History of Modern Weather Forecasting
SUNDAY JANUARY 24, 2010
Until the barometer was invented, (with Galileo’s help) by Torricelli in 1643, the forecasting of weather was considered to be part of natural astrology, which encompassed a geocosmic relationships between celestial phenomena and the natural environment of the earth. It was considered a unified, reputable body of knowledge. It formed an important part of natural philosophy. The study of natural astrology reflected the view of the universe that everything is connected. All earthly forms were seen as bound to the power of the Sun, Moon, stars, planets, and other phenomena of celestial or atmospheric origin, such as meteor showers, comets, aurora borealis, lightning, and precipitation.
In the late eighteenth and early nineteenth centuries weather records were kept however the recorder wished, as weather recording was still an amateur hobby. John Dalton began keeping daily meteorological records in 1787. Dalton constructed his own rain gauges, barometers, thermometers and hygrometers. In 1793 he published tables of rainfall, barometric pressure, temperature, wind and humidity in his Meteorological Observations. He continued to record his observations until the morning of the day he died in 1844.
Meteorological phenomena were still little understood in the early nineteenth century. One problem was that although instruments were now available to measure atmospheric change, and with the advent of the Fahrenheit system temperature measurements could be recorded in a single system, there was no quick way of transferring weather data from one location to the next. Often the weather being warned about would arrive before the data. When Samuel Morse invented the electric telegraph it became possible for weather reports to be circulated widely and speedily. The age of modern weather forecasting had begun.
Weather forecasting in the nineteenth century was dealt with by a strange combination of scientists and hobbyists. As weather predictions from every source sometimes failed, no single approach established its authority through a string of successful predictions. A wide range of weather forecasts had plausibility. Forecasting was played out in the public arena. Anyone could supply themselves with data from which to compare predicted and actual weather.
Government meteorology was originally established to collect observations. Its competitors usually based their forecasts on theories of planetary or lunar influences on the Earth's atmosphere. Some of the amateurs concerned would later be redefined as scientists because of their influence.
One such example is Luke Howard, (1772-1864), the man credited with developing the basis for our cloud classification system, a manufacturing chemist and pharmacist. Howard produced several works including On the Modification of Clouds, The Climate of London, and Seven Lectures on Meteorology. Howard became a devoted observer of the atmosphere for the rest of his life, augmenting his visual observations with readings from barometer and thermometer. For over thirty years, Howard maintained a record of accurate meteorological observations. His publications of the 1840s showed his interest in a lunar influence on the weather. In Barometrographica he concluded that the fluctuations of the barometer showed a periodicity, which could be linked speculatively to the Moon's gravitational effect on the atmosphere and he analysed an eighteen-year period of the Moon's relative position to the Sun and Earth, which influenced the weather.
In 1854, the British government created the Meteorological Department on the recommendation of the Royal Society. The first head of the department was Capt. Robert Fitzroy, who was expected to collect data from ships and, because of the recent invention of the telegraph, create weather charts. From these charts, Fitzroy issued some of the first forecasts.
Fitzroy's mandate was to compile statistics on wind to assist the efficiency of navigation. He enlarged on that role promoting weather observations, establishing barometer stations, telegraphic reporting and, in 1861, the first storm warnings. These he soon extended to routine weather forecasts. In 1862 he published The Weather Book.
Fitzroy's techniques used were not up to the standard needed to establish credibility. He was criticized in Parliament, in newspapers and by other scientists. His role in assisting Darwin also greatly troubled him, and he oscillated between being a firm “lunarist” and denying the moon played a part in weather, which may have come from his strong religious faith and the knowledge that the moon was the pre-Christian symbol of paganism. On April 30th 1865 at his home at Upper Norwood, outside London., he fatally slit his own wrists. The inquest attributed his action to overwork.
Soon after, a special committee was formed to investigate the Meteorological Department and concluded: "The first step toward placing the predication of storms on a clear and accurate basis was taken by Fitzroy. He compelled scientists and society in general to take note of this branch of meteorology, and he gave his life in the effort." The Meteorological Office was closed, to be reopened in December 1867 in response to public demand, and forecasts were resumed.
Between the stools of astro-meteorologists and scientists fell the lunarists, those who predicted according to the cycles of the Moon. These were not accepted by either side, but each claimed that lunarists belonged to the other.
