A National Weather Service Publication In Support of The Celebration of American Weather Services
Past, Present and Future
Personal View of Professor Cleveland Abbe
Editor's Note - Cleveland Abbe was a highly respected civilian meteorologist who worked for the Signal Service, and later the Weather Bureau. Essentially, he was considered the expert on forecasting.
The Joint Resolution enacted by the Congress of the United States, February 2, 1870, and signed into law by President Grant on February 9, 1870, marks an important epoch in the history of meteorology in America. It was well known that weather systems moved from west to east, or from the southwest to northeast across the United States. Previous investigators such as Redfield, Loomis, and Espy had shown the basis on which weather predictions could be safely made and Ferrel had unraveled the mechanics of the atmosphere. About this time the electromagnetic telegraph was being used to disseminate knowledge of the coming storms and weather. Professor Henry, on behalf of eminent meteorologists, had not only explained how the telegraph could be utilized for weather predictions, but had systematically done this for many years at the Smithsonian.
The Joint Resolution, provided for taking meteorological observations at the military stations in the interior of the continent, and at other points in the States and Territories of the United States, and for giving notice on the Northern Lakes and the seacoast by magnetic telegraph and marine signals of the approach and force of storms. In January 1872, and in compliance with the appropriations bill of 1871, reports relative to the stages of water in the rivers were added, and in the following spring, and again in 1873, the floods of the Lower Mississippi were pre-announced in general bulletins, so that from this time forward that branch of work became a regulalr part of the duties of the Signal Service.
The appropriation bill, approved June 10, 1872, provided: "For expenses of storm-signals announcing the probable approach and force of storms throughout the United States for the benefit of commerce and agriculture;" and again, in the same bill provided: "That the Secretary of War be, and hereby is, authorized and required to provide in the system of observations and reports in charge of the Chief Signal Officer for such stations, reports and signals as may be found necessary for the benefit of agriculture and commercial interests." Thus, in a few years, the Signal Office came to officially include every form of meteorological observations or prediction that could affect the interests of agriculture and of our commerce on the Great Lakes, the oceans, and the rivers.
The demands of the new agency required a trained work force familiar with observational, theoretical, and operational meteorology. To educate the officers, meteorological experts from around the country provided training in the form of classes at local offices, as well as educational notes. Experts which were used included:
Professor Henry (Smithsonian Institute)
Dr. B. F. Craig (Army Medical Corps)
C. A. Schott (Coast Survey)
Admiral Thornton A. Jenkins (U. S. Navy)
Professor J. H. C. Coffin (Nautical Almanac)
Professor Loomis (Yale University)
Dr. Daniel Draper
Mr. T. B. Maury
A major task for General Myer was to educate weather observers of the Signal Service. He therefore added a school of meteorology to his school for instruction in telegraphy and military signaling, located at Fort Whipple, or Arlington, near the City of Washington. This school continued during his lifetime; after his death the name was changed to Fort Myer; it was abolished as a school of the Signal Service by order of the Secretary of War in 1886. At first the instruction at this school embraced courses in military signaling, the meteorological text-books of Loomis and Buchan, the meteorological instructions relating to the special work of the Signal Office, the building and equipment of telegraph lines, and such other duties as officers and men were liable to be called upon to perform.
In the early 1880s, a more formal course of meteorological instruction was designed to supplement the more informal training started in 1870. The instruction began in 1881 when four lieutenants of the Signal Corps began a course of Deschanel's Physics and a wide range of meteorological literature; it was subsequently enlarged from time to time until, in 1885-86, a class of six officers attended an extensive course of lectures and instructions supplemented by monthly examinations on the following subjects, i.e., the theory of instruments, chartography, general meteorology, thermodynamics of the atmosphere (taught by Abbe); theoretical meteorology without mathematics (Ferrel); practical meteorology and weather predictions; topographic surveying and drawing (Dunwoody); electricity and laboratory manipulation (Mendenhall).
In 1886, the school of instruction for our observers in their duties was abolished, notwithstanding General Hazen's remonstrance, and its work was relegated to the sergeants at the respective stations, where it was consisted mainly in the study of Loomis' text-book and of the volume of meteorological instructions and the acquisitions of good habits as observers and telegraphers.
