USGS Logo Geological Survey Professional Paper 1044—C
The Waters of Hot Springs National Park, Arkansas—Their Nature and Origin



The thermal springs of the Hot Springs National Park, at Hot Springs, Ark. (fig. 1), have been a natural resource of international renown for many years. The springs were known to President Thomas Jefferson, who initiated the first scientific study in 1804. This study, by William Dunbar and George Hunter, marked the beginning of an era of scientific curiosity as to the origin and heat source of the springs.

FIGURE 1.—Location of study area.

Public interest in the hot springs has been focused primarily on the therapeutic value of the waters, and, in serving such interest, this also has been the focus of Federal management since the area was established as the Hot Springs Reservation in 1832. This emphasis did not change when direct Federal supervision was implemented in 1877, nor when the area was designated as a National Park in 1921. The purpose of the park today is to preserve and protect the hot springs for present and future generations.

Long-range planning for park uses takes into consideration, however, the prospect of a shifting of emphasis from therapeutic values of the spring waters to the scientific, esthetic, and recreational values of the park as a whole. The existence of the springs and their geologic and hydrologic setting as well as their thermal characteristics play an important role in attracting visitors to the area. The purpose of this report is to describe the hot springs with reference to their flow, temperature, and chemical quality; to present information on the geologic framework of the hot springs flow system; and to define the nature of the hydrologic and geothermal flow systems as completely as possible with the data available.


For those readers interested in using the metric system, metric equivalents of English units are given in parentheses. The English units in this report may be converted to metric units as follows:

To convert from—
(English unit)
Multiply by—
(conversion factor)
To obtain—
(metric unit)
Inches (in)25.4Millimeters (mm)
Feet (ft)3.048x10-1Meters (m)
Square miles (mi2)2.59x106Square meters (m2)
Gallons per day (gal/d)4.38x10-8Cubic meters per second (m3/s)
Feet per day (ft/d)3.53x10-6Meters per second (m/s)
Curie3.7x1010Becquerel (Bq)
Atmosphere1.013x105Pascal (Pa)

Chemical concentrations are given only in metric units—milligrams per liter (mg/L). For concentrations less than 7,000 mg/L, the numerical value is about the same as for concentrations in the English unit, parts per million.

The conversion from temperature in degrees Fahrenheit (°F) to temperature in degrees Celsius (°C) is expressed by: °C = (5/9) (°F - 32). Kelvin=degrees Celsius+273.15.


The authors are grateful for the advice, suggestions, and information given by many associates in Federal and State agencies, and for facts drawn from many published and unpublished reports. We are especially grateful to N. F. Williams, State Geologist, Arkansas Geological Commission, for his advice and interest in the study. Raymond B. Stroud, U.S. Bureau of Mines, and B. R. Haley, U.S. Geological Survey, provided advice and assistance on the geologic study. T. W. Carney, R. C. Cross, and P. A. Bailey, chemists, Arkansas Geological Commission, provided chemical analyses of rock samples. In the presentation of geothermal models, the authors especially acknowledge the assistance of M. L. Sorey of the U.S. Geological Survey. Mr. Sorey critically examined the flow models presented in the report and provided constructive criticism and technical assistance.

The authors are grateful for the cooperation and assistance of Garland Moore, Richard Ketcham, and Rock Comstock, of the National Park Service; Bernard Campbell, Superintendent, and Franklyn Hambly, Management Assistant, of the Hot Springs National Park, Ark., and Richard Rasp, Earl Adams, Lewis May, Bernie Braughton, Eugene Stringer, David Carpenter, and Alice Horner, also of the Hot Springs National Park. We also thank M. D. Bodden, J. C. Chemrys, C. T. Rightmire, and T. A. Wyerman, who provided 14C, chemical, stable isotope, and tritium analyses of the samples collected in 1972.


The history of the hot-springs area has been documented in numerous publications, many of which present detailed accounts of some aspect of the springs' environment and the cultural development of the area. For the purposes of this report, therefore, and to minimize duplication, only those historic events and developmental practices that relate to the technical management of the springs are cited.

Early descriptions of the hot springs give different accounts of as many as 72 spring openings, in a belt about one-fourth mile long and a few hundred feet wide, along the southwest slope of Hot Springs Mountain. Excavation and covering of springs, to increase and concentrate flows and to protect the springs from contamination, have so altered the natural spring environment that it bears no resemblance to the original condition. Among the early investigators, Owen (1860) reported 42 springs; Glasgow (1860), 54 springs; Haywood (1902), 46 springs; and Hamilton (1932), 48 springs. In his detailed history of Hot Springs, Scully (1966, p. 139) reported 47 active springs, including 2 exhibition springs.

Prior to 1877 some of the springs were walled in and covered by masonry arches to protect them from contamination (Scully, 1966, p. 118). By the 1890's, most of the springs were covered and a complicated piping system had evolved for supplying the bathhouses with hot water. In 1901 the springs were uncovered to give access for sampling, and chemical analyses were made by Haywood (1902). The spring enclosures were opened again in 1931 for cleaning; some of them were deepened, and the present-day (1974) collection system was constructed. The collecting system diverts the flow of 44 springs to a central reservoir, from which the water is redistributed to individual bathhouses. A map of the collection system is shown in figure 2. Since 1948 all the water delivered to the bathhouses has been metered. Excess water overflows into Hot Springs Creek when storage reservoirs are full.

