10 years, but 20 to 25 years out into the future.”
The project is motivated by the intention of Sioux Falls to get rid of old, inefficient wells and replace them with new ones. “When you’re trying to pull water out of these older wells that are not functioning properly, we’re wasting a lot of money on pumping and maintenance. Hopefully when we put in these new wells, we’ll just take the old ones offline and abandon them,” Dunn says.
The final groundwater model, which will be generated by the USGS MODFLOW software system, will give Sioux Falls key information from which to base decisions for new well location and type, he says. “The model will tell us, of course, where the most porous soils are; the model will tell us how the water moves in and around the aquifer; the model will tell us how the aquifer works during droughts and periods of heavy rain.”
Understanding those characteristics will help planners forecast the most plentiful parts of the Big Sioux Aquifer. It will help them cut down on more expensive horizontal collector wells by making them aware of spots capable of pumping just as much water with cheaper vertical wells.
Referencing the groundwater model will be easier than ever, considering the advancements in visual mapping technology that have taken place. “In the old days, it was just a sea of numbers,” Anderson says. “We’d get lots of values and we’d have to interpret the values based on grids and so forth, but now it’s all loaded into visual tools and you can actually just fly through the aquifer in 3D. The model will be able to provide a visualization on a screen of what the aquifer looks like inside, and if you’ve got this well and the well starts pumping through 50 years, and then what happens to the water table.”
Getting to the end point of a comprehensive model will take a good deal of time. The estimate is two to three years before a final technical memorandum is released.
In the meantime, Dunn says, some good preliminary results have already been obtained from USGS. “We kind of had an idea where some of the more porous soils are, but we were surprised and pleased to learn that there are still other areas within city property that we could go to immediately if we wanted to drop in a new well or wells.”
As with all USGS MODFLOW data and aquifer reports, the information will ultimately be made available to the public and, while the airborne aquifer mapping approach decreases the need for test well drilling work, it increases the certainty of where the best place for a water well is for drillers. “It ought to improve their success; in other words, being able to drill water wells in areas that can produce more water for their clients,” Anderson says. “So, it’s a way of looking into the subsurface without having to say, ‘Well, let’s see. Where should we drill? Oh, I guess we’ll put it over here.’”