How Meteorologists Work

to talk about that? I mean, you're not gonna have an emergency weather broadcast burst into your favorite viewing time of watching ALF and say warning, things are gonna be really nice this afternoon. So we're gonna talk about today about some of the tools that meteorologists use in order to get information about what's going on at this very

moment as far as the weather is concerned. In our last episode, we covered the basics of how weather works, so that man, it was more of a science based show, less technology, but it was important so that we can build the foundation to understand why meteorology is so incredibly complicated and so process or heavy. Like when you talk about supercomputers. One of the main uses for supercomputers these

days is to run weather models. And that's because you're talking about an enormous amount of data that has to be processed running you know, various types of calculations on it in order to create the outputs that we use for weather forecasts. And as our weather models become more and more high resolution, more precise, using more more readings per hour, that demand of processing power increases quite a bit. So let's talk about just the stuff that meteorologists used

to gather that data. This isn't even about the weather models. In our next episode, we will cover how weather models work, because it's fascinating all in its own. Today we're gonna focus on the stuff that meteorologists used to get the data that feeds into those weather models. Uh So, meteorologists

do a few different things. They do make observations of current weather conditions, so they're telling you what's happening right now, and that's what I'm focusing on for this episode and the next one will look more at the forecasting side. How do they actually predict what will happen next. Observations let you know the current conditions. They include data like wind speed, wind direction, air pressure, temperature, humidity, u V, radiation numbers, smog fog, all of these things and more.

Are you need observation stations throughout any given region, the region that you are responsible for, let's say, and you need to have a lot of them in order for you to get a more complete picture of what is really actually happening. If you have only a few observation points, then you would have what we would call low resolution. You would not have a very accurate accurate view of what was going on within the region, unless your region

is particularly small. Let's say that you are responsible for gathering the information across an entire county. Well, a county could be lots of square mileage, and depending upon where your observation stations are, you might have information that's very relevant for a specific part of that county, but it might not be so relevant for other parts of the county. And yet you have to create weather forecasts based on

your observations. It means the further out you go from those observation stations, the less reliable that data is going to be because it may not reflect what the actual conditions are further away from those observation stations. So having a high density of observation stations is critical for having very precise, whether a picture of what's going on with

the weather and thus making more accurate weather forecasts. So it could cause a problem, right Like, you could end up giving a weather forecast that ends up being useless for a significant percentage of the people who are relying upon your weather forecasts if they happen to live far away from where your observation stations are. So let's think of this in more concrete terms and you'll really understand

how this can become a problem. Let's say that you have a region that includes some different areas within with different levels of elevation. So let's say that you're at the foothills of a mountain range. Well, we'll say this the appal Achians. So, uh, I live in Atlanta, Georgia, where on what is called the Piedmont it is h

hilly but not mountainous area. If I were further northeast of here, or even just further north of here, I would start getting into the foothills of the Appalachian Mountains. So let's say that I'm on Just like that, I'm looking at a region where it's covering the end of the Piedmont into the foothills. That's several different areas of

different elevation. That means that those different areas are going to experience different weather patterns because weather is all about atmospheric movement and things that are going on within the atmosphere. Atmosphere is going to move over the topographical features of any given region, the geography, and with hills and mountains, that means the weather is going to do different things

in those different areas. So you're going to have folks checking the weather report and wondering why it says there's no chance brain while they're being rained on, and it may be because they live on the opposite side of a mountain from where you are with your observation stations, and at the observation stations everything's bone dry because that's what you know, that's what you've based your forecast on,

because that was where the data was coming from. But on the other side of the mountain, because you've had warm air masses pushed up to higher elevations as they moved over a mountain, They've then cooled down, water has condensed, and precipitation has begun to fall. People on the other side of the mountain they get rained on, and yet they're living in that same region that's being covered by you, because that's the county they are in, or the zip code.

This is why it gets really problematic when you start making weather forecasts, because they depend so heavily upon the geography of the respective regions and the geography and and observation stations within that geography. Now, if you have observation stations throughout your area, you can give much more precise weather forecasts. And say, in this part of the county, you would expect conditions to be dry, but up here over in the mountains, you're gonna start seeing some rain.

So you start seeing where the complexity comes into play. And this is a very small scale example of this problem. Once you start looking at it from say a statewide or nationwide or global perspective, you begin to realize this is really complicated stuff, and this is a big challenge

for meteorologists. You might notice that many weather apps allow you to search for weather forecasts by zip codes, but again, zip codes do not necessarily conform to geography, and you could have lots of different geographic regions within a single zip code, and that means that the weather forecast for one of those parts of the zip code may not be accurate for all the other parts. So if I check my app and it says hey, it's gonna rain, and I step outside and it's sunny, I might think, well,

what the heck is wrong with this weather forecast app? Again, they have to try and give you a forecast that's going to be relevant for the entire zip code, even if different parts of that zip code are in different geographical regions, like different geographical features mountains or streams or lakes or ocean or whatever it may be. So you can't really take all that into account if your criteria for generating a forecast is based solely upon zip codes.