In addition to Howard, mentioned above, one such was Stephen Martin Saxby, born in 1804 in Kent. He gained employment as a naval engineering instructor, published several books and the spherograph, for navigation on a sphere. He is best known for weather prediction using the moon. Saxby was careful to make a distinction between his predictions based on the orbit of the Moon and astrology, which he derided. The distinction was not recognized by many meteorologists who were struggling to establish their own credibility in their predictive science. In 1868 Saxby predicted a major storm - and published the forecast in the London Standard. There was no mention of a specific location in the warning.
The particular storm that coincided in time with Saxby's prediction was a hurricane that made landfall in the area of the Maine/New Brunswick border. Strong winds and high tides caused extensive property damage and loss of life in New Brunswick and surrounding areas. Saxby ended his letter with a promise that come September - with the newspaper's permission - he would send along a brief reminder of his prediction, for the safety of all mariners. In the meantime, he wrote that the weeks and months leading up to the impending disturbance should provide ample time for the repair of unsafe sea walls. He concluded with, "I am quite aware that in taking this step I am allowing a sense of social duty to outweigh personal considerations; but I accept the consequences ..."
Frederick Allison was living in Nova Scotia when he heard of Saxby's prediction and writing a monthly column for the Halifax Evening Express on the weather of the past month. The October 1st edition of the Evening Express had a letter to the editor written by Allison providing a more explicit warning than Saxby's.
Little did Allison realise that, as his prediction of a major gale rapidly spread across the Canadian Maritimes, a powerful tropical hurricane was working its way northwestward through the western Atlantic waters. As this storm passed over the colder waters of the North Atlantic, it was transformed from a hurricane into a deep, extra-tropical depression. Its centre made landfall on October 4th and 5th in the area of the Maine/New Brunswick border. Its strong winds and excessive rainfall created one of the worst natural disasters for this part of the world. It was responsible for the death of about one hundred people. For much of New England, rainfall records were set that still stand today. The storm was also responsible for the world's highest known tidal excursion, at the head of the Bay of Fundy, flooding salt marshes. This storm went down in the weather annals as The Great Storm, The Saxby Flood or The Saxby Gale.
Astrology had peaked during the seventeenth century with the career of William Lilly. During the Restoration it fell into a gradual decline. Its rejection by religious and philosophical authorities forced astrology into two directions - one in which cosmological and medicinal elements were assimilated into respectable areas of though while the second into popular almanacs in the form of horoscopes. The split between natural and judicial astrology as defined by Ptolemy was wider than ever. Astrology had gone through a rebirth in the early nineteenth century following the work of Robert Cross Smith (Raphael). Raphael's' and Zadkiel's work sold hugely and new books were flourishing.
The work of John Goad was broadly followed by astro-meteorologists. It was accepted that the Earth's atmosphere was vulnerable to planetary influence, that its position relative to other planets was the key to its weather, that a mathematical analysis of the angular distance between the planets coupled with observation would expose causal relationships and that this information would be valuable for farmers and sailors. Both astro-meteorologists and lunarists were concerned with long term prediction. The 48-hour forecasts offered by the Met. Office seemed inadequate by comparison.
Lunarists relied upon an analogy with the tides. If the Moon exerted a gravitational effect on the oceans, it could do the same with the atmosphere. The mechanism varied. Some thought that the Moon disturbed the electric tension of the atmosphere, others looked towards gravitational forces put in motion by the proximity of the Moon. Scientists saw lunarism as a form of astrology. Astro-meteorologists argued a more mathematical science calculating intensity and types of influence depending on the specific combinations of planets and positions. Like many lunarists they sought the cause of these effects in electrical, magnetic or gravitational forces. A prestigious name associated with lunar meteorological theories was that of Sir William Herschel. His son, John Herschel, found it difficult to shake the public's belief that he too was a lunarist. However, he acknowledged that too little was known about the emerging science to make categorical statements about methodology. "Meteorology, so far as prediction of the weather is concerned (which most people consider, very erroneously, to be its only practical object) may be regarded as a science still in its infancy" and did not dismiss the effect of the Moon totally.
