Weather forecasts have been created and distributed to the public in the U.S. for 155 years. The means and methods have seemingly improved by light years. Communication of important weather information and forecasts continues to become a work in progress, however. The situation is complicated by the amount of information that is available and the number of sources that the public uses to receive that information.
Evolution Of Weather Forecasting
Some of our founding fathers dabbled in the weather. Ben Franklin was the first to conjecture that weather systems move from west to east or southwest to northeast.
George Washington and Thomas Jefferson observed chronicled weather information and I have used some of that information in past articles dealing with historical weather events.
An official agency wasn’t established until 1870 when Congress established a Weather Bureau that was assigned to the War Department. The bill was signed by President Ulysses S. Grant. Congress transferred all weather responsibilities to the Department of Agriculture in 1890 and the US. Weather Bureau began in 1891.
Their forecasts were simply worded and only went out for a day. Just after the turn of the century, it became apparent to physicists that weather forecasts could possibly be made by a system of mathematical equations that could project the movement of fluids and gases in the atmosphere.
The forecasts were communicated primarily through telegraphic reports, with observations from field stations being sent to a central location for analysis and then disseminated back to observers, railroad stations, and the news media.
A “primitive” weather map from the early1870s. Map Credit- U.S. War Department.
A British scientist, Lewis Richardson, attempted to make a six-hour forecast but his calculations took 6 weeks to finish.
During World War Two, B-29 pilots flying missions over the Pacific came in contact with strong winds aloft, and the jet stream was discovered. This was an extremely important element in weather forecasting After the war, there was finally some progress made with computer technology. a 24-hour forecast could be made in 24 hours. It wasn’t useable but there was a considerable improvement.
By the mid-1950s, computer models were quite primitive as they began to create forecasts. The model most used was called the barotropic. It has some severe limitations, however.
Weather forecasts were communicated to the public on radio stations but many television stations were devoting more time on their local news to a weather segment.
In 1960, TIROS 1 weather satellite was launched and it became a new day for weather forecasting. We’ve come a long way from a grainy photo of the Earth to the complex GOES satellites where we can see multi-channel visible, infrared, and water vapor photos. In addition a complete profile of the atmosphere, including lightning data is recorded. It can also collect solar data.
Left- First image of the Earth taken by the TIROS 1 satellite in April 1960. Right- GOES satellite photo of a weakening tropical cyclone along the Australian Coast in April 2019. Image Credits- NOAA.
By the early 1970s, a new model (LFM) was at first experimental before becoming operational. This model had limitations, compared to what we have today, but it was a giant leap forward. Forecasts for two to three days became more reliable.
In the 1980s computer models really evolved and they could forecast farther ahead in time. I remember using the (84-hour) “Spectral” model. Three-day forecasts were becoming more accurate and you could at least get an idea of four to five-day forecasts in many situations.
Today, there is a myriad of models that can be used for forecasting. We have to remember that models are only a projection based on atmospheric data that is fed into them. These include mathematical equations and even current conditions and radar. Some models can project out for weeks and even months at a time. Accuracy does fall off with time, however. Much of this information is available to the public on various websites for free.
There are models that are long term, short term, and medium range. Some can be used for severe weather situations and some for hurricanes.
Below is an example of a model menu from the College of DuPage website. .
The total amount of information that can be communicated to the public is eye-popping, over the years that I have been a meteorologist. The various graphics and broadcast/digital media that can be used is also incredible.
An image from inside the NWS office in Albuquerque, New Mexico. Image Credit-YouTube.
The accuracy of extended forecasts has also increased dramatically. When I started out in this business, three-day forecasts were inconsistent and beyond that was a long shot at best. Hurricane track forecasts are so much better than they were in the past. Even mesoscale (small scale) forecasts like thunderstorms have improved to a great degree over the years.
GFS computer model surface and precipitation forecast for 21Z Sunday April 6, 2025. Map Credit-College of DuPage Weather.
Advances have also been made in radar technology since the Doppler network was launched by NOAA in the 1990s. Incredible 3-D analysis of thunderstorms is among these advances, along with tornado detection.