That's why a lot of different weather apps and services are now using geolocation data to give you a more precise weather forecast for your your requests. So, if you're carrying a mobile device and you have a weather app on it and it's asking hey, can I have access

to your GPS or location data? If you say yes, then the weather app can look for your location and try and base the forecast for you closer to observation stations that are relevant to your your actual location So if I happen to be in a part of the county that is not the main population center and thus not the place that the forecast is really gonna cater to, I might get something more personalized saying, hey, well, based upon where you are, you're gonna be Charlie Brown and

have a little cloud follow you all day raining just on you, which has never quite happened to me, but some days it feels that way. It's sidesteps that zip code problem, so that the weather forecasting system can consult the observation stations really closest to you at that time and to project weather based upon that. But even then

you can still run into issues. So, for example, if you happen to be closest to an airport observation station, I mean most really all airports have some form of meteorological meteor logical boy man, that's gonna be a lot of fun today. Meteorological observation stations. Almost all air airports have at least some element of that, because clearly the weather conditions are very important when it comes to air travel.

One problem with that is that airports have very big runways and uh tarmax things that absorb a lot of heat and give off a lot of heat, more so than the surrounding area. And as we learned in our last episode, heat is a big important factor when it

comes to impacting weather systems. So if you happen to be close to an airport, that might end up throwing off some of the readings that you would get because it has this island effect right at the airport, But that might not extend very far outside of the airport's borders. So while you're getting observation data that's close to you,

it may still not be relevant to you. A lot of this sounds like I'm apologizing for weather forecasts that aren't entirely accurate, and I guess you could argue that I kind of am. But mostly what I want to stress is just, um, what a tough job this is. It's an interesting and fascinating thing, and we've learned so much about atmospheric fluid dynamics and just how weather works in general as a result of lots and lots of men and women putting their heads together and and making

observations and sussing all this out. But it's still really really hard to do, which is why it's one of the reasons why we don't do it perfectly now. Besides resolution, as in how many observation stations you have within a given region, you have to take those observations frequently. This helps with precision. If you take one observation in the day, early in the day, and that's it, then you can't take into account any changes that happen afterward, which means

that you can never really update your forecast. And that means that the further away you get in time from that initial observation, the less precise and accurate your forecast is going to be. If you continuously or at least regularly take observations, you can then update your forecast as a result when conditions change, and that way you can

keep your at least your immediate forecast more accurate. You may notice that the further out you go from the point of forecast from the current time, the less accurate these forecasts tend to tend to be. So if you're looking at a date that's ten days out, it might be more or less right on the money, or maybe way off base. Because weather is again very complicated and constantly changing. Uh So you want to make sure that the information you're relying on doesn't age so much as

to be irrelevant. So you need lots of observation stations. You need to take a lot of observations per given unit of time, and the combination of those two requirements means that you are generating an enormous amount of data that then has to go someplace and then be processed for you to get forecasts. This is why you need those hefty computers like supercomputers to process meteorological data. And

more on that in the next episode. But first let's talk about some of the basic instruments these meteorologists are using to gather actual weather data. Then in our next episode will move on to how they use that to create models for weather. And also why are there different models. You've probably heard of various weather models. Why do we not just have one? Why are there multiples? And why are some quote unquote better than others in very specific scenarios.

We're going to cover that in the next episode. Now, generally speaking, meteorological instruments fall into two broadcat gregories. You have direct sensors also called institute sensors. They're inside the situation like a thermometer that is left out to measure the temperature of the air. It's institute. It is a direct sensor. It's directly measuring the temperature of the air outside. Then you have remote sensors that are measuring something that's

much further away. The name pretty much gives you the indication of what it does. Now, we're going to talk about both types in this episode. Direct sensors include lots of stuff that you're familiar with, and we're gonna start off with a good old, easy one to to talk about thermometers, the noble thermometer telling us such information like dude, it's wicked cold out that today, but on your coat, or seriously, buddy, it's boiling out there. Let's play some

player unknowns battlegrounds in the air conditioned house. But how do thermommeters actually work? So let's start with your basic mercury thermometer. It's something that is still being used around the world. It's your basic bulb thermometer. It depends upon a simple physical principle, which is that a liquids volume changes relative to the temperature of the liquid. So when you heat a liquid up, it's molecules decide to do the equivalent of the guy I'm always seated next to

on a long distance plane ride. In other words, it spreads out well beyond its normal parameters. You know who I'm talking about. Give me my arm rest back. But that's what happens with liquids. You heat them up, the molecules get more energy, they get excited, they move around more, and they spread apart. So as a result, the liquid expands, and if, of course you do this long enough, the liquid will end up turning into a gas, which is

even more free form than liquid. Obvious asleep. When liquids get cold, those molecules end up huddling together, not so much for warmth, but because they have less energy they don't move around so much, so the liquid actually takes up less space than it normally would. So your basic bulb thermometer consists of a small bulb at the base and a narrow, long, closed tube leading up from that

bulb base. These physical proportions accentuate the change in volume of the liquid, So you want the bulb to be small because you want any changes in temperature in whatever environment you are measuring to rapidly be reflected within the thermometer itself. So, for example, if you're talking about outside. Any drastic change in the outside temperature, you want that to be reflected in an outside thermometer pretty quickly. Typically

those temperatures don't increase or drop that fast. But let's say you want to go from room temperature thermometer and you're testing someone's temperature. You want to find out they have a fever or not. You don't want to have to wait a very long time for that change in temperature to UH to happen within the mercury inside that thermometer. So that's why that bulb is small. It's just a

small amount of liquid. Doesn't change take very long for that change in temperature to move through the liquid, and thus the volume increases inside the the thermometer. UH. With weather, it means you you'd want something that will remain liquid at temperatures found across most of the planet, which is why we use mercury. Mercury is a metal. It is a metal that is liquid at room temperature and a

lot of temperatures that you will find on Earth. Very useful in that sense, it's not going to boil away rapidly at high a temperatures, nor does it freeze at your typical low temperatures. When you get two very low, which does happen on Earth, mercury will free So it's not perfect, but it's reliable and it's easy to read the differences. It's um it shows up well in a glass thermometer, and as it turns out, uh, mercury liquid

mercury is is pretty reliable. It remains liquid at temperatures that range from six D seventy four degrees fahrenheit, which is three D fifty six point seventy three degrees celsius, all the way down to its freezing point of minus thirty eight fahrenheit or minus thirty eight point eight three celsius. But hey, here's a fun fact. The most extreme cold temperatures on Earth get way below minus thirty eight celsius.

