HORTON, ROBERT ELMER

(b. Parma, Michigan, 18 May 1875; d. Voorheesville, New York, 22 April 1945),

hydrology, quantitative geomorphology, meteorology, hydraulic engineering.

Horton is sometimes called the father of American hydrology. His investigations led to important new insights about the relationship between the physical features of drainage basins and water flow on and beneath the surface. He is best known for his theory that relates the infiltration of soils to the generation of floods by surface runoff. Among the first American hydrologists to stress interdisciplinary skills, he emphasized the influence of soils and vegetation on runoff, and his significant contributions to the mechanics of soil erosion contributed to the development of soil conservation techniques. His career culminated in a major landmark paper (1945) in which he analyzed the relationship between the mechanics of slope and channel erosion and runoff processes. The paper’s emphasis on quantitative geomorphology for understanding hydrological processes helped turned hydrology from a largely descriptive science into one at once both more theoretical and rational.

Early Development . Horton was the son of Van Rensselaer and Rowena (Rafter) Horton. In 1897 he graduated with a BS degree from Albion College, a small Michigan liberal arts school close to his home. Horton extended by about a year the customary four years for undergraduate work in order to help his father with the family farm and, possibly, to earn money to cover college costs. While in college, Horton devoted the turbulent political summer of 1896 to supporting the Republican presidential candidate William McKinley after concluding that Democratic presidential candidate William Jennings Bryan’s pro-silver platform would tighten the money supply and worsen economic conditions. He continued his support of the Republican Party for the rest of his life.

The engineering education Horton received at Albion was rather limited. His scientific courses included geology, physics, astronomy, chemistry, mathematics, and possibly some botany. The mathematics courses comprised algebra, plane and solid geometry, trigonometry, and elementary calculus. Probably these courses did not challenge Horton too much, since he had earlier studied mathematics at home. One story has it that he studied trigonometry while helping his brother haul manure. By working hard and fast, he could fill the wagon faster than his brother could haul the manure away. While waiting for his brother’s return, he studied mathematics. At Albion, Horton easily passed an examination that gave him class credit for trigonometry. His record in grammar was equally impressive. After spending two weeks in a grammar class, he was given credit and excused from further attendance. Horton’s writing ability largely resulted from the influence of his father, who insisted that his children learn the foundations of good writing. In 1932 Albion College awarded Horton an honorary doctorate.

Horton’s uncle on his mother’s side was George W. Rafter, a well-known hydrologist who directed his nephew toward the same profession. In 1896, a year before Horton graduated from college, he and Rafter coauthored a paper for the New York State Engineer that analyzed rainfall and runoff in the Upper Hudson River Drainage Basin. The authors used probability theory, an approach then in its infancy among hydrologists. Specifically, they applied new ideas about the importance of frequency to determine probable maximum flood stages. Later, Horton worked for his uncle as an assistant engineer for the Board of Engineers on Deep Waterways, which was authorized by Congress to study the possibility of a ship canal from the Great Lakes to the Atlantic seaboard. Horton’s own ambition was undeniable. It is reflected in a notebook found in his papers called “Hydraulic Tables and Formulae” (1898). Begun when Horton was still in Michigan, the notebook contains (among some forgettable poetic verse) annotations concerning the formulas of numerous engineers for calculating everything from discharge to coefficients of roughness to mean velocity.

Rafter was a nervous man, caustic and quick to criticize, and Horton occasionally found it difficult to work with him. Yet Rafter effectively opened doors for his nephew. While working on plans for constructing the Great Lakes–Atlantic Ocean ship canal, Rafter had commissioned the hydraulic laboratory at Cornell University, his alma mater, to perform a number of weir studies to identify discharge coefficients for dams of various cross sections. In 1899, he obtained a position for Horton at the laboratory working on the project under the supervision of Professor Gardner Williams, the laboratory’s director. He and Horton did not get along. They argued over methodology, and Williams blamed Horton unjustly for some inaccurate readings (the problem lay with the equipment, rather than with Horton) and tried to fire him. Horton appealed to his uncle, who wrote Williams. Subsequently, Williams appeared resigned to Horton’s presence.

In the fall of 1899 Horton obtained a per diem appointment with the U.S. Geological Survey (USGS). Again, Rafter’s hand is evident. The survey, at Rafter’s urging, had recently taken over the gauging stations established by the Deep Waterways Board. A year later, Horton became the survey’s resident hydrographer in New York State and obtained full-time civil service status within a few months. With his office established in Utica, New York, Horton oversaw the establishment of gauging stations throughout New York State and Michigan, when that state was added to his district in 1902. His staff included two assistant engineers and from three to five field assistants, including three young college professors. Winter conditions posed special problems. Horton employed “coach candles” to warm current meters during the winter time before lowering them into icy water, where, without being warmed first, they would have frozen. Horton developed his own snow sampler and was among the first to keep a continuous record of snow depth and water equivalent, taking weekly readings for two winters in Utica, from 1903 to 1905. During this time as hydrographer, Horton began to appreciate the significant contribution of groundwater to runoff and began wondering about the process by which surface water reached the aquifer.

Horton’s Accomplishments . In May and June 1903, Horton headed a team of researchers at Cornell that performed a variety of experiments on weirs similar to those at USGS gauging stations. This time his dealings with Professor Williams were evidently more cordial. The conclusions were published by the USGS in 1906 as Water-Supply and Irrigation Paper 150. They were revised the following year in Water-Supply and Irrigation Paper 200, which became a standard reference on the subject of weirs and streamflow.

Horton married Ella H. Young on 19 June 1901. Though the couple was (and remained) childless and a government salary might have satisfied their needs, inadequate government funding for the survey’s field offices contributed to Horton’s decision to seek new opportunities in the private sector. In 1906 he resigned from the survey and established an office as consulting hydraulic engineer in Albany, New York. He advertised himself as a “Specialist in Hydrology and Critical Reports on Adequacy and Safety of Water Supply and Structures for Power, Public Use and Irrigation.” Horton also contributed articles to various engineering journals, including the Transactions of the American Society of Civil Engineers and Engineering News Record. The articles reflected his broad engineering interest, covering such topics as flood frequency, snow evaporation, and deforestation. His numerous publications following World War I also reflect his increasing interest in meteorology. Just between the years 1919 and 1923, Horton published articles in the Monthly Weather Review on such diverse topics as rainfall interception, evaporation observations, areal rainfall estimates, transpiration by forest trees, and rainfall interpolation. In 1918, he moved his residence to Voorheesville, New York, just outside of Albany, where he established a private hydrological (as opposed to hydraulic) laboratory on an 80-acre site complete with an old mill, a stream, and waterfalls.

For a number of years, Horton developed his ideas about infiltration capacity, a term he introduced in 1924 to describe the maximum rate at which a given soil surface could absorb falling rain over a specified period of time. In succeeding years, he refined his idea through a number of investigations. He published a major study in 1933 titled “The Role of Infiltration in the Hydrologic Cycle.” In this article Horton argued that precipitation reaching the ground divides into either overland flow or infiltrated water. The former becomes storm runoff once it enters a stream. The latter, with net losses to transpiration and evaporation, percolates down to become groundwater and then may slowly seep downhill and eventually enter a stream. Overland flow increases when rainfall intensities or snowmelt rates exceed the infiltration rate. After initial detention in uneven surface areas, it commences to flow uniformly throughout the whole basin, gradually increasing in depth as the water flows to lower elevations. This became known as the theory of infiltration-excess overland flow. One hydrologist, J. A. A. Jones, called this article “the first scientific study of storm runoff” (1997, p. 78).