Communication Of Weather Information
There is so much information available and so many ways to communicate it. The trend of weather information communicated to the public gradually increased over the years but cable networks like The Weather Channel significantly pushed that trend ahead dramatically in the 1980’s ! The demand for specific weather information has been ever increasing. The major question is how much does the public understand what is being communicated to them?
There are so many situations. Winter weather, severe weather, and tropical weather are SO different and pose different challenges to the meteorologist. Although I have personally witnessed the evolution and accuracy in forecasts (especially extended forecasts) it doesn’t seem to translate to the public to that degree.
There are many points that I could make regarding this subject, but I just can’t accomplish that on one article. Perhaps I can continue to revisit this subject. I’ll start with misconceptions.
I have been in this business for many decades and I can assure you, meteorologists attempt to make the best possible forecast with the information that has to use and the experience that they have.
Meteorologists are products of vigorous programs of colleges and universities. They have to take courses not only in meteorology but also advanced math and physics. Many meteorologists have advanced degrees.
For meteorologists who are employed by TV stations or other media, I can speak to that from personal experience. Although TV producers look for what can draw viewers in, the meteorologists work separately. Often the meteorologist will work with the producer to craft headlines that draw attention but aren’t “over the top”.
The most effective communicators in TV weather are those who can tell the story to the viewer and weave in facts and figures if it’s important. There has been such an improvement in weather graphics over the years but these tools should be used to help to tell the story, than having the meteorologists read the information directly.
Tons Of Websites
There are plenty of places that you can go to on the Internet to obtain weather information. The National Weather Service and other government agencies are a good example. There are local TV stations and national TV networks. There is also a myriad of private weather services. There is no shortage of websites for “weather”. Some of these sites are run by weather enthusiasts rather then meteorologists or even high school and college students and those seeking attention to their website Once that happens the “word” of the forecast event spreads like wildfire (with exaggerations to boot).
If we could turn the clock back 25 years, there were far fewer entities to get your forecast. Perhaps that was better. All I can say is just find a reputable source and get to know how they convey weather information and submit questions to them.
The website of NWS Charleston, West Virginia on April 19, 2025.
App Explosion
Weather apps are commonplace now and the are so many of them. The emergence of these apps in the early 2000s provided weather data that was much more accessible to the pubic . Initially, the apps provided basic information like temperature and rainfall with static icons. With the emergence od smartphones, apps became able to provide more detailed forecasts (out several days) and included maps and radar. Now, the apps can give the user immediate notifications such as weather watches and warnings. This is especially critical when tornado warnings are issued !
There are many weather maps these days that can be downloaded on your smartphone. I won’t get into rating them breaking them down but apps like The Weather Channel can also feature weather, and weather related news features. The apps have become a major source of weather information and alerts ! The information has become more personalized and information like forecasts integrates AI for more accurate forecasts. The visual presentation of this information has improved by leaps and bounds over the old static images of 20 years ago.
A photo showing information from a weather app on a smartphone. Photo Credit- Lloyd Dirks on Unsplash.
Types Of Weather
Different types of weather present their own set of challenges for both the forecast side and the communications side.
Severe Weather
I’ll go to severe weather first because it can be complicated. Severe weather is on a mesoscale level rather than on a grander scale like winter weather and tropical weather. Trying to forecast where particular thunderstorms will be
or where a tornado will form is almost impossible until radar indicated that it has happened.
Forecasters can hone in on areas with computer models and upper-level forecasts for certain time periods, They also look at atmospheric parameters where severe weather can develop.
The only way to communicate this is to show a percentage chance of severe weather over a particular area and a time frame for the event. The chance for particular forms of severe weather like large hail, damaging winds, and tornadoes.
I’ll give you an example from Wednesday, April 18th. in northern Texas. The forecast was for that area, including Dallas-Ft-Worth, for severe thunderstorms with large hail, and damaging winds. My forecast would have been the same, by the way.
In this case, the hail would come from individual thunderstorm cells that would erupt. These would only cover a small area, so many residents wouldn’t see hail at all. The damaging winds would come later, from a line of thunderstorms. Only some thunderstorms in a line like that would be severe. So, really only a small part of the area would see severe weather.
From some Facebook sites, many thought that the hail and damaging winds would cover most of the area. Look at the storm reports below.
