« PreviousContinue »
Aquifer Thermal Energy Storage
In May 1980, the University of Minnesota and the U.S. Geological Survey began a project to evaluate the storage of hightemperature water in the deep FranconiaIronton-Galesville sandstone aquifer (hereafter called ''the aquifer'') and later recovering the hot water for space heating. High-temperature water (302°F.) generated during the cooling of the University's electrical generation facilities will supply heat for storage at about 800 feet below the surface. Figure 1 shows the Aquifer Thermal Energy Storage (ATES) system which uses two wells 860 feet
apart. Water is pumped from one of the wells in the aquifer through a heat exchanger where heat is added or removed. The water is then returned to the aquifer through the other well. Experimental testing of the ATES system consists of a series of hot water injection, storage, and withdrawal cycles. Each cycle is 24 days long, and each injection, storage, and withdrawal step of the cycle takes 8 days.
The U.S. Geological Survey designed the data-collection network, collected and analyzed the data to describe the hydrologic properties of the aquifer, and con
Figure 1.- Block diagram of
the Aquifer Thermal Energy Storage site, University of Minnesota, St. Paul.
structed a ground-water-flow and thermalenergy-transport model for evaluating the efficiency of the ATES system.
A comprehensive network for data collection, data storage, and data reduction has been designed to monitor temperature and pressure changes during the ATES test cycles.
Temperature and pressure measurements are combined in observation-well tests at approximate distances of either 20 or 40 feet away from the production wells (fig. 1). All pressure and temperature data are transmitted from the observation wells by way of buried cables to a central data logger where measurements are viewed independently or stored on computer magnetic tape for later analysis.
Two computer models have been constructed to simulate the movement of ground water and heat. The computer code for the models, developed under contract to the Geological Survey, has been used for calculating the effects of liquidwaste disposal in deep saline aquifers. The code has been successfully used to model storage of warm (130°F.) water in a shallow unconfined aquifer in Alabama but never for high-temperature water in a deep confined aquifer, such as in Minnesota.
Application of the calibrated model will consist of simulating various combinations of heat injection, storage, and withdrawal periods and various pumping-injection rates and analyzing the effects of the combinations on efficiency of the ATES system. One measure of system efficiency is the ratio of heat recovered to heat injected, which can be expressed as a percentage. The higher the percentage, the more efficient the ATES system.
Preliminary model simulations indicate that, under the condition of injecting 300°F. water for five 1-year cycles, the least-efficient operation would be injection
of hot water for 8 months and withdrawal of hot water for 4 months; both periods have the same injection-withdrawal rate of 300 gallons per minute. This operation would be only 39-percent efficient at the end of the 5-year period.
The most efficient operation that has been simulated describes injection and withdrawal of hot water for equal 6-month periods but with a withdrawal rate of 600 gallons per minute, which is twice the injection rate of 300 gallons per minute. Under these conditions, 84 percent of the stored thermal energy would be removed at the end of 5 years.
Although it may appear that the best operation of the ATES system is the one that is most efficient in terms of heat recovered, other analyses indicate that the most efficient operation may not be the most economical. The economic efficiency of the ATES system is a function of the minimum temperature requirements for use of recovered water.
Preliminary models indicate that, under conditions of injecting 300°F. water with a minimum withdrawal temperature requirement of 110°F., the maximum system efficiency at the end of five 1-year cycles would be 84 percent, based on 6-month injection and withdrawal periods with a withdrawal rate of 600 gallons per minute and an injection rate of 300 gallons per minute. If the minimum withdrawal temperature required is 140°F., the maximum system efficiency at the end of five 1-year cycles would be 61 percent based on 8 months of injection and 4 months of withdrawal, again with a withdrawal rate of 600 gallons per minute and an injection rate of 300 gallons per minute.
The model simulations are encouraging and indicate that the ATES system is a viable means for conserving a valuable resource that presently is being wasted.
Regional Ground-Water Flow in the Floridan Aquifer System
The Floridan aquifer system is one of the major sources of ground-water supplies in the United States. This highly productive aquifer system underlies all of Florida, southeastern Georgia, and small parts of adjoining Alabama and South Carolina, for a total area of about 100,000 square miles. A total of about 3 billion gallons of water per day is withdrawn from the aquifer, and, in many areas, the Floridan is the sole source of fresh water.
The Floridan aquifer provides public water supplies for many cities including Daytona Beach, Jacksonville, Orlando, Tallahassee, and St. Petersburg in Florida and Brunswick and Savannah in Georgia. The amount of water withdrawn for irrigation and industrial use is greater than that withdrawn for public supply. Although pumping of ground water has caused extensive local declines of ground-water levels, more than one-half the aquifer area has not yet had significant declines in water levels. Despite the enormous amount of untapped water in the Floridan, the water ia not always locally available for use.
This aquifer system is a sequence of hydraulically connected carbonate rocks (principally limestone with some dolomite) that vary in thickness from a featheredge where they crop out to more than 1,500 feet where the aquifer is confined. In parts of southern Georgia and northwestern and central Florida, the rocks of the Floridan outcrop are covered by a thin layer of soil. In these areas, the Floridan usually contains one aquifer. In parts of coastal Georgia, the Florida panhandle, and all of southern Florida, the Floridan aquifer rocks are covered (or confined) by hundreds of feet of clay and sand, and, in these areas, two aquifers generally are separated by rock that has low to extremely low permeability. Most of the natural recharge, flow, and discharge occurs where the rocks crop out or where the aquifer is thinly covered. The large springs of northwest-central Florida occur in the outcrop areas.
From 1978 to 1982, the U.S. Geological Survey conducted a regional assessment of the Floridan aquifer system by studying and summarizing previous studies, collecting new data in selected areas, and using computers to model and simu
late ground-water flow. The Floridan study was one of 15 comprising the Survey's Regional Aquifer Systems Analysis Program, which is commonly known as the RASA program. In the early 1930's, Victor Stringfield, a Survey geologist, published a Water-Supply Paper, Artesian Water in the Florida Peninsula. In it, he identified for the first time a regional flow system in the carbonate rocks of Florida. His potentiometric surface map of the Floridan aquifer showed the natural recharge areas (intakes to the aquifer), discharge areas (outlets), and the general direction of ground-water movement from the recharge to discharge areas. A major potential recharge area was shown in central Florida. Major discharge areas were shown in the coastal areas of Florida and Georgia.
In the early 1940's, Stringfield, H. H. Cooper, and M. A. Warren documented a widespread decline of water levels, or artesian pressure, along the Georgianortheastern Florida coast. The areal decline was caused by large ground-water withdrawals from the Floridan aquifer at the coastal cities of Savannah and Brunswick, Georgia, and Fernandina Beach and Jacksonville, Florida. Pumping, which began in the late 1800's, caused 'cones of depression' to form in the potentiometric surface around these cities. However, by the 1940's, these cones had coalesced to form a troughlike depression (in the potentiometric surface) extending from Jacksonville, Florida, to near Hilton Head, South Carolina, a distance of 150 miles.
Since this early work of Stringfield, other maps of the potentiometric surface have been prepared. However, to do a rigorous analysis of the flow system, an aquiferwide map based on simultaneous measurements throughout the four-State area was needed. To meet this need, the Survey, during a 12-day period in May 1980, measured the depth to water level, or artesian pressure, in more than 2,700 wells tapping the upper Floridan aquifer. The resulting potentiometric-surface map, as prepared by R. H. Johnston and several coworkers, is shown in the accompanying illustration. Most of the major features of the regional flow system can be demonstrated from this map; for example, con