Subsequent studies showed that Horton’s ideas of infiltration and overland flow worked better in semiarid than in humid regions, and that the entire process is more complicated than Horton described. Nevertheless, Horton’s contribution provided fundamental insights. In particular, his work showed the critical roles soil type and vegetation had on runoff and soil erosion. One might reasonably argue that showing these impacts and providing a functional vocabulary, thereby increasing the hydrologist’s explanatory power, were significant contributions in their own right. The work helped in later computer modeling and provided significant rationale for the acceptance of the unit hydrograph, a method devised by Leroy K. Sherman in 1932 to measure the average shape of storm hydrographs. Unlike earlier hydrographs, which simply plotted stream discharge against time, the unit hydro-graph enabled engineers to combine daily precipitation figures with rainfall and runoff data from a 24-hour hydrograph in order to calculate runoff and streamflow for the same or (with somewhat less accuracy) similar watersheds. Sherman did this by calculating the average runoff pattern for a specified unit of rainfall, usually 1 inch or 1 centimeter. Still, without detailed chemical and physical analysis, Horton could not describe in any great detail how the process of runoff actually worked. To do so required more data, more accurate instruments, and numerous techniques. Rather, Horton emphasized what he called hydrophysical relationships; today, it would be called it quantitative geomorphology.

Horton worked on rainfall and runoff throughout his life. The gestation of his ideas from unpublished papers through increasingly sophisticated articles typified Horton’s approach on many subjects. His seminal papers reflect years and even decades of thought and investigation. He was often the first to recognize the shortcomings of his own work and was very cautious in advancing his ideas. Only in 1939 did he commit to print a predictive equation for infiltration capacity. More empirical than the work of Henri Darcy and his followers, the equation obtained limited acceptance.

Horton’s work on the impact of hydrophysical processes on drainage flow, flood discharge, and other factors culminated in 1945 in a paper that provided new data and theoretical insights, but summarized, with some modifications, many of his earlier ideas. It was Horton’s final and most important contribution. In the paper, Horton quantitatively suggested four new laws or principles: the law of steam lengths, the law of stream numbers, limits on infiltration capacity, and the relationship between runoff and detention of water. Horton reversed the European stream order and designated the “fingertip” streams as order one and assigned the main stem stream the highest numerical order. He then used this classification system to compare and contrast the physical features of streams of different orders. With some modification, this approach was commonly used in American hydrology into the mid-1980s. Horton also expanded upon his idea of drainage density, which he had first suggested over a decade before. The term could be defined mathematically:

Drainage density,

where ∑L is the total length of streams and A is the drainage basin area, both in units of the same system. Horton’s law of stream lengths suggested that “The average lengths of streams of each of the different orders in a drainage basin tend closely to approximate a direct geometric series in which the first term is the average length of streams of the 1st order” (1945, p. 291). In other words, there is a regular relationship between stream lengths of one order and those of other orders of a given stream. Horton demonstrated a constant relationship between the number of streams of different orders within a basin.

Horton’s article stimulated a reevaluation of historical and qualitative geomorphology that could be traced to Grove Karl Gilbert and William Morris Davis. Geomorphic typologies that depended on descriptions such as youth, maturity, and old age came under fire as inexact and unscientific. Horton’s classification system has been superseded by ecological concepts of energy expenditure and efficiency to understand stream basin organization. Still, Horton deserves credit for helping to put hydrology on a more scientific footing.

Horton’s Personality and Professional Involvement . There is little in Horton’s papers that reveal much about family or social life. He joined the Cosmos Club in Washington, D.C., in 1934 and remained a member until his death, but even here the record is ambiguous; the Cosmos Club had always attracted a large number of geographers, geologists, and hydrologists (it was founded by John Wesley Powell in 1878), and one can imagine Horton joining colleagues from various federal agencies for lunchtime discussions of professional issues.

In any such discussions, Horton would have adopted a practical and rather hardheaded approach. He had no time for unscientific ruminations. He observed in 1931 that, even though hydrologists did not know the exact maximum rainfall that might fall on Washington, D.C., meteorological research persuasively demonstrated that it could never reach 1,000 inches a month, regardless of what frequency analysis might show. As he put it, “Rock Creek [a small creek in Washington] cannot produce a Mississippi River flood—any more than a barnyard fowl can lay an ostrich egg, and for very much the same reason, namely, it would transcend nature’s capabilities under the circumstances” (p. 201). In short, theoretical arguments that fly in the face of empirical evidence must be rejected.