“Severe” storm reports from northern Texas on April 18, 2019. Green dots denote hail and blue is damaging winds. Notice that much of the DFW metro area was spared (circle).
Many residents of the Dallas- Ft-Worth area probably exclaimed “they” said it as going to be bad and it wasn’t as bad as they said it would be. The information and forecast were communicated. On this “meso” scale, we just can’t pinpoint where the severe thunderstorms will be.
Another thing that I wonder about is the SPC Convective Outlook. This actually goes out for eight days now.
The forecasts show areas that have a threat of severe thunderstorms and how high the threat is, by categories based on the probability that a severe weather event will occur within 25 miles of a given location. The threat levels include marginal, sight, enhanced, moderate and high risk.
An example of the SPC Convective Outlook for April 3, 2025.
This outlook was modified in recent years but I wonder how the general public understands this product. I will leave that for another day, however.
Several NWS offices post “storyboards” on their website when a significant weather system will impact their area. Take a look at an example below. I think that this is a very good idea.
An NWS Nashville storyboard showing that severe flooding is possible in parts of Tennessee. Posted on April 3, 2025.
Tropical Weather
I could write forever about this but I’ll just let you know that hurricane models have improved so much over the years and the track forecasts, indicated by a cone of uncertainty, are quite remarkable in recent times, as compared to many years ago.
Intensity forecasts have been slow to catch up but intensity models have improved dramatically in just the past few years. What the National Hurricane Center will do is to trend the intensity in their forecasts (usually in small increments).
As we have seen, there are times when hurricanes can intensify very rapidly in a short period of time. NHC impact products have improved markedly in recent times, as well.
There is a media frenzy over tropical weather systems, however. The can be lots of conjecture across various media platforms. Many folks are caught up in the category of a hurricane. That only gives us wind speed. There are so many other impacts such as total rainfall that can produce catastrophic flooding of high storm surge that can cause excessive damage. Loss of life in hurricanes and tropical storms are much greater from water than wind.
If a hurricane could possibly affect your area, it’s good to know what type of impacts it could in ict.
It’s incumbent on meteorologists to convey this information. For example in 2018, Hurricane
Florence wasn’t that intense but flooding was catastrophic in North Carolina, whereas Hurricane Michael was extremely intense and it produced catastrophic wind and storm surge damage in the Gulf Coast area.
One more thing, if you hear that the maximum sustained winds in a hurricane are 110 miles per hour, that doesn’t mean that every place that the hurricane hits will have winds with that force.. For example, if a hurricane overs a large area and only one point within it has sustained winds (1 minute) of 110 miles per hour then it’s listed at 110 miles per hour. So, most areas will NOT see winds of that intensity. Winds in a hurricane also drop off pretty rapidly away from the center.
Hurricane Michael “peak gusts” forecast frame issued by NWS Tallahassee on October 9, 2018.
Winter Storms
These can be tricky to forecast and to communicate. Some snow forecasts are easier than other’s especially if the temperature is solidly cold. When temperatures flirt with 32 degrees, especially where the snow forms in the clouds, just a slight difference in temperature could mean something other than snow and it will ruin the forecast.
There are also large scale winter storm and localized events (such as “lake effect”. These can be tricky to explain and I sometimes wide how much is understood.
Once again, I have never seen a meteorologist “hype” a snow event by inflating the forecast. They make the best forecast based on the information that they have to work with. I have seen what I would call “hype” by national news outlets or media websites that are NOT weather based.
Another thing that I notice is that sometimes a large amount of snow is forecast for an area four or five days ahead of time. Over time the forecast sometimes changes, to various degrees, but the original forecast is spread around and the updates don’t seem to resonate.
How Can Communication Be Improved?
So, as you can see, we in the weather communication business see a dilemma. How can we better communicate weather information? How can we make the general public realize that various weather situations can differ? How can we better educate the public about various weather events and what type of forecasts that can be reasonably communicated? There is no cookie cutter method to cover it all.
There just seems to be some kind of “disconnect”. Maybe it’s because are living in cynical times. Perhaps some feel that forecasts are exaggerated for ulterior motives. Maybe the public is not trusting of what information is conveyed to them. Is the debate about climate change spilling over to weather.? Are there too many weather sources with too much varying information. Is the information being conveyed not complete or not being properly understood?