In fact, in NASA released satellite data that measured the lowest recorded temperature at minus ninety four point seven celsius, which is minus one thirty five point eight fahrenheit. At that temperature, mercury itself would freeze. So for very low temperatures you cannot use a mercury thermometer. You just you're not gonna it's gonna be as cold as it gets. It's already frozen. It's a solid, so you actually have to use a different kind of liquid. If you want

to use a liquid thermometer, alcohol works. Uh. They call them spirit thermometers in the old days because you're talking about spirits alcohol. Alcohol is a very low freezing point, and it's boiling point, however, is much lower than mercury, so you can't use it for very high temperature things, you know, but it works great for low temperature applications. And uh, this actually leads me to an interesting question that I think is fun to tackle, even though it's

not not quite as technical. How did we come up with the fahrenheit and celsius scales? Well, it all has to do with the freezing and boiling points of water, which makes sense. Water is very prevalent here on Earth. Most of our surface of our planet is covered in water. We depend upon water for our survival, so the temperatures at which water will freeze or boil are obviously important to us. So fahrenheit will start with that because that

scale was proposed first. That came from Daniel Danny Boy fahrenheit in fourteen and as far as I know, no one else called him Danny Boy, but I'm waiting for it to catch On fourteen, he decided to use a scale designed actually by a predecessor of his. It was not it was not completely invented by Fahrenheit. He took a scale that was made by a man named Olaus Rummer. Rummer's thermometer listed zero as the lowest point. That was not the freezing point, but it was as low as

the thermometer could register. Was zero and at seven point five on Rumors scale, that's where I would melt into water, or if you prefer where water would freeze into ice. It is the freezing point or melting point, depending upon your perspective. That was seven and a half on or scale at twenty two and a half that was considered body temperature, and sixty was the temperature for boiling water.

So it went from zero to sixty with boiling water being at the top and freezing being at about seven and a half fahrenheit to create a mercury thermometer, and it was capable of making more precise measurements than the the spirit thermometer that Rumor had been using. And because of that precision, since you could measure smaller changes in temperature, fair Kneit felt that there needed to be a scale that would be broader than than Rumor's scale so that

you could easily talk about tiny changes in temperature. Right, it's just getting an extra level of precision in there, and it means that you don't have to subdivide those units into further and further decimal points in order to describe the differences of temperature changes. So Fahrenheit ended up first taking rumor scale and he multiplied it by four. Uh.

He then adjusted the scale. He started doing some research and realized that just multiplying it by four it meant that it wasn't as accurate as it needed to be. He had more levels of precision, but the accuracy was off. Multiplying it by four it multiplied not just the scale but also the imprecision of that original scale. So he

started to try and refine it. Fahrenheit ended up after he passed away, people took his scale and began to refine it more, and they began to establish the freezing point and boiling point of water, and decided to set the freezing point and boiling point apart by one eighty degrees, which is important in math, but not so great for just casual conversation, So some people would say it was kind of an arbitrary decision to make it a one

degree difference between freezing and boiling. The temperature of freezing water was eventually established as thirty two degrees, which means boiling water would be two hundred twelve degrees. One benefit of the scale was that the units would allow for subtle descriptions of temperature changes without the need for decimals. So if you were to describe the temperature as rising from eighty six to eighty seven in fahrenheit, that's easy.

But if you wanted to say the same thing in celsius, to take those same two temperatures and talk about that increase, you'd say it went from thirty degrees celsius to thirty point six degrees celsius, or so if you said thirty one celsius, you're not being as precise because that's a greater change in temperature than what you're actually referring to. Now that being at having the temperature for freezing water set at thirty two degrees is a bit frustrating, but

I think I have an explanation for this. This is Jonathan's supposition corner. I kind of wish I could get Dylan to make a musical sting for this. That just sounds confusing. All right, here's here's my pitch. Why is water freezing at thirty two degrees? Why is it thirty two degrees? Why is it not zero? Why would you not start at zero for the freezing point of water? If water is the really important part on your scale, it's because it can get colder than freezing here on Earth.

And so you've got fahrenheit saying, well, I want a scale that I can measure the temperature even when it's colder than freezing, because some days it is colder than freezing, so I need my scale to be able to reflect that. But how do I measure that? If I've set freezing as zero? Where do I go from there? I mean, I can't have the mercury go further down. I've got it. I'll set freezing higher up on the scale, And that way you can still describe stuff that's colder than freezing water,

but not colder than what I can measure. So, uh, you set the bottom of your scale lower than the temperature for freezing. In that way, every measurement you take is a positive unit, you don't have to create negative units, So you just set your scales based at the temperature lower than what it tends to get down to. Right. So Fair Knight, say as well, doesn't often get colder than today, So today is gonna be zero, and everything above that will be fine, because how could it get

colder than today? That's Jonathan's supposition, corner, y'all? All right, but what about Celsius. Celsius was the brainchild of Andres Celsius in seventy two, so a few decades after Fair and Height, he had the clever idea of creating a temperature scale where the freezing and boiling points of water would be separated by one hundred degrees, making it much

easier to talk about, especially for mathematic uh purposes. So zero to one, taking that simple decimal scale, it makes sense, uh, Although originally he intended to make one hundred the freezing point of water and zero the boiling point of water, so in other words, the scale was inverted. The higher the unit number, the lower the temperature would be. So really he was creating a scale to measure cold, not heat.