Nor did Horton have much tolerance for moral ambiguities, especially as it applied to his own work. According to Howard L. Cook, one of Horton’s protégés, in 1907 Horton complained that John C. Hoyt and Nathan C. Grover, two well-known Geological Survey employees, appropriated without attribution some of the tables Horton had published in Geological Survey Paper 200 in their book River Discharge. Arthur Powell Davis, a senior engineer in the Survey, came to the defense of Hoyt and Grover, insisting that these sorts of engineering tables were common property and their use in engineering texts did not constitute plagiarism. The answer did nothing to assuage Horton’s outrage. Horton also considered himself a guardian of hydrologic terminology, especially of terms he coined. In 1942, when the Hydrology Committee of the American Society of Agricultural Engineers initiated a project to standardize hydrologic terms, Horton replied with a withering criticism of those who had misused or wrongly defined his term infiltration capacity.

Horton contributed to many professional associations. In the late 1920s, he campaigned with other prominent hydrologists for the creation of a hydrology section within the American Geophysical Union (AGU), an effort that needed to overcome some anxiety within the scientific community over whether hydrology was true science or not. In May 1930, the AGU amended its bylaws to create the new section. In the organizational meeting in November, section members elected Horton the vice chairman and Oscar Meinzer, chief of the USGS Division of Ground Water, the chair. In subsequent years, Horton was a major contributor to the AGU Transactions on hydrologic issues. He also served as president of the American Meteorological Society in 1939, consultant to the National Resources Commission (1934–1937), and consultant to the Soil Conservation Service (1939–1941). At various times, he advised state and local agencies in New York State.

Horton died of a heart attack on 22 April 1945. He left the bulk of his estate and most of his papers to the AGU. In honor of Horton’s contributions, the AGU established the Robert E. Horton Medal, which recognizes outstanding contributions to the geophysical aspects of hydrology. The American Meteorological Society also honored him with the establishment of the Horton Lectureship.

BIBLIOGRAPHY

WORKS BY HORTON

Papers. Record Group 189, 94 Boxes. National Archives and Records Administration, College Park, MD. Includes correspondence, and drafts of reports, articles, and studies.

Weir Experiments, Coefficients and Formulas. U.S. Geological Survey Water-Supply and Irrigation Paper 150. Issued as U.S. Congress. House. 59th Cong., 1st sess. Washington, DC: U.S. Government Printing Office, 1906. H. Doc. 231. Revised and republished as Weir Experiments, Coefficients and Formulas. U.S. Geological Survey Water-Supply and Irrigation Paper 200. U.S. Congress. House. 59th Cong., 2nd sess. Washington, DC: U.S. Government Printing Office, 1907. H. Doc. 794.

“The Field, Scope, and Status of the Science of Hydrology.” Transactions of the American Geophysical Union 12 (1931): 189–202.

“The Role of Infiltration in the Hydrologic Cycle.” Transactions of the American Geophysical Union 14 (1933): 446–460.

“Erosional Development of Streams and Their Drainage Basins: Hydrophysical Approach to Quantitative Morphology.” Bulletin of the Geological Society of America 56 (1945): 275–370.

OTHER SOURCE

Bevan, Keith. “Robert E. Horton’s Perceptual Model of Infiltration Processes.” Hydrological Processes 18 (2004): 3447–3460.

Bras, Rafael L. “A Brief History of Hydrology.” Bulletin of the American Meteorological Society 80 (1999): 1151–1164. Robert E. Horton Lecture.

Cook, Howard L., Papers. Folder 682. Research Collections. Office of History, Headquarters, U.S. Army Corps of Engineers, Alexandria, VA. Contains Cook’s handwritten notes pertaining to Horton’s childhood and early career.

Hall, Francis R. “Contributions of Robert E. Horton.” In History of Geophysics, vol. 3, The History of Hydrology, edited by Edward R. Landa and Simon Ince. Washington, DC: American Geophysical Union, 1987.

Jones, J. A. A. Global Hydrology: Processes, Resources and Environmental Management. Harlow, U.K.: Longman, 1997.

National Research Council, Water Science and Technology Board. Opportunities in the Hydrologic Sciences. Washington, DC: National Academy Press, 1991.

Paynter, Henry M. “Robert E. Horton (1875–1945).” American Geophysical Union. Available from http://www.agu.org/inside/awards/horton2.html.

Martin Reuss