Fitzroy attached great importance to "lunisolar" effects arguing that the Moon and Sun exerted a gravitational pull on the atmosphere, generally undetectable as it simply reinforced the wind patterns first detailed by Halley, that of warmed air rising at the equator and sinking at the poles. He argued that the combined effect of Sun and Moon was significant enough to cause disturbances at the equinoxes. While he denied that he was a lunarist it is easy to see why others would jump to that conclusion. Astro-meteorologists acknowledged the effects of the Moon but believed them to be minor in comparison with planetary influences.
Astro-meteorology shared forms and audiences with phrenology, mesmerists, scientific demonstrations, and instrumental exhibitions in the Victorian market for scientific amusement. However, astrology was illegal and its practitioners in danger of persecution and imprisonment. The astro-meteorologist William Joseph Simmonite refused personal interviews following a colleague's prosecution and carried out his practice solely by correspondence to avoid entrapment. Simmonite, in his Scientific and Literary Messenger, (devoted to astrometeorology, astronomy, predictive astronomy, astrology, geology, botany, chemistry and the physical sciences), was fairly traditional in his approach. He described the influence of the Moon on the atmosphere, stating that the atmosphere changed when the Moon reached the Sun's declination. He also listed the angular positions of planets by zodiacal sign.
Astro-meteorology had an immense audience, which it reached through inexpensive almanacs. Moore's Almanac had a circulation in the hundreds of thousands. Zadkiel's Almanac, which flourished until the end of the century, sold from 22,000 to 70,000 a year. There were dozens of others, some of which specialised in weather predictions.
In 1837 Patrick Murphy issued a volume called The Weather Almanack (on Scientific Principles, showing the State of the Weather for every day of the year 1838). Almost immediately it became a best seller and ran through forty-five editions. The success he achieved though is based on his forecast for one day, January 20th. Under this date he wrote 'Fair, probably lowest degree of winter temperature'. January 20th turned out to be a remarkably cold day and at sunrise the temperature was four degrees below zero. At Walton-on-Thames a reading of fourteen degrees below zero was reached. This event made a celebrity out of Murphy, and almost brought about a catastrophe to his publishers. So many people rushed there to buy his book that they almost wrecked the place. It wasn't long before that winter became known as 'Murphy's Winter'. His fame was such that a one act play Murphy's Weather Almanac was performed at Sadlers Wells in 1838, partly based on his story. Murphy tried again to give a daily forecast for the whole year, though he never repeated this success. Respectable competitors emerged; for example, Robert Scott of the Met endorsed The Seaman's Almanac Office in 1869 when in its 35th year. Weekly journals also provided a venue for astro-meteorology. The English Mechanic published exclusively on astro-meteorological theories.
Some astro-meteorologists claimed that astrology and planetary weather science were unconnected, going so far as to call the latter "astrono-meteorology" to stress the connection to astronomy rather than astrology. "They have chosen an Astronomical title because they consider that all great meteoric phenomena are the result of various electric currents, produced in this globe, from other orbs composing the Copernican system, as the earth makes her orbital revolution among them." The Journal of Astronomic Meteorology for one claimed that is was different to other publications as astro-meteorology was based on the fact that in Ptolemaic astronomy, planets in their progress around the earth formed a constant association of angular positions or aspects with each other. This publication also claimed that astro-meteorology in the past had failed as a system because of the lack of knowledge of earlier ages in the areas of astronomy and philosophy. They could be scathing of the work of others "Has this scene of grandeur and magnificence been called into being merely to amuse a few astronomers - to gratify the fancies of a few star-gazers?" This publication also sees the assertion of not being adept at astrology. Typical of its type, the Journal covered a lot of ground and devoted much of its space to commenting on other publications. It reminded readers that atmospheric conditions being analysed allowed the forecasting of disease and epidemics, carried obituaries, and discussed how forecasts could be derived from plants animals and inanimate objects, for example, watching how pigeons washed before rain or how a leech lies. White in particular had a fondness for comets. Other publications reminded readers of such facts as Jupiter having a great influence on the death rate.
George MacKenzie's Manual of Weather devoted space to discussing the principle of cycles. Although nowadays this is one of the most acceptable of early theories in many quarters, at this time it was seen by the Journal as "unreliable".
The 1862 Journal claimed to promote the study of atmospheric phenomena on practical astronomic principles. However, many Astro-meteorologists were either astrologers or sympathetic to astrology. Two of the best-known Richard Morrison (Zadkiel) and William Simmonite, published their weather predictions in almanacs, which also contained horoscopes, political predictions and advertisements for personal astrological services. A little later AJ Pearce (Zadkiel 2) published a weather almanac in the 1860s and wrote The Weather Guide Book (1864) as well as a Textbook of Astrology (1879). There were dozens of almanacs. Old Moore's sold as many as 560,000 in 1839.
An offshoot of astro-meteorology was psychological meteorology, which centered on the assertion that celestial bodies influenced terrestrial matter by rays of magnetic fluid or electrical nature. Fitzroy was also interested in the investigation of invisible fluids in the atmosphere. His detailed analysis of a storm in 1859 stressed the electrical and magnetic disturbances that accompanied it.
One area in which astro-meteorology and science met more comfortably was that of "sunspottery". Evidence of a sunspot cycle was brought to light in 1851 when Alexander Humboldt publicised the work of an observer in his book Cosmos. Twenty years later sunspot observation became one of the pillars of solar physics. It was argued that sunspots could be used to predict weather cycles and produce long term forecasts.
The main figure in Britain associated with sunspots and weather was Norman Lockyer who began to promote the connection between sunspots and weather in 1871. The parallels between sunspot claims and astro-meteorology were obvious to contemporaries. Astro-meteorologists in the 1860's frequently cited spectroscopy work, suggesting that chemical properties could explain the influence of different planets. It was claimed that sunspot theory indeed belonged to astro-meteorology by those supporting the more conventional scientific approach.
One sunspot investigator openly acknowledged the connection with astrology. In 1875 William Stanley Jevons, the economist, turned to correlations between sunspot and metrological data o supply an explanation for economic crises. His theory was regarded by many as a joke but he himself placed great weight on the theory, defending it with vigour and collecting his sunspot articles for publication with other writings on economic fluctuations. His speculations were not unprecedented. At the end of the eighteenth century William Herschel had tried to find a link between corn price series and sunspots. Jevons recognised that if his theories were proven to have validity then science would go in the direction of what would appear to be akin to the astrology of the Middle Ages.
Jevons and Lockyer supported weather prediction research using sophisticated instrumentation and rigorous numerical analysis. The question of statistical analysis was becoming fundamental to meteorology. The statistician Francis Galton played an important role in this development and was a member of a committee investigating meteorology in 1865-6. Forecasting in Fitzroy's hands was a matter of individual judgment, one reason why it ceased after his death. Fitzroy himself spoke of using intuition and mentally channelling. The Royal Society designed the new government meteorology to avoid this unscientific tendency, establishing a chain of eight model observatories. From henceforth instruments and numerical analysis offered an escape to the close identification of the person of the scientist with his work.
Sunspot prediction was based on numbers gathered from precise instruments. Zadkiel's response to the reforms of 1867 warned against both dangerous atmospheric conditions on 26th August and the misleading meteorology of the "helpless old ladies, who figure away at the British Association." Saxby wrote scathingly of the mistaking of mechanical facilities for scientific progress. There was a widespread popular critique of instruments as they convinced only those schooled to interpret and accept their evidence.
Critics of sunpottery shared with astro-meteorologists a suspicion of numbers. The sunspot cycle itself was an uncertain figure and numerical analysis offered a process that was closed to public scrutiny. Today sun pottery is used by supporters of astrology as an explanation for how it works. For example, Dr Percy Seymour - "...astrology is not mystical or magical but magnetic. It can be explained by the tumultuous activity of the sun, churned to a lather by the motions of the planets, borne earthward on the solar wind, and perceived by us via the earth's magnetic field while we grow inside our mother's wombs" - speaks in terms instantly recognisable to the nineteenth century astro-meteorologists.