So a one hundred degrees Celsius in the original implementation would be freezing temperatures, whereas a zero degrees celsius in the original implementation would be boiling water. Guests arguing that why would you ever get hotter than boiling water? Maybe Uh, this did not stick obviously. His contemporaries ended up deciding that that was not logical and perhaps a bit whacka noodle, and so they flipped it so that zero is freezing on is boiling. Uh. Celsius did not live to see

his his his standard became a standard. It was not a standard during his lifetime. He actually passed away only a couple of years after proposing it, so he did not get to see how it was adopted by almost the entire world, with the exception of some notable places such as the United States of America, where we still use fahrenheit and not celsius. Although I still like fahrenheit because it is easier to talk about more subtle changes

in temperature than it is with celsius. Also, I just grew up with it, so at this point it's hard for me to sit there and think, like, if someone tells me it's twenty five degrees celsius, I have no real, no real correlation of that in my brain. Like I couldn't tell you how warm or cold twenty five degrees celsius is um but if you tell me that it's eighty nine degrees fahrenheit, I know how hot you're talking.

So you know fun times. But what if you wanted to convert fare kneit to celsius and you don't have access to Google, which makes it really easy to do and how I did it whenever I needed to make conversions. Let's say that you want to do a conversion of fahrenheit to celsius, but you don't have access to a temperature calculator. Well, you just follow this handy dandy guide. You take your fahreneit temperature, you subtract thirty two from that temperature. Then you multiply your new number by the

number five. Then you divide that new number by the number nine. Now you know what temperature it was half an hour ago before you had to deal with all that math. It's a joke. I'm a liberal arts major, so make a lot of jokes about being bad at math. They're mostly jokes, there's some truth to them. Honestly, I don't think anyone really knows how math works. That's also a joke. These days, more thermometers are actually electronic. They're not based upon some liquid moving up or down a tube.

Based on changes in temperature, they end up using thermo resistors, or sometimes they're just called thermistors. That's also not a joke. They really are called that. These are materials they experience a change in electrical resistance due to temperature fluctuation. So remember, electrical resistance is the tendency of a material to resist or impede the flow of electrons through that material. If you have a material that has a very low resistance

like copper, those are good conductors. Metals that have our materials rather that have a very high resistance like rubber are really good insulators. And temperature turns out can affect some materials with their electrical resistance, which means that at certain temperatures electricity may pass more easily through that material than at other temperatures. So that is the basis for electronic thermometers. The most common thermistors have a resistance that

decreases as temperatures increase. In other words, they become more conductive the warmer it gets. This is called a negative temperature coefficient or NTC thermistor because it's this uh see saw kind of relationship, right, the temperature increases, resistance decreases. There are also some that have a resistance that increases with temperature. These are called positive temperature coefficient or PTC therm mists. So with these, as the temperature goes up

in the material, so does its electrical resistance. But most electronic thermometers use NTC therm mists. The relationship between temperature change and variation resistance is not a constant, so you cannot say that resistance changes by the same amount of ohms. Those are the units we use to measure electrical resistance. But you can't say that it's gonna change by the

same number of ohm's per degree of celsius. So if you go from twenty three to twenty four degrees celsius and then twenty four to twenty five degrees celsius and twenty five to twenty six degrees celsius, the difference in resistance by ohms is not going to be linear in

relation to those changes in temperature. Typically, thermistor resistance varies in a nonlinear way, but in a way that you can still factor by using a formula, So you have to design a formula that takes all of this into account in order for you to relay a change in resistance as being a change in temperature. So an electronic thermometers microprocessor detects and measures these changes in resistance, takes that formula into account, converts those measurements into temperature units.

Their mists are also used in other applications as well, so you might use one to protect a circuit from electrical overload. You've got electricity running through a circuit. A current is running through the circuit. Let's say that the current increases uh to a point where it's going to cause issues with the circuit if it continues on this path. If you have a thermistor positive a PTC thermistor in that circuit, then as it warms up, its resistance will increase.

