Tikalon Blog by Dev Gualtieri

Weather Forecasting

April 8, 2019

Weather was a foremost topic of the many jokes I learned in my childhood. The following are the most memorable:

* Where does a snowman keep his money?
   In a snow bank.

* What do you get when you cross a snowman and a vampire?
   Frostbite.

* When do you realize that it's raining cats and dogs?
   When you step in a poodle.

Medieval children "getting medieval." Winter weather in Upstate New York is extreme, but I enjoyed the snow when I was a child. Now, the snow-covered driveways and roadways bring fewer winter delights. (Detail of folio 12r, Codex 103, "Book of Hours" (Dijon, 1524), via Wikimedia Commons.)

The origin of the unusual expression, "raining cats and dogs," is unknown, but conjectures about its possible origin are given in an article on the United States Library of Congress website.[1] Its first recorded use was in the 15th century, and its origin is sometimes associated with Odin, the Norse god of storms, being depicted with dogs and wolves, the symbols of wind. That's a rather weak conjecture since there are no cats involved in that depiction. Knowing the 15th century origin of the expression and the classical education of that time, I place my bets with the Greek expression, κατα δοξα (kata doxa), against belief.

Early man must have been amazed that something invisible like the wind could exert a force. This must have fueled a belief in invisible being such as ghosts. The minor Greek god, Aeolus, was known as the keeper of the winds. There's a saying that everyone talks about the weather, but no one does anything about it. Primitive man did try to do something about it by performing rituals, like rain dances.

Weather forecasting has been practiced for more than 2,500 years, albeit not too successfully. Aristotle (384 BC - 322 BC) gave some forecasting tips in his 350 BC Meteorologica,[2] as did Theophrastus (c.371 - c.287 BC) in his Book of Signs (De Signis).[3] In later times we've had the old nautical adage, "Red skies at night, sailors' delight. Red skies in morning, sailors take warning," that I mentioned in an earlier article (Red Skies at Night..., April 25, 2014).

The invention of atmospheric sensing instruments like the thermometer and the barometer, when finally coupled with reliable telecommunication, offered a scientific means of understanding and forecasting the weather. This finally gave the ability to map air masses over large areas, such as the contiguous United States (see figure), that enabled improved weather forecasting.

Weather map of the contiguous United States for July 19, 2003. I selected this date since that was when I observed an exceptional episode of a radio propagation phenomenon called tropospheric ducting on the FM radio band. From 10:30 AM to 3:30 PM I was able to detect, at my Northern New Jersey location, WYYX, Bonifay, Florida on 97.7 MHz as if it were a local station. (National Oceanic and Atmospheric Administration map. Click for larger image.)

Computers have advanced many scientific disciplines, and that includes weather forecasting using numerical methods of weather prediction. The first significant milestone in this regard was a forecasting program implemented on the ENIAC computer in 1950.[4] This forecast happened in real time; that is, it took an entire day to forecast the day's weather. It was realized that faster computers would give weather forecasts for future days. Of course, this hope was predicated on the development of adequate atmospheric models. On the 100th anniversary of the American Meteorological Society, a short perspective on advances in weather prediction has been published in Science by geoscientists from Pennsylvania State University (University Park, Pennsylvania) and the Massachusetts Institute of Technology (Cambridge, Massachusetts).[5]

As stated in this article, the economic benefit of weather forecasting is 3-10 times its cost.[5] In a 2009 study, the value of weather forecasting derived from from $3.4 billion in public expenditure and $1.7 billion in private expenditures is $31.5 billion to US households.[5] Adults in the US access weather forecasts 300 billion times per year, and geographically specific weather information is available instantly on smartphones and other devices.[5] I get my forecasts from the National Weather Service and Weather Underground, generally from a desktop computer.

Many parts of the world are now experiencing high wildfire activity; and, as we glean from news reports, the developing world is quite vulnerable to weather disasters. The number of people living in flood-prone areas has more than doubled over the past 30 years.[5] The World Bank reports that upgrades to national meteorological and hydrological services of about $2 billion, with annual expenditures of about $500 million, could save 23,000 lives per year and provide many billions of dollars annually in economic benefits.[5]

The accuracy of weather forecasting is improving through improved observations, better models, and better data acquisition and data management.[5] Accuracy of predictions has continued to improve for such weather events as Hurricanes, blizzards, flash floods, hailstorms, and tornadoes.[5] Presently, 72-hour predictions of hurricane tracks are more accurate than 24-hour forecasts were 40 years ago (see figure).[5] Today's 5-day forecast is as accurate as a 1-day forecast was in 1980, and forecasts up to 10 days provide useful information.[5-6]

Advances in hurricane prediction. These are the track errors of Atlantic Basin Tropical Storms and Hurricanes from 1970 - 2017, with least-squares trend lines superimposed. (Graph from the National Hurricane Center, National Oceanic and Atmospheric Administration, reformatted using Inkscape. Click for larger image.)

Such progress has been made despite spatially incomplete and uncertain data. There are also limits imposed by sensitivity to initial conditions, a principle known as the butterfly effect.[5] Such sensitivity to initial conditions can be mitigated by using a large number of computer forecasts starting from slightly different initial states.[5] Forecasts are aided by known seasonal and annual cycles, such as the Madden-Julian Oscillation that moves eastwards around the tropics over 30-90 days.[5]

Weather forecasting is important to renewable energy, since forecasting of weather-related effects such as sunlight and wind availability is needed to balance electrical grids.[5] Increased use of unmanned aerial vehicles for monitoring hurricanes and other extreme weather events have great potential.[5] Improvements in high-speed computing offer the opportunity for improved weather forecasts.[5] While newer technologies are expensive, governments should resist putting weather forecasts behind paywalls.[5]

Monsoons at sea.

The dynamic imagery of clouds and extreme weather have been an inspiration to artists.

(Colored wood engraving by C. Whymper, a Wellcome Trust image, Photo V0025039, via Wikimedia Commons

Weather Forecasting

References: