Water Well Journal
Issue link: http://read.dmtmag.com/i/498253
I worked in my family-owned water well business when I was growing up. I remember performing pump tests for customers on newly installed water wells and existing wells when we would change out pumps. Our test would consist of choosing the pump we felt would be sufficient for meeting the maximum yield of the well. We would install the pump after devel- oping the well until it was pretty clear. We would not develop so long as to pre- vent us from finishing the install and testing the well before the end of the day. We would perform several maxi- mum discharge surges following devel- opment to help finish clearing the well. Once we had developed the well in this fashion, we would let the well re- cover over lunch. Following lunch, we would measure the static level and acti- vate the pump. The pump was allowed to operate at maximum output for an hour or so. If the well pumped off, the discharge was throttled back with a choke valve until it quit pumping off. It would be allowed to run for an addi- tional 30 minutes to an hour. If the discharge remained stable, we would measure the discharge rate with a 5-gallon bucket and watch, or a measur- ing tape square, of the flow from the discharge pipe. With static water level data and discharge data in hand, we would notify the well owner of the test outcome. I know the eyes of some groundwater scientists reading this may be glazing over because this is not your standard practice. I only mention it because for many groundwater service providers this has been their method of practice. For the purpose of this article, we will classify the above as the pump test method as opposed to the pumping test method for aquifer characterization. We will look at the various pumping test methods used to assess aquifer char- acteristic and aquifer performance—and not the performance of the pump itself. There are numerous types of tests designed for specific applications and a multiple of different test analysis meth- ods to use in determining aquifer char- acteristics. We'll focus on a few of the more common methods to help you un- derstand some of their major differences and uses. Reasoning for Aquifer Characterization The need to characterize aquifer per- formance comes from the need to quan- tify and qualify groundwater availability for use. Many managers and developers when conceptualizing, designing, and building residential, industrial, and com- mercial projects seriously underestimate the need to characterize groundwater as a resource prior to construction. There are serious ramifications for overestimating available groundwater resources. I've heard of multimillion- dollar water-intensive projects being built only to find no groundwater or surface water resources were available. Type of Aquifer Tests The primary function of an aquifer test is to stress the aquifer and see how it performs. There are many ways to apply stress to an aquifer. We'll focus here on the pumping of water from wells as the method of testing the aquifer. The NGWA Press book Compendium of Hydrogeology explains an aquifer test as: Withdrawal or injection of measured quantities of water from or to a well and the associated measurement of re- sulting change in head during and/or after the period of discharge or injec- tion. Aquifer tests are performed to determine hydraulic properties of an aquifer. It further defines a pumping test as: Test performed by pumping a well for a period of time and observing the change in hydraulic head in the aquifer in order to determine the ca- pacity of the well and the hydraulic characteristics of the aquifer. (Porges and Hammer 2001) Pumping Tests Pumping tests are defined by the manner in which water is withdrawn from an aquifer. The type of information and data that can be derived from a pumping test is a direct function of the type and the inherent limitations of each test. The tests are classified by the number of wells or observation points needed and discharge rates along with the type of aquifer in which the test is performed. A multi-well test is designed to use a control well (pumping well) and obser- vation wells. Water is withdrawn from the control well and the changes in hy- draulic head are measured in the obser- vation wells. The measurements of the water level in the observation wells are collected in reference to time and distance from the control well. The single well test is designed to use a single production well and meas- ured observations are collected from the measurement of changes in hydraulic head or water level within the pumping well and are referenced to time since pumping began. Single well tests have limitations in available data. It cannot be used effectively to measure storativ- ity of an aquifer due to inherent loss of head associated with friction as water flows from the aquifer into the wellbore. The step or step drawdown test is most often used in the production or control well to estimate the flow rate for the constant rate test. It can also be used to estimate transmissivity, hydraulic conductivity, well efficiency, and maximum flow/acceptance rate. A step test generally consists of at least three or more pumping intervals of equal length of time, usually 30 minutes to two hours per interval (Weight and Sonderegger 2001). Pumping during each interval is held at a constant rate with an increase in production at the be- ginning of each interval. The production rate for each interval is an equally spaced fraction of full capacity (Kasenow 2006). The constant rate/steady state test can be used in a single or multi-well test configuration. The test is generally per- formed by pumping water from the pro- duction well at a constant rate for the duration of the test. The variations in water levels from the production well and observation wells are recorded at prescribed intervals based on logarith- mic cycles. The constant rate test is most often used for aquifer characterization associ- ated with the application of predeter- mined mathematically-calculated performance curves or models to data derived from the test. Each model is based on a specific set of assumptions Analysis of pumping test results laying over image of flowing spring. Courtesy of Raymond L. Straub Jr., PG FIELD NOTES continues on page 42 WWJ May 2015 41 Twitter @WaterWellJournl