So current runs through the thermistor that makes it generate heat that ends up changing its electrical resistance, and eventually it ends up becoming a barrier to current, so that the current cannot continue to flow through the circuit, and thus the thermistor will end up protecting the rest of the circuit from electrical overload. So that's one potential application of a thermistor outside of electronic thermometers. I thought it was pretty nifty. Well, I've got more nifty things to say,

and we're going to move away from temperature. But before I do that, let's take a quick break to thank our sponsor. All Right, so we've covered temperature. Now let's tackle our next sensor, the barometer. Barometers measure air pressure, and as I mentioned in that last episode about whether air pressure plays an enormous role in how weather behaves. So knowing the current air pressure conditions helps meteorologists understand

what might happen next. So, for example, if your area happens to be under high air pressure, that's an indicator that you're not likely to see very much rain that day. High air pressure systems tend to keep rain systems out. Typically, if the pressure is starting to drop, so you're seeing a change in air pressure, that could indicate that it's

going to get windy. It might possibly indicate that there's some wet weather on the way because a low pressure system is moving into what was a high pressure system. But you have to have something to measure that air pressure changes. An air pressure at any given altitude are typically too subtle for humans to really pick up on. Right, Like, if I'm at sea level, I'm not likely to detect very subtle changes in air pressure, but I would notice the difference if I were to go from say, Death

Valley to Mount Everest. Those changes in altitude are so dramatic that the differences an air pressure would be noticeable and actually a fatal problem. On Mount Everest, I wouldn't be able to adjust to that remarkable drop an air pressure, not to mention temperature, uh that rapidly. I would need to acclimate to it to avoid getting sick and potentially having a really fatal problems. So, yes, you can detect differences an air pressure, but typically if you're at a

single altitude, you're not moving up or down. You're just experiencing changes in air pressure due to pressure systems. You probably aren't going to be conscious of that change in air pressure because it's it tends to be fairly subtle, even though it can mean some major changes in weather. Barometers measure air pressure. They detect how much air is pressing down on them, so in a way, it's kind of like a set of scales for the atmosphere. So

how do you do that? How do you make something that can actually detect how much air is pushing down on them? Well, the simplest type really is called a Tori Chilian barometer, and it's named after its inventor, Evangelista Barometer. No, I'm sorry, wait, Evangelista Toricelli. He was an Italian physicist and a mathematician of the seventeen century. Torchelli is one of the mathematicians who laid the groundwork for integral calculus,

but I'm not going to hold that against him. Toward Chelli worked with Galileo who gave tor Chelly the idea that he should experiment with glass tubes and mercury to study things like vacuums as well as other physical properties. And this was in sixteen three. So tore Chelly took a tube that was four ft long or about one point two meters. It was sealed at one end, so

think of like a test tube. He filled it with liquid mercury, and then he overtook, turned the tube and put the end of it in a dish that had raised sides. So at first mercury started to come out of the tube and into the dish, but eventually the mercury settled and it was at a level above the dish. It was, you know, like several inches above where the base of the dish was above the level of the rest of the mercury. And he thought, uh, that's kind

of interesting. The mercury did not sink all the way down to the level of the dish. It actually remained up quite a bit. And the area behind the mercury at the top of the tube, so near the sealed end, that was a vacuum. He had created a vacuum in this way. There was nothing, no air in that part of the tube. Then he noted that the mercury's level would change day to day, and some days the mercury would actually end up being higher in the tube than it was the day before. So this meant the mercury

wasn't just leaking out. Right. If you kept coming out day after day and the mercury level is getting gradually lower every single day, your conclusion might be this mercury is very gradually leaking out of the tube into the dish. But if you come back one day and the mercury is actually higher up in the tube than it was the day before, something else has to be happening. It can't just be leaking out. So towards Shelly figured that atmospheric pressure was the reason for the changes in the

height of the mercury in the tube. On high pressure days, when the air is pressing down harder because there's essentially more dense air above you, then it ends up pressing down on the mercury in the dish, which forces mercury up the tube. Because again, the mercury that's in the tube with the vacuum in it, it's not being affected by the changes in air pressure. It's only the mercury

that's in the dish that gets that effect. On low pressure days, there's not as much air pressing down against the mercury in the dish, and so more of it starts to come out of the tube into the dish itself. So he never actually published his findings on this. Despite the fact that this was a really remarkable discovery, he

wasn't really concerned with it. He didn't think of it as being particularly important, particularly in regards with his interest in advancing mathematics toward Chilian barometers tend to use mercury instead of other liquids, but it's not because of temperature. Way thermometers are you don't necessarily worry about your barometer overheating or freezing. It's because mercury is more dense than water is, so you could create a barometer using water.

In fact, there were barometers that used water that predate the tour Chilian barometers. But the problem is that water's density is so much less that you need a much longer tube to be able to see that the changes. Otherwise you're gonna max out very early on because water is less dense, right, it doesn't take as much pressure to force water up a tube, so you have to have much longer tube in order to be able to

to see these changes. Uh and it means that it would be very difficult to take measurements, so mercury being more dense made more sense. Also, it was very easy to read it inside the barometer because again you're using clear glass. Mercury is a silvery liquid, so it was very easy to read the changes in the levels in

uh torre Chelian barometer. Now, at at sea level under normal circumstances, under one atmosphere of pressure, mercury would rise up to about the seventy six centimeter or thirty mark in a torrey Chellian barometer, and while changes in air pressure will be measurable with such a barometer, you wouldn't see dramatic differences in the height of the mercury unless you were to take the whole thing to a mountaintop

or something. So in other words, you could watch the mercury in the barometer and it could very accurately and with great precision, show you the changes in barometric pressure. But those changes wouldn't necessarily be really dramatic because you're talking about again the same altitude. Were you to take a mercury barometer at Death Valley and then magically transport yourself to Mount Everest, you would see a very dramatic

change in the height of the mercury in that barometer. Now, maybe you've seen barometers that haven't entical dial that either turns left or right along a semicircular scale that tells you what the barometric pressure is. So how did those work? Well, these are called aneroid barometers, and they're pretty clever. Inside of these, there's a sealed, air tight metal box, and

attached to that metal box is a spring. Now, when the air pressure is high, it compresses this metal box, and that ends up pulling on the string on the spring rather, which then creates the force necessary to move the dial, so it indicates a high pressure system is moved in and low pressure the little metal box. The air type metal metal box expands and this ends up pushing against the spring, which means that the dial will move toward the other side, showing a low pressure uh

system has moved in. So the dial turning to the left or right is all dependent upon whether or not this metal box is compressed or expanded. It's actually incredibly simple when you think about it, and thus I think a pretty elegant way of measuring air pressure. Mercury barometers are more accurate than aneroid barometers, but there's a disadvantage to mercury barometers, which is that stuff's poison. Y'all, mercury is toxic, so aneroid barometers are safer to have around.