FIGURE 2.—Locations of the hot springs and hot-water-collection lines. (click on image for a PDF version)

Through the years (1860 to the present), at least 20 scientific investigations, directly or indirectly involving the hot springs, have been made. Although each study generally had a separate and specific objective, many of the investigators became sufficiently interested in the hot springs to try to explain the origin of the water and the source of the heat.

The chemical quality of water from the hot springs in Arkansas has been of great interest to man, probably since the hot springs were first discovered. One of the earliest scientific approaches to determine the concentration of the hot-springs waters is found in Branner's (1892) Annual Report for 1891, in which analyses of water samples collected in 1890 are tabulated in grains per gallon. At random times since 1890, analyses have been made for investigations. The purpose of many of these investigations has been to support some theraputic claim for the water or to determine whether the chemical concentration of the water has changed.

Most earlier investigators concluded that the waters discharged from the hot springs are of meteoric origin, having fallen as precipitation and recharged to the Bigfork Chert in the anticlinal valley lying just northwest, north, and northeast of Hot Springs. Some investigators have attributed some of the recharge to the outcrop of Arkansas Novaculite to the east of the hot springs. Another theory that has been regarded by some as having a degree of scientific validity is that the water may be of juvenile origin, that is, derived from the interior of the Earth and not having previously existed as atmospheric water.

Bryan (1922, p. 426) posed the question as to the meteoric, juvenile, or mixed origin of the waters discharged from the hot springs. He indicated (p. 447-448) that the juvenile theory is perhaps more satisfactory, although it rests on an insecure foundation in postulating (1) a special igneous mass that is discharging water owing to cooling and recrystallization and (2) a special fault fissure through which the water rises to the land surface. Bryan (p. 444) analyzed the merits of both the juvenile and the meteoric theories, but conceded that "a definite conclusion as to the ultimate origin of the water in the Hot Springs cannot now be reached." He pointed out (p. 443-444) that "If the water is juvenile there is presumably a constant supply, diminishing very gradually through the centuries in quantity and temperature *** If *** the water has a meteoric origin, it is variable in quantity, fluctuating with the seasons or with groups of years having heavy or light rainfall." Thus, Bryan recognized the critical value of precise measurements of temperature, discharge, and other parameters during a sufficient period of time to provide adequate data on which to base conclusions as to the water's origin.

Arndt and Stroud (1953) suggested a dual origin for the water. Meteoric water, they believed, entered the spring system through the lower division of the Arkansas Novaculite on Hot Springs and North Mountains. They calculated that this source of meteoric water could supply about one-sixth of the flow of the springs. The rest of the water, they considered, could be juvenile water rising from depth.

Proponents of the theory of meteoric origin of the spring waters include Weed (1902), Purdue (1910), and Purdue and Miser (1923). Purdue (1910, p. 283) described the geologic conditions supporting this view and identified the collecting area as the anticlinal valley between Sugarloaf and West Mountains. Most of the valley is underlain by the Bigfork Chert, a much-fractured formation of high permeability into which infiltrates water from precipitation. According to Purdue (p. 284), the occurrence of this formation in anticlinal valleys, with its highly inclined beds, affords the most favorable condition for the intake of water. He postulates further that the water passes through the Bigfork Chert, beneath the North Mountain syncline, and is forced upward into the Hot Springs anticline to emerge as the hot springs.

This suggested movement of water from the recharge area required geologic conditions that account for passage of the water through the Polk Creek Shale, Missouri Mountain Shale, and the Arkansas Novaculite, to discharge as it does from the Hot Springs Sandstone Member of the Stanley Shale. Such conditions would ordinarily require a fault or faults, with associated jointing and fissuring, that would provide passage through these formations. Several authors have shown such a fault (Arndt and Stroud, 1953; Fellows, 1966). Recent mapping by Haley and Stone (pl. 1) confirms the presence of a complex fault system in the area but indicates no conclusive evidence of a large fault at the hot springs. In addition to the faulting, the intensive folding and overturning of formations in the vicinity of Hot Springs are attended by intensive jointing and fissuring of the sandstone, chert, slate, shale, and novaculite. It is concluded from the present study that the faults and the associated joints and fissures provide conduits for the water.

Prof. D. D. Owen, in his report of 1860 on the hot springs, said the following:

"When we reflect on the boundless and never-ceasing flow of thermal waters that must have bathed the sides of Hot Springs Ridge for countless ages *** and however inexplicable such wonderful phenomena and changes may at first appear, yet, when the chemical principles become properly understood, disclosed by the enlightened and accurate chemical analyses, these obscure geological transformations [and the origin of the water and operation of the springs] can be satisfactorily and clearly explained."

<<< Previous <<< Contents >>> Next >>>

Last Updated: 09-Mar-2009