You’ve all seen it: Besides reporting when the weather is noteworthy (such as a blizzard), an important role of the local news’ weather person is to give the upcoming forecast. Usually, it’s a multi-day prediction, which they explain via green screen maps which show the macro progression of weather patterns across the landscape. Those maps come courtesy of satellite imagery.
Before the late 50s, there were no man made satellites. That changed with Sputnik. At some point, satellite imagery became available (the 60s?). So how did the weather man give the forecast before this technology? Television predates satellite, right? Didn’t news broadcasts with weather segments exist before satellites, too?
More generally, how was weather predicted before we could get a view above the weather patterns? I’m aware that hurricanes used to be a lot more devastating because people didn’t have advance notice.
Before there were weather satellites, systematic measurement of the upper atmosphere was done with radiosondes, expendable wireless weather stations launched from balloons. Thousands of these were launched daily in the US and used to compile weather maps. I remember when these maps were available by primitive FAX machines for rapid transmission within the US.
When I was growing up in Des Moines, we had a weatherman who would stand behind a glass (plexiglas?) map of Iowa and write temperatures by dots representing various cities. He was writing backwards and seeing the map backwards, and I never saw him make a mistake.
We tried to imitate him whenever we played television, but that never ended well.
I couldn’t find any videos of him doing this on You Tube.
Forecasts were also a lot less accurate. Even in the 80s and 90s, I assumed that tomorrow’s forecast was pretty good, the 3 day forecast had a decent chance, and anything beyond that was an educated guess.
Hey, I remember that too, though I was nowhere near Des Moines – must have been pretty common practice! Although I have a vague recollection of an even more primitive TV weather guy with a blackboard map, writing things in chalk, believe it or not – not sure if I ever actually saw this or just heard about it. It was once explained that the backwards-writing guys were able to see themselves in a mirror and with a bit of practice that made it a lot easier. Never quite understood why they couldn’t just flip the image electronically, but I guess those were Olde Tymes technologically.
As for weather satellites, the first really useful ones were launched beginning in 1964 with the Nimbus program. There were a couple of earlier ones but they didn’t provide much useful data. The 1970s and onwards was when weather satellites really started to come into their own. And at some point, sadly, the backwards-writing weather guy with all his finely honed backwards skills disappeared.
The usage of historical data hasn’t been emphasized enough here.
Meteorologists kept extensive and detailed weather data going back in the late 1800’s, as already noted. They used this to make statistical predictions.
So, they’d look at all the available data for today, say. Then they’d look at all the available data for a hundred years, for days that that weather similar to today. And they would note, for example, how many of those days were followed by rain the next day.
So, if 80% of the prior recorded days similar to today had rain, they would forecast an 80% chance of rain tomorrow.
That was the basic meaning of using percentages to give the probability of particular weather patterns. If they said 80% chance of rain tomorrow, that meant that 80% of prior similar days on record were followed by rain.
The smaller, rural stations still used map boards with Velcro pieces for the weathermap well into the late 1970’s.
There were two or three map boards that slid to the side and revealed the next one underneath. Each map was already set up before the newscast.
They had stick on pieces to represent highs, lows, thunderstorms, sunny, snow etc. They also had stick-ons for the temperature map.
Once in awhile one would dangle loose or even drop off during the weather segment. The map got jarred sliding back and forth. I think YouTube has some of these old bloopers.
Here’s one of the early weathermaps with stick-on symbols. https://goo.gl/images/a0k4Yh
I recall it took awhile for all the local stations to switch to green screen. The smaller stations waited a few years.
Radar was the same. The big local stations with a state wide audience got it first.
Ha! Here’s the chalk guy I was talking about [Youtube video]. This is from 1969 but he was doing this until at least 1972. He didn’t have to write backwards but he made up for that missing skill by throwing the chalk up in the air at the end of the broadcast and catching it each and every time, except for missing it once in his entire career!
I think one thing that is being missed is that weather forecasting is based upon an understanding of the physics of weather, and this has been understood in some form for a very long time.
You start with the Earth as a rotating sphere, add solar radiation, convention, Coriolis force, thermodynamics of water cycle, and you have a pretty good clue about what s going on. This gets you high and low pressure systems, warm and cold fronts, tilted axis of the Earth adds seasons, and we see a fairly constant rate of progression of weather systems. The most important tool in the set for a weather forecaster is the barometer. A national array of weather stations measuring barometric pressure will tell you exactly where the high and low pressure systems are, add in rainfall and you know where the fronts are. Temperature measurements and wind help. By now you can build a pretty good synoptic map, and it isn’t too hard to apply an understanding of how these systems evolve to make a remarkably good prediction of the next day’s weather, and not too bad a prediction of the next couple of days. But weather systems can be unstable, so there is always the chance they won’t evolve smoothly, and you get surprised.