They also are more portable, so you could put them on stuff like sailing ships and not have to worry about mercury spilling out all over the place because they were mechanical didn't depend upon mercury at all. But if you wanted something that had more precision and accuracy. You wanted a mercury barometer, not an aneroid barometer. Uh. However, we can go with microelectronics too. We don't have to use mercury or anneroid, although the microelectronics version uses a

very similar approach to aneroid barometers. So we do have barometric pressure sensors which rely on the piezo resistant of effect. Now, this is kind of similar to what we were talking about with thermistors, only in this case we're not talking about temperature. We're talking about stuff that's under pressure. Do you remember that has that same baseline as ice ice baby? Anyway, you may have heard about the piezo electric effect, right.

Piezo electric effect refers to the tendency of certain materials um certain types of crystals in particular like quartz, that when you put a mechanical stress on those materials, such as you mush mush mushed them up in some way, these materials would generate an electric charge or there's a reverse piece of electric effect. If you were to subject these materials to an applied electric field, they would produce a mechanical force like they vibrate and stuff. The crystal

courts are the quartz crystals. I guess I should say in old watches. That's why they're used. It's has this piezo electric effect. Piezo resistive materials are similar to that, except that, as you would expect, their electrical resistance changes as mechanical force applied to them changes. So typically you'd put this piezo resistive material around a hermetically sealed cavity

similar to what you'd find in an aneroid barometer. So you have this little area that you have hermetically sealed and it's lined with this piezo resistive material, and as the cavity reacts to changes in the air pressure, it places mechanical stresses on the piezo resistive material inside, and

that again changes its electrical resistance. A microprocessor will measure fluctuations in current passing through this piezo resistive material and then convert those changes in current into a digital signal that can be used to approximate pressure. All right, So now we've got temperature and air pressure out of the way, two of the big ones. But man, there's so much more. So I'm gonna try and summarize some of the other many sensors that are used by meteorological observation stations UH

today in order to gather information about the weather. But I am going to summarize because otherwise this episode is gonna last six hours long, and I got stuff to do, y'all. So let's get another basic measurement out of the way, and that would be wind, something I generate a lot of from multiple ends. As it turns out, you want to know where wind is coming from, and you want to know how strong the wind is because this will inform lots of other stuff like incoming changes to weather

such as storms, et cetera. For wind direction, we use something that's been around for hundreds of years, weather vain. We have lots of fancy ones today, but they all are still working on the same general principle. I mean, you can use more high tech ways to detect wind direction, but it's really not necessary. So whether veins typically consist of a counterweight on one end of a rotating UH peace on the weather vein. On the other end, you have some sort of thin that is covering a much

larger area than the counterweight is. So when wind blows, it hits against the fin the weather vein because this part of it can rotate freely along its axis. UH. In the horizontal plane, it will rotate so that the area being hit by the wind is facing away from the wind. The counterweight will point into the wind. Often the counterweight is in the form of an arrow, so it might be the point of an arrow, and the back may look like the fletching of an arrow, and

this tells you where the wind is coming from. So if you have a traditional weather vein and the arrow is pointing northeast, that tells you wind is coming from the northeast. It is not blowing to the northeast. It it's coming from the northeast. The counterweight is needed so that there's equal mass on either ends of this rotating part of the weather vein. But you also want to

make sure that there is an unequal area. In other words, the back half of the weather vein of that rotating piece needs to have more area to it so that the wind pushes it in the right way. You want more area on one side so that you can get that into the right position, and that indicates where the wind is coming from. UH. Once you look at where the counterweight is, wind direction can give you a general idea of what sort of weather you might encounter based

upon what's going on in that direction. So let's say that you are in Georgia, the state that's where I'm in, and you are in the winter, and you see that the wind is coming from the northwest. You're looking at a weather vein, it's pointing to the northwest. That's where winds are coming from. And you happen to know that there's a cold air mass that was moving down from

Canada through the United States. So you would say, well, based upon the fact that wind is coming from the northwest, that's the direction where if you were to go that way, you hit Canada. And I happen to know that there's a cold air mass moving down. I suspect that means that pretty soon our temperatures are going to drop further and we're going to get what is called the Devil's dan droff down here in Georgia. Thank you, Saturday Night Live.

Most people know it as snow. We know it as the stuff what shuts down our entire infrastructure at a given heartbeat. Anyway, that's why wind direction is important, because if you know what's going on elsewhere, then you can and you know that the wind is coming from that direction you, you can expect to get some of it yourself. By the way, if you hear that winds are let's say, northeasterly, that tells you where the winds are coming from, that

they're coming from the northeast. But if you hear the suffix ward ended at the end of a direction, that tells you the direction the winds are blowing toward, So an eastward wind or eastward wind if you prefer, I don't eastward wind that means winds are blowing to the east, and easterly wind means winds are blowing from the east, clear as mud, right. But wind direction is just one thing.