While attempting to build up a strictly military organization, General Myer had a clear appreciation of the uses that he could make of civilian employees. Having been intimately associated with Paine and Lapham, of Milwaukee, in securing the legislation that authorized the Weather Service, the Chief first secured Professor Lapham as his civilian assistant. After Professor Lapham had declined a permanent appointment, on account of his health, I was called to what I then supposed would be a temporary engagement and in June 1871, Professor T. B. Maury accepted a similar position; the electrician, or telegrapher, Mr. G. W. Maynard, was also a civilian appointee. It was all the more necessary to secure a few civilian employees in view of the fact that the officers of the Army, temporarily detailed to Signal Service duty, were liable at any time to be ordered back to their regiments. Up to the middle of 1872 the duty of weather predictions and storm warnings devolved upon the civilians; during the remaining years of General Myer's administration it was equally divided between them and the detailed officers, Lieutenants Craig, Dunwoody, Story, Kilbourne, and Greely. During the administration of General Hazen those who had become second lieutenants in the Army by promotion from the corps of observers in the Signal Service, and especially Lieutenants Powell and Glassford, also became "indications officers," while the special work of predicting tornadoes was assigned to Lieutenant Finley.
After the appointments of Upton, Waldo, Hazen, Russell, Sawyer, Marvin, and especially to Professor Ferrel on August 10, 1882, and Professor T. C. Mendenhall on January 1, 1885, it came to be recognized that there were multifarious fundamental labors appropriate to the civilian experts besides the making of weather predictions, which, as a purely empirical matter, had already been brought to a satisfactory degree of perfection. General Hazen's policy of introducing into both the military and the civilian ranks as high a grade of intellectual attainment as was practicable accomplished much for the scientific reputation of the Signal Service and enabled it to accomplish far more for meteorology than could otherwise have been done.
True science is never speculative; it employs hypotheses as suggesting points for inquiry, but it never adopts the hypotheses as though they were demonstrated propositions. There should be no mystery in our use of the word science; it means knowledge, not theory nor speculation; nor hypothesis, but hard facts, and the framework of laws to which they belong; the observed phenomena of meteorology and the well-established laws of physics are the two extremes of the science of meteorology between which we trace the connection of cause and effect; insofar as we can do this successfully, meteorology becomes an exact deductive science.
That relation between the Signal Service and the science of meteorology, which is of fundamental importance, consists in the character of the observations made at the stations. In reference to this branch of our subject, I need only to say that the whole time of the regular observers was given to the work of the Signal Service, and there was a multitude of duties to engross their attention. The fact that the Signal Service employed hundreds of men whose lives were concentrated upon the maintenance of a complete record of the weather demonstrates the thoroughness of its equipment for this work. It was indeed impracticable to maintain hourly observations, but for many years four simultaneous or telegraphic and three local time or climatic observations were made besides the records of the maximum and minimum thermometers and the continuous record of the wind.
With regard to the uniform character and general accuracy of these observations, there can be no doubt that the system of instructions and inspection, and especially the intercomparison of the telegraphic reports that attended the study of the tri-daily weather map, served to prevent and detect any appreciable variation from the desired standard of accuracy; very few instances are on record in which an observer's observations proved to be so unreliable that they had to be rejected. In 1891, about two-thirds of the observers had been over five years in the Service and about one-third over ten years.
With regard to the locations of the stations it must be acknowledged that they were not selected for climatic purposes, but almost wholly with a view to dynamic meteorology and the publication of storm warnings and weather predictions. The Service needed to know the pressure, temperature, and the rainfall approximately, but it needed to know the winds, and that too the strongest winds, quite accurately. The first great problem of dynamic meteorology is to know the local and the general motions of the air, and in fact, climatology may also be said to depend upon the same knowledge.
In order to bring about intimate and direct relations with the national business interests, General Myer requested the respective cities to organize committees which should feel themselves, in some sense, responsible to him as the representatives of popular interests and needs. From the chairman of each committee there was usually received an annual or semiannual report, together with many intermediate letters; the committees were kept duly apprised by our observers of the progress of the Service at Washington, and, on the other hand, it gave General Myer timely notice of the character of the work done by the local weather observer. The duties of these committees included both praise and criticism, and occasionally, some excellent suggestions were received from them; their very existence always demonstrated the desire of the Government to labor in the interests of the people.