The simplest form of astro-meteorology is weather lore, sayings and omens used for many centuries to forecast weather, many based on simple astronomical phenomena. They are generally accepted as nonsensical and cited to show the deficiencies of astro-meteorology. Unfortunately for this argument, many of these sayings have been proven in recent times to have validity. The most commonly known saying comes in many forms, one being, "Red sky in the morning, shepherd's warning; red sky at night, shepherd's delight". This is a derived from a biblical quotation from Matthew 16:2 "He answered and said unto them, When it is evening, ye say, fair weather: for the sky is red." And it is also correct. Near sunset and sunrise rays from the low sun travel a much longer distance through the atmosphere encountering more dust particles. These scatter at the longer wavelength red end of the light spectrum (when the sun is high in the sky, that is why the sky appears a blue colour). A red sky at night suggests a clear sky for hundreds of miles beyond the western horizon, and no imminent frontal systems bringing rain. Red sky in the morning tells us little about what is approaching from the west, but only that eastern skies are largely clear. However when this corresponds to the zone of fine weather, which often exists between two depressions, then it suggests that it is moving away eastwards with more bad weather not far away to the west.
It was argued that weather prophets were no more than jumped up fortune tellers, who by adopting scientific pretensions had taken all the amusement out of the subject. Despite the best efforts of some astro-meteorologists, the link with astrology was undeniable.
In true Victorian style, numerous societies were formed to study the weather during the nineteenth century, comprising members of the scientific establishment and astro-meteorologists, often together. The British Association for the Advancement of Science, founded in 1830, included meteorology among the sciences which it intended to advance. The leading figures in the Association were quick to recognise the need for the standardisation of instruments and for theoretical frameworks to guide the gathering of data.
On 3rd April 1850, ten men assembled in the library of Hartwell House, near Aylesbury, Buckinghamshire. According to the minutes of the meeting, they gathered "to form a society the objects of which should be the advancement and extension of meteorological science by determining the laws of climate and of meteorological phenomena in general". They called the society the British Meteorological Society and appointed as its president Samuel Charles Whitbread, a grandson of the founder of the famous brewing firm. The society they formed still exists and flourishes. It became The Meteorological Society in 1866, when it was incorporated by Royal Charter, and the Royal Meteorological Society in 1883, when Queen Victoria granted the privilege of adding Royal to the title. It worked hard to escape controversy rejecting Morrison and White's overtures. Official weather forecasting began 1860s and it kept its distance from the disputes surrounding it. The forecasting work of the Met. Dept. gave meteorology an official and public identity.
The "advancement of meteorological science" remains the primary aim of the Royal Meteorological Society today, with meteorological science including climatology, physical oceanography and other related disciplines. Among its 3,500 members, the Society numbers not only professional meteorologists and other scientists but also people whose work is affected by the weather or climate and many who take an interest in the weather as a hobby. Weather observation was born as a hobby and even today it is impossible to ignore the many amateurs involved. The British Meteorological Society was not the first meteorological society to be formed in the United Kingdom. On 15 October 1823, at the London Coffee House, a group of scientists founded the Meteorological Society of London. This society flourished for seven months and then, despite the initial enthusiasm and activity, languished until 1836, when it was revived, only to fail again in 1843 partly through overspending. William White was responsible for reviving this society and served on its committee.
Five years later, another Meteorological Society of London was founded, but this, too, was short lived as in the summer of 1850, its Council dissolved the Society and recommended the members to join the British Meteorological Society. Many of the books and papers of the Meteorological Societies of London were handed over to the British Meteorological Society in 1851 and have survived. They are now cared for in the National Meteorological Archive, along with papers and books of the Royal Meteorological Society from 1850 onwards. The Scottish Meteorological Society, founded in 1855, amalgamated with the Royal Meteorological Society in 1921.
Whether made by government scientists, astro-meteorologists, lunarists or hobbyists, weather predictions were, in general terms, notoriously unreliable throughout the nineteenth century. The major difference so far as the public were concerned was that forecasts made by those who weren't associated with the governments efforts were far longer ranging than those that were. From 1860, the forecasting work of the Meteorological Department changed the status of meteorology, giving the emerging science a new official and public identity.
The work of the department was closely associated with Fitzroy, whose initiative the forecasting was. With Fitzroy's publicising of his work weather science became highly visible, rather than being a part time occupation of hobbyists, or a sideline for astrologers. Writing frequently to The Times and attending meetings to canvass support for his system Fitzroy wrote his Weather Book (1863) in the hope of bringing meteorology to a general audience.
Fitzroy's suicide in 1865 damaged the reputation of government weather forecasting owing to the close association his name had with it. He died in debt and a public campaign had to be launched to raise funds for his widow and family. In Victorian England suicide meant either sin or madness depending on your viewpoint. Weather forecasts temporarily ceased, although storm warnings were continued. Fitzroy was seen as a fraud and attacked in the press.