It's also useful to know wind speed. Now. Traditionally wind speed was measured in knots or nautical miles per hour. But was it not? I'll tell you, but not right now. I'll tell you after we take another quick break to thank our sponsors. So you want to measure wind speed wind speed, you would measure and knots. Knots stands for a nautical mile, although it's spelled like a knot like you would tie and a thread. A nautical mile is equal to one point one five miles per hour or

one point nine kilometers per hour. So if you hear there's a northeasterly wind blowing at fifteen knots, you know that the wind is coming from the northeast, and you know that it's blowing at seventeen point to five miles per hour or twenty eight and a half kilometers per hour. But how do you determine wind speed? How do you know how fast the wind is blowing? Well, meteorologists use

an instrument called an anemometer. The old anemometers had moving parts in them that made them sort of look like pin wheels. Uh. They had arms extending out from a hub, with each arm ending in a little cup to catch the wind, and then they would rotate along their axis

on the horizontal plane. So think of like a windmill, but on its side, so it's the fans are not standing up vertically, they're spread out horizontally and instead of it being ends or fins, some some of them were, but most of them ended with these cups that would catch the wind. So the wind would blow and force the hub to rotate along its axis in that horizontal plane. And then it just depended on the type of anemometer you're looking at. A lot of them worked in a

very similar way to an electric generator. So you might remember this from our discussion about the history of electricity. It's based on electro magnetism. If you have a magnet and you turn it and it's surrounded by a conductive material, or it's itself around a conductive material like a coil of insulated copper wire, it will induce current to flow through that conductor. Right. That's the basis of the electric generator.

So let's say you've got this spinning anemometer and it's turning a magnet around a conductive material, and this creates a current flowing through that duct of material. You have then a electronic circuit that's specifically designed to measure how much current has been produced and then converts that to a digital readout that indicates wind speed. So you calibrate it. You first have to calibrate this device so that it quote unquote knows how much current relates to which you

know what the wind speed is. But once you've calibrated it, that's how you can measure wind speed. You just look at how much electric current is generated in one of these devices. And now there are also other anemometers that do not use this approach, they don't resemble an electric generator in that way. They instead will count the number of rotations of the cups within a given amount of time in order to convert that to a wind speed.

So you might look at how many times per minute did this rotate based upon the you know, the speed of the wind, we're going to say that that means it's blah blah blah. Typically these anemometers have a simple switch that gets activated upon each rotation, and the switch makes a notation, and you just look at the number of notations per minute or per whatever unit of time you're using to measure, and you convert that information over to create the figure for miles per hour or knots

of wind speed. Or you could have a light sensor, so imagine an anemometer. It still is one of these pinwheel like devices. It's still spinning in the horizontal plane, but it has a little disc that can cover up a hole that otherwise leads down to a light sensor. When the anemometer rotates, this disc ends up being pulled away from the sensor so light can hit it, and then as it continues to rotate, it covers the sensor. Again, it just does the circular path where it is covering

and uncovering the sensor. Doing this, the sensor counts the number of times that light is hitting the sensor. It's very similar to that other electronic switch I was just talking about. And again it makes a notation. And again you just look at the number of notations per unit of time and you use that to convert it over two miles per hour for wind speed. But that's not all. You don't even have to have a rotating element at all to calculate wind speed these days, because you can

use what are called sonic anemometers. Now these are more or less the standard for a lot of observation uh points these days. For meteorological observations, they use ultrasonic signal emitters and receivers mounted at right angles to each other. So from the top it might look like a square. You have receivers and transmitters that are mounted in a square in relation to one another. And it's it's important

to remember sound as physical phenomenon, right. Sound is all about molecules bashing into each other, vibrations spreading across a medium.

So when we're hearing things, we're hearing the sense of hearing is all based upon air moving, vibrating at oscillating at the speed of whatever caused it to move in the first place, and it continues to make other air molecules do this same thing until some of them end up hitting the ear drums in our ears, which transfers this uh this vibration to some very tiny bones in our ears in our ears, which then transmit that vibration to the cochlea, which ultimately interprets this as sound. That's

how how we perceive it. But that means that sound itself is a physical thing, and you can affect it by changing things in the air, like if wind is blowing, it affects how sound travels. And if you've ever tried to talk to someone on a windy day, then you've probably experienced this at least a little bit. So the way these ultrasonic wind anemometers work is that they transmit signals at an ultrasonic frequency. It's too high for humans to hear, but the receivers can pick up on it.

When wind is blowing, it's going to affect the timing of when a transmission is sent out and when it gets received by the other side this timing. It's super super subtle. Uh, It's not like it's something that we humans could detect, but these instruments can detect it, and by detecting those changes, it can convert that into an

interpretation of how fast the wind is blowing. It's really the difference that the it took for the sound to get from its point of origin to its destination impaired against what it normally would take under still conditions with no wind present. If you ever look at technical readouts of wind speed and direction, you might notice that there's a lot of symbols that are used. Typically, you would

see what's called wind barbs. So first you start with two lines representing north, south and east west in a crosshair layout, you know, your typical north southeast west compass rose sort of thing. And from the center, you would have a line that would extend out towards the direction where wind is coming from. So let's say it's coming from the northwest. This line would extend out halfway between north and west. If it was truly coming from the northwest.