I must stop a minute to call attention to that feature of the work of the Service which has enabled it to get up its daily weather maps with a celerity and regularity that have always been the wonder and admiration of those accustomed to ordinary commercial telegraphy. Notwithstanding the numerous telegraph wires that connect our principal cities, yet it ordinarily happens that individual dispatches must take their turn, and thereby suffer a delay that may amount to many hours. But General Myer saw plainly that this would never do for the work he had in mind; his experience during the war (Civil War) had accustomed him to attain the utmost possible dispatch, and he demanded this also in his new application of military signaling to the commercial needs of the country. It required much argument to induce the telegraph companies to accept the scheme that he proposed; they entered into it, at first, only on agreement that after a few months' trial it might be modified, and, in fact, such was the friction between the conflicting interests that on the 4th of March, 1871, all telegraphic dispatches were suddenly refused by the Western Union Telegraph Company, and for several days I made weather predictions based on such few reports as we could obtain from our stations through rival telegraph companies.
The same trouble occurred in the following year, but eventually General Myer's circuit system triumphed.
By this simple arrangement the observers deliver their short cipher dispatches to the respective telegraph observers at prearranged minutes; all the men on a specific line of wire, or "circuit," are at hand simultaneously, and any dispatch put on that circuit wire is received simultaneously by all the observers. As soon as any one dispatch of a few cipher words is telegraphed another succeeds it, and thus in a minute a number of stations have interchanged their reports so far as that circuit is concerned. The next minute another circuit, joining on to the preceding one, is opened, and its own, together with all the accumulated reports, are interchanged. In this manner it is found to require only from twenty minutes to a half hour to interchange reports between all the important telegraph centers of the United States as they converge toward Washington, so that in an hour after the observations are made the observers at all the larger cities begin the construction of weather charts similar to the standard chart that is published at Washington.
Cautionary Signals for Wind and Weather
The synopses and probabilities (forecasts) that were furnished to the daily press of the country reached the public eye after the lapse of considerable time, nor was it at all certain that they would in any way reach the mariners for whom the service was designed. It was, therefore, necessary to supplement these by a system of visible signals that could be hoisted immediately by telegraphic orders from Washington, and this system of cautionary storm-wind signals was instituted in the summer of 1871, as soon as it was demonstrated that our knowledge of the movements of the storms justified taking that step. The display of the square red flag with the black center, or the cautionary danger signal, mared a passage from the general weather probabilities to the definite special prediction of a specific velocity of wind within a specific time and a small region. The region was defined as within a radius of 100 miles, and the time limit was eight hours. Subsequently, a modification of the signal was introduced showing the general direction of the expected winds, and again, another modification showing whether the winds would be above or below a certain velocity.
A decided advance in our methods of communicating with the public was made early in General Hazen's administration by the adoption of a special signal for the so-called "cold wave." In many localities also special signal devices began to be used by the people -- flags and balls, steam whistles and bells, and in Ohio the so-called railroad weather signal had been devised. By means of a few flags, white, blue, and black, the probable local weather for the next day is indicated in every town and almost every telegraph and telephone station in the country, so that any one may know what to expect and prepare for.
River Floods and Their Predictions
Telegraphic reports of the condition of the rivers began to be received by the Signal Service January 1, 1872. This work was so natural and desirable and so easy an expansion of the work originally authorized that there could be no doubt of its propriety. At first it seemed sufficient to publish the reports of stages of water as received at the office, but soon some general indications of probable rising and falling water began to be added to the weather probabilities. The gauge readings, above which a stage of water was considered to be dangerous, as well as the times required for flood waves to descend along the channels of the rivers, as first adopted approximately by me, were revised in a report by General Greely in 1874.
One of the matters most clearly foreseen, and, in fact, often urged by me upon General Myer, was the propriety and the necessity of stationing experienced officers at such centers as Chicago, St. Louis, New Orleans, and San Francisco, for the purpose of making forecasts for those more remote sections of the country, as well as looking after the general interests of the people and the Service. A beginning of this kind was eventually made by establishing Lieutenant Woodruff at St. Paul and Lieutenants Maxfield and Finley at San Francisco. But the original argument was to the effect that at every city where the map was published daily a local forecaster ought to be able to do better or, at least, as well as the Central Office in Washington, and that on many accounts it was best for the Service to develop a large board of forecasters rather than to confine the work to a few military officers in Washington. The force of this argument was finally felt, and the preparations for the work of the local forecasters was already being made by the Signal Service when transferred (to the Department of Agriculture) in 1891.