I would say that the most important tool in modern forecasting is not the satellite at all. It is the supercomputer. A satellite can tell you more about what is going on right now, and is invaluable seeing severe weather developing out to sea. But it can’t predict the future. Right from the start weather forecasting has been a major customer for supercomputers, and they still are. Weather is difficult, and famously chaos theory was born out of an understanding of how unstable naive models were. But over time, as computers became more and more powerful, and mathematical and programming tools advanced, it became possible to build much more sophisticated models (coupled models being a big step), and just as importantly, run ensemble models, where you run the program many times with slightly perturbed inputs and look at the overall pattern of results across many runs. This is what gets us the 7 day forecasts, Not satellites.
The forecast that delayed D-day was probably the most crucial ever and used information from every available source. The Americans wanted to go, but the British, led by Group Captain James Stagg, told Eisenhower to wait.
As I understand it, the chalkboard was pre-prepared, with markings written in a color that was invisible to the black-and-white camera. The weatherman simply traced over them in a color that the camera could see.
The need to establish stations to report on what the weather is doing over there right now–in order to predict what the weather will be doing over here the day after tomorrow, and specifically in a military context–led to fascinating little side stories of history like secret German weather stations in Greenland during World War II (Wikipedia link), which in turn sound like they could be the starting point of anything from a two-fisted war story to a comic book superhero plot to some kind of Lovecraftian science fiction/horror tale. (“A secret Nazi base in the remote frozen Arctic wastes! Dun dun dunnn!”)
Now now now … weather forecasters have a well deserved reputation about their accuracy …
Francis Vaughan laid out the basics above quite well … I just want to add that weather satellite data isn’t all that great, doesn’t really say much or give much information … it shows it’s cloudy in Des Moines but our partner there already told us that … weather satellites give us very limited information on the vertical structure of the atmosphere … so to this day we still launch radiosondes regularly all around the globe … it’s still the only way to get good data aloft; pressure, temperature, humidity, wind speed and direction …
If the wind is blowing 10 mph, and we want to know the weather in 6 hours … we simply look at the weather 60 miles upwind …
Computers are the big break through … but it’s just a tool … we still need experienced forecasters to look at the output of the various programs that are run to decide what the forecast should be … using spaghetti charts … when all the models focus on one 3 day solution, great, you have your forecast … but when the models are scattered all around, then the forecaster has to try and decide which model is best for the given environment … and hope … and these things help accuracy some, but the BIG change is we can say HOW accurate a forecast is … here where I live, we regularly get statements from the NWS basically saying the models are everyplace so tomorrow’s forecast is just the average result … the weather may or may not be anywhere close to what they’re saying … although as a rule we don’t hear that on commercial news casts … they have product to sell us …
Yeah, they made a note of the readings on what instruments they had (barometer and … humidity thingy), and also what the weather was the next day. And then radar became a thing in the 1950s, so they could see the storm clouds forming.
a lot of people pick their local news based on the weather people. One local station has had their main weather guy since 1981 , they are tops in ratings.
If you see a local TV weather guy/gal get fired, it could be that ratings are down . I lived next door to a local TV weather person and she lasted only a few years. She was their lead weather person at 6 and 11 newcasts.
Thomas Jefferson kept a detailed daily account of the weather his entire adult life. That wasn’t unusual among gentlemen of the era, when “natural philosophy” was considered an aspect of a well-rounded mind alongside knowledge of Greek and Latin.
Predicting the weather was one of the most serious parts of life before people started living in cities. Farmers needed to know when to plant, when to reap, and what conditions they could expect in between. Sailors had to worry about winds and rain every day. Nearly everybody in America was one or the other.
You may have heard of the still existing Old Farmer’s Almanac. The original Farmer’s Almanac started in 1792. It took the available data and predicted coming conditions, along with other critical data likes sunrise and sunset, tides, times for planting. Was its accuracy on future weather better than chance? Doubtful. But it was one of dozens of almanacs so you could always try a competitor to see whether it had better weather. That Benjamin Franklin guy’s Poor Richard’s Almanack was a bestseller for years starting in 1732. In fact, these were among the bestselling books of any type in the colonies and after the revolution. Franklin made his sell by including his pithy saying that have become famous but all of them did some variation on this, knowing that almanacs were essential reading every day of the coming year.
My dad was an FAA Flight Service Specialist. He would gather hourly weather data from the Flight Service Station for the aviation weather reports, as would FAA personnel from all over the country. Weather maps showing winds and fronts, etc. would be faxed to each station. (I recall the fax being rather damp when it came out of the machine.) Warm fronts were gone over with a red marker, and cold fronts would be gone over with a blue marker. Occluded fronts would get both. The maps were then made available to pilots.