From that line, you might notice one or more short barbs, or even what is called a pennant. It looks like a little flag at the end of it. Those barbs actually represent wind speed, and the number of barbs on there tell you how strong the wind is blowing. So a line that has in short barb extending from it

indicates calm winds that approximately five knots. If you have the opposite, If you have a line with has a pennant on the end of it and two barbs extending from it out to the side, that would indicate very strong winds at like sixty five knots now. One of the simpler tools in the meteorological tool kit is the precipitation gauge. This is telling you how much precipitation has fallen over a given amount of time. Essentially, this comes down to a container designed to catch precipitation, so you

can see how much has fallen in that area. Now, your basic rain gage consists of a funnel which can capture precipitation falling precipitation. It has a measuring tube that the funnel feeds into, and the measuring tube itself tends to be fairly you know, not like maybe like an inch in diameter, maybe about eight inches long. Typically, um, it's it's a tube, it's open at the top and

close at the bottom. Then that itself is inside a larger collecting vessel, and the collecting vessel's mouth is the same diameter as the funnel that's at the very top, so the funnel prevents water from falling directly into the containment vessel. Instead it funnels the water into the tube

that's inside the collecting vessel. The tubes are calibrated to measure the amount of rainfall based upon the diameter of that collecting vessel's mouth, so each one is very specific to the collecting vessel, and the scale that you will see on these tubes has been written out to reflect

that collecting vessel. That's why if you pick up one of these measuring tubes that's using the big funnels, and you look on it and it looks like there's maybe six inches of water inside the tube, but the tube breeds that as b being uh, three quarters of an inch of rain. You're like, well, why is that? I mean, there's there's six inches of water inside the tube. How can it be three quarters of an inch of rain? It's because the funnel that's catching all that rain and

funneling it down into the tube. It's a larger diameter, it's got more surface area, so more rain is hitting that funnel than would have hit the tube just on its own. Now, the reason why you have the collecting vessel there is that sometimes you get more rain than

what the tube can handle. If the tube is calibrated so that it can hold up to one inch of rain compared based upon the size of the collecting vessel, it is in what happens if you get more than an inch of rain, well, water will start to overflow from the top of the tube and pour into the collecting vessel. When rain is done and you're wanting to see how much rain has fallen, you go out, you remove the funnel, you remove the tube, and you say,

all right, we start with one inch of rain. Because this tube is full, that means that it rained at least an inch. You pour that out it. Then you take the water from the collecting vessel, you pour that back into the tube, and you use that to measure how much in addition to one inch has fallen. So that's the reason for the collecting vessel and for the calibration.

Otherwise you would just have to keep building tubes that are taller and taller and taller, and you know, if they're narrow enough, it can give you an unrealistic account for how much rain has fallen. That's your basic rain gage. Um there there are other types of rain gages beyond the basic one. There's one called the tipping bucket rain gage.

I love these. If you've ever seen fountains where there's a small container that gradually fills up with water and when it fills up to the top, it tips over, dumping all the water out into the base of the fountain, and then it tips back up again because now the water's gone, that that counterweight has gone, so the bucket returns to its normal, uh normal orientation. That's exactly the

way these tipping bucket rain gage is work. They work as they have two buckets, typically on a see saw like device, so they swivel on the seesaw device as one fills up. It gets heavy enough once it reaches a certain point for it to tip, pouring its water out into a containment vessel below. The other bucket is then tilted upward to catch the water from that point forward until it fills up, and then it tilts again

and dumps its water. These buckets hold a very small amount of water, typically one an inch of rain essentially, and you have a switch that is connected to this see saw like device, and every time it tilts, the switches registers it. And because it registers that, it makes a mark on the device, makes a mark on a piece of paper or otherwise activates a counter, and that tells you that an inch of rain has just been

uh registered counted. And then you just add up all the different one one hundreds of an inch, and you can tell within a hundredth of an inch how much rain has fallen in that given area. There are even versions of this that are heated, where they have little heating elements, typically coils of wire that will heat up as current passes through them, and the purpose of that

is so that they can measure frozen precipitation. As frozen precipitation hits the buckets, that heats up, it converts into water, and also it prevents the gauge itself from freezing over in cold weather, something that we don't get a whole lot of in my neck of the woods, but I think it's a pretty cool way of measuring rainfall. There are other types of rain gages. They are weighing gauges. Weight gauges you could say that estimate the amount of

rain based upon the weight change. And UH, those are the major types of precipitation gauges. Then you have various devices that can detect electrical storms. UH. Basically what you need is an antenna. Meteorologists use thunderstorm detectors. That our antenna that registers spikes of electro magnetic radiation or lightning strikes.

So if you've ever listened to a m radio during a thunderstorm, you may have noticed that there's this burst of static whenever there's a lightning strike somewhere in the area. Thunderstorm detectors pick up electrical discharges, typically within a couple of hundred miles of the detectors, so it doesn't have to be that close in order to pick up on it. It just is this little spike of electric discharge that the antenna can pick up on. Your basic system is

a simple receiver. There's no transmitter, it just records it and using several of those sensors across the region will help you determine where and not just when lightning strikes. You can use triangulation using three or well really three points to figure out where did the lightning strike These three different detectors picked it up. Based on the timing of the three detections, we can say the lightning strike must have happened at these coordinates. It's a very simple

way of doing it. But there are also mobile lightning detectors that you are typically would put in an aircraft. You fly a plane around a weather plane and look for these electra electric discharges. These will use attenuation signal attenuation to determine the location of lightning strikes, but