The State of Missouri, through the Missouri Geological Survey, and later the Missouri Department of Natural Resources, has operated and maintained a network of groundwater-level observation wells across the state for nearly 50 years. The wells are equipped with recorders that measure and record the distance from land surface to the water level in the well. Water levels in the wells change in response to changes in natural conditions as well as changes caused by groundwater production from the aquifers. During extended periods of dry weather, depth-to-water in most wells, especially those penetrating relatively shallow, unconfined aquifers, increases as water drains from the aquifer to emerge at springs and streams. Groundwater recharge is ultimately provided by precipitation, so periods of wet weather cause depth-to-water to decrease in the shallow aquifers. Wells drilled and cased into deeper zones of unconfined aquifers generally respond more slowly to recharge, except in some Ozark areas where groundwater recharge is extremely rapid. Confined aquifers commonly show little response to precipitation.

Groundwater levels also change in response to pumping. Groundwater produced from private and municipal water supply wells throughout the state removes billions of gallons of water each year from Missouri’s aquifers. But this accounts for only a fraction of the total amount of groundwater that is used. Industrial and agricultural water use continues to increase yearly, placing additional stresses on the groundwater resources of the state. When groundwater production exceeds recharge, the water level in the aquifer begins to decline. The observation well network is a very effective tool for monitoring the quantity of water available from Missouri’s major aquifers. The observation wells are akin to the oil dipstick in a car. It allows the owner to monitor the level of a vital fluid that protects the engine from damage. Observation wells perform much the same task. They also allow the level of a vital fluid…..groundwater….. to be monitored, hopefully allowing problems to be recognized and corrected before long-term effects result.

Systematic groundwater-level data collection has been an ongoing activity since the first observation wells were installed in 1956. The late 1950s were a time of severe regional drought. Many shallow wells were failing due to declines in water level. The first observation wells were installed to help answer water-supply questions. The initial network consisted of about 23 wells, most of which were drilled by the Missouri Geological Survey and Water Resources (predecessor of the Missouri Geological Survey and the Water Resources Center) on state-owned property such as at Missouri Department of Transportation (MoDOT) maintenance facilities. A few of the early observation wells were unused water-supply wells donated by individuals, municipalities, or companies. Despite the nearly 50 years that have passed since the first observation well was installed, eighteen of the 23 observation wells established in late 1950s, which are considered landmark stations, are still in use today. These long-term stations are Lamar, Duck Creek, St. Joseph, Jefferson City, Delta, Malden, St. Clair, Atlas Powder, Longview, Fredericktown, Hannibal, East Prairie, Steele, Halfway, Naylor, Columbia Bottoms, Sikeston and Fairview.

The early equipment that was used to measure and record water level consisted of mechanical recorder that used a moving ink pen and long roll of graph paper to record changes in water level. A float was attached to one end of a beaded cable and lowered into the well where it rested on the surface of the water. The beaded cable passed over a float pulley on the recorder, and went back into the well. A counterweight was attached to the opposite end of the cable to keep it taut. The beads on the cable mated with indentations on the pulley, which kept the cable from slipping. When water level changed, it caused the float to rise and fall. That movement was transferred to the pulley on the recorder by the cable. As the pulley rotated it moved the pen back and fourth, drawing a line on the paper. The long roll of chart paper moved through the recorder at a constant rate of 1.2 inches per day. The recorder was powered by gravity; a 15-pound weight attached to a long, thin cable, was suspended in the well directly below the float pulley. The weight-driven cable would slowly unwind, providing the energy necessary to advance the chart paper. A very accurate mechanical clock governed the speed at which the chart paper was transported through the recorder. The clock-weight cable had to be rewound every few months. These recorders were extremely durable and reliable. In fact, many of the recorders purchased in the 1950s and 1960s were still in use in 2000 when the network was updated with new equipment.

Leupold & Stevens A35 - Mechanical water-level recorder
Figure 1 - Chart-type (Leupold and Stevens A35) mechanical
water-level recorder.

The mechanical recorders worked well but had two major drawbacks; one of which was not apparent until computers came into use. A major drawback of the mechanical recorder is the time necessary to process the data that they collected. The recorders are geared so that a water-level change of 1 foot causes the pen to move 2 inches across the width of the paper. This allowed very minor changes in water level to be accurately recorded. But since the paper is only 10 inches wide, it would record a maximum water-level change of only 5 feet before the pen reached the edge of the paper. When the pen reached the edge of the paper it reversed direction and continued to record data. Thus, at each pen reversal, the up-down direction on the chart paper changed. It took more than 36 feet of paper to record a year of data. Even if the charts were collected every few weeks, they sometimes contained more than a dozen reversals. Accurate manual water-level measurements must be made each time the chart paper is retrieved in order to accurately interpret the chart data. All of this made data interpretation a complicated and time-consuming activity. Furthermore, the data were analog. For many years, the charts were processed by hand and average 5-day water levels were calculated then plotted along with area precipitation information onto long-term paper graphs. Later, when computers came into use for data storage and display, another step was required to create computer-compatible digital data, either by using a digitizer or manually entering values interpreted from the chart into a computer data file.

Chart Paper from mechanical water-level recorder.
Figure 2 - Chart paper from mechanical water-level recorder.

The next generation of equipment solved some of these problems, but in the process created a few new ones. In 1979 and 1980, electromechanical digital water-level recorders were purchased to partly replace the aging mechanical recorders. By this time, the network had grown to about 32 wells. Almost all of the wells added since the late 1950s were abandoned or otherwise unused water-supply wells that were loaned to the department to use as observation wells. The digital recorders measure depth-to-water much like the mechanical recorders. They, too, used a float that rested on the water surface in the well. But instead of a beaded cable a 3/8-inch wide, flat stainless steel tape was used to transfer movement of the float to the recorder. The stainless steel tape contained 0.1-inch diameter perforations spaced precisely 2.4 inches apart (5 perforations per foot of tape) that mated with pins on the float pulley. This prevented the flat tape from slipping on the pulley and maintained a high degree of accuracy.

Long Term Plot
Figure 3 - Raw data from a mechanical recorder (top) and
long-term data plot (bottom).

A counterweight to maintain tension was attached to the free end of the stainless steel tape, which went back down the well. One revolution of the recorder pulley was equal to a water-level change of precisely 1-foot. The recorders would record a maximum water-level change of 99.99 feet before reaching zero. Unlike the mechanical recorders, which plotted data continuously, the digital recorders used electronic timers that could be set to record data in 5-minute time increments ranging from a minimum of 5 minutes to a maximum of 60 minutes. The digital timers that were used were very accurate, generally to within a few minutes per year. To extend battery life, all of the recorders were set to record every 60 minutes, on the hour.

Stevens Model 7001 Front View
Figure 4
- Stevens model 7001 electromechanical digital
water-level recorder.

The data were recorded on paper punch tape. The float pulley, moving in response to changes in water level, changed the positions of a complicated set of drums. When the timer triggered the recorder, an electric motor moved the paper tape toward the encoder drums. Ridges on the drums caused pins to punch a line of holes in the paper tape. After the holes were punched, the paper tape was advanced one unit, and the recorder remained idle until the timer reached the end of its next timing cycle. Each of the holes left in the paper corresponded to one of 16 values. The values were .01, .02, .04, .08, .1, .2, .4, .8, 1, 2, 4, 8, 10, 20, 40, and 80. To record a depth of 63.17 feet, for example, the recorder would punch a line of holes in the tape corresponding to 40, 20, 2, 1, .1, .04, .02, and .01. To read the depth-to-water from the tape, the values on each line were added together. It is possible to decipher the tapes by hand, but a digital tape reader is generally used to read them. The tape reader can process a year of data collected at hourly intervals (8,760 data points) in about 4 minutes, and enters those values into a computer data file. Correction values were added to the tape values for water levels deeper than 99.99 feet.

Stevens Model 7001
Figure 5 - Stevens Model 7001 electro-
mechanical digital water-level recorder.

The greatest advantage the digital recorder had over the mechanical recorder was the relative ease that data could be processed and used. But like the mechanical recorders, the digital recorders employed many moving parts that were subject to mechanical wear. Because of how and where they are used, the recorders must operate in a harsh environment. Temperatures can range from well below zero to more than 120 degrees Fahrenheit. Humidity can be as high as 100 percent. Such conditions are conducive to corrosion and mechanical failure. The high-torque, low-speed motors that powered the mechanical parts of the units were very expensive to replace, and long-term wear decreased their mechanical efficiency. The digital timers, like all solid state electrical devices, also occasionally failed. In the 20 years that the digital recorders were in use, many of them needed a new motor, a new timer, or both. The large dry-cell batteries that typically powered the units needed replacing about every 6 months. By the late 1990s, the digital recorders were obsolete. The company that manufactured them had ceased supporting them, and it became nearly impossible to purchase replacement parts for them.

Both the mechanical and digital water-level recorders required regular field visits to recover the raw data and to perform routine maintenance. With both instruments, it was impossible to know if they were working properly without physically visiting the stations and recovering the charts or tapes. Most of the time the units work quite well and collected data uninterruptedly. However, mechanical or electrical failures, vandalism, extreme weather conditions, and other factors caused them to occasionally fail. Failures could not be identified without field visits, which resulted in lost data.

By the early 1990s, the network had grown to include about 46 wells. The total number fluctuated, though, from 40 to as high as 50 installations. Occasionally, recorders were placed on wells to address a particular problem or concern, and may be removed after collecting only a year or so of data.

Digital Tape Reader
Figure 6 - Digital tape reader.

In 1999, an expansion approved by the General Assembly resulted in the purchasing of new equipment to completely replace the aging recorders. Data collection platforms consisting of electronic data recorders and digital encoders were installed at the existing observation wells, and the size of the network was expanded to 70 wells. There are some similarities between the new data-collection system and the old ones. The same floats and stainless steel tapes used by the digital recorders can also be used by the data collection platforms. The movements of the float are measured and translated into a computer-compatible format by a device called a digital encoder. Data from the encoder is fed into the recorder where it is processed and electronically recorded. The only moving parts are the encoder pulley and shaft, and even it uses an optical counting system to detect the movementrather than a mechanical system. Pressure transducers are replacing the digital encoders at some of the installations. A pressure transducer is a device that is placed in the well below water level that can accurately measure the height of water above it. It is linked by cable to the data recorder. A small vent tube incorporated into the cable allows transducer to correct for changes in atmospheric pressure. The transducers are most commonly being used in wells that are crooked or have irregularities that cause the float to catch, or in wells that have very deep water levels.

Sutron 8210 data collection platform
Figure 7 -
Sutron 8210 data collection platform

The data recorder or data logger is the heart of the system. It is essentially a dedicated microcomputer that is programmed to control all of the aspects of data collection. The most significant advance is how the data are recovered. Like the mechanical and digital recorders of yesteryear, data collected by the new system are stored internally in the recorders. The old systems used paper, the new one uses solid-state memory chips. But unlike the old equipment, the data are also transmitted back to the office using the GOES satellite system. Each recorder unit contains a small UHF radio transmitter that has a power output of about 7 watts, a little more than a typical CB radio. Its target is a GOES satellite in a geostationary orbit above the equator some 22,000 miles from Earth. A small, directional hi-gain antenna and a slow data transmission rate allows the low-power radios to accurately and reliably send the data over such a distance. Each recorder unit is assigned a unique identification number and is given a time window during which the data must be transmitted and a data channel that must be used for that station. So, for a one-minute period every four hours, the GOES satellite “listens” on that data channel for a signal from only one particular unit. No other stations should be transmitting to that satellite at that time on that data channel.

Sutron 8210 data recorder and digital encoder
Figure 8
- Sutron 8210 data recorder (top) and
digital encoder (bottom).

The data are received at an Earth station in Little Rock, Arkansas, and automatically sent to Rolla via fiber-optic cable. The Department of Natural Resources partners with the U.S. Geological Survey to retrieve and store the data, and to distribute it via the Internet. Data are generally available within a few hours after collection. Besides water-level data, diagnostic data for each of the data recorders is also transmitted. This allows the performance of each of the data collection platforms to be remotely monitored on a regular basis. If a station quits transmitting data, a technician can quickly make a field visit and repair the problem. In many cases, even if the station quits sending data via satellite, the data are still stored in the data recorder. The data collection platforms are designed to be self-sufficient. High-capacity sealed lead acid batteries power them. Photovoltaic solar panels keep the batteries fully charged. The batteries have ample capacity to power the installations a full year, even if the solar panel becomes inoperative.

The data recorders are capable of receiving information from a variety of sensors. Currently, only water level data are being collected. In the future, precipitation gages, soil moisture sensors, or barometric pressure sensors may be installed at some sites to better monitor hydrologic conditions in those areas.

Note icon.  Note: The inclusion of manufacturer’s names and equipment model numbers is only for purposes of technical accuracy. It does not constitute endorsement by the Missouri Department of Natural Resources or the State of Missouri.