The reputation of astro-meteorology had already been damaged by this time, owing to Morrison's libel trial in 1861. Morrison had managed until this point to keep his scientific and astrological identities separate. In Zadkiel's Almanac of 1861, Morrison had predicted ill health for Prince Albert. When the Prince died of typhoid on 14th December 1861, "Zadkiel's" prestige and sales shot up. The Daily Telegraph mounted an attack against astrology and Zadkiel in particular. Unaware of Zadkiel's identity they were informed by a letter from an anonymous reader. The writer, Rear-Admiral Sir Edward Belcher, urged that Morrison be ferreted out and prosecuted as a rogue and vagabond - referring to the statute under which astrologers had been punished since 1824. Morrison immediately sued for libel.
The trial took place in June 1863 and many eminent witnesses testified on Morrison's behalf agreeing that he received no money for his ventures inn the 1850s. Although he won the case, Morrison was only awarded £1 in damages and nothing for costs.
Astrology became high profile just as Fitzroy's forecasts were getting under way. The Astro-Meteorological Society were meeting in Pall Mall under Morrison's presidency, (he was elected president in 1840) and Zadkiel was publicising White's theories in his almanac. Although Morrison's identity as Zadkiel was hidden in weather circles, with the connection now being made the issue of scientific versus traditional weather prediction was placed firmly in the public arena. In many ways it was reduced to a battle between Morrison and Fitzroy.
When Morrison died in 1875 Alfred James Pearce took over editing the Almanac. Pearce was also interested in meteorology having joined the Astro-Meteorological Society shortly after his twenty first birthday in 1861. Morrison was dead but Zadkiel continued to predict the weather.
In the nineteenth century Zadkiel was a byword for weather prediction. Meteorological office scientists were known as the government Zadkiel. During the nineteenth century the majority of weather predictions failed, whatever their source. The other side of this of course is that the same number were successful, whatever their source. Forecasting required public explanations and justifications. The object was to warn people of bad weather and anyone within reach of an almanac could supply themselves with data to compare predicted and actual weather. Fitzroy and Zadkiel both wrote letters and published their achievements as much as they could. Where lay the difference?
The scientific consensus is that weather forecasting has only been possible to any level of accuracy in recent years because of the development of technology. However, even today, observers such as Theodore Fujita, long-time professor at the University of Chicago, still use their own eyes as one of their principal observing tools. Although sunspot theory was first advocated in the 1870s it was seen as part of astro-meteorology and so largely ignored. It was difficult to separate it from astrology and so threatened to undermine the role of solar physicists as representatives of a modern science. It was necessary for the theory to be "discovered" by a scientist before it could gain acceptance. The Serbian astrophysicist Milutin Milankovitch is now credited for developing theories relating Earth motions and long-term climate change. The Milankovitch Theory, developed in the 1920s, states that as the Earth travels through space around the sun, cyclical variations in three elements of Earth-sun geometry combine to produce variations in the amount of solar energy that reaches Earth.
Some weather models can predict as far in time as nine to fifteen days, but the accuracy goes down to 50 percent only 48 hours into the forecast. This is mainly due to the fluidity of the atmosphere. The eye of a cyclonic event may be 1000kms out after only 3 days. For every day after that the error doubles. Many forecasters see meteorology as an art, as the data has to be interpreted by an intuitive and creative meteorologist. Isn't weather forecasting strictly scientific? Don't forecasters use the latest in computers, satellites, laser beams, and other high-tech aids undreamed of just a short time ago? Yes, they do. And still, they can't predict weather more than three days into the future with complete confidence. The scientific part of forecasting the weather doesn't seem to be that difficult. Atmospheric motions follow laws of physics that have been known about for years. One reason give for the forecasters failure is that, unlike most other scientists, meteorologists cannot verify their calculations by making measurements in a controlled laboratory environment.
Unfortunately, research has not been produced which analyses the success or otherwise of nineteenth century weather forecasts in terms of their provenance. There is little written on the modern history of meteorology and that which exists generally dismisses astro-meteorology as a superstitious hangover from an earlier age. Despite claims to the contrary weather forecasts are no more accurate today than they were a century and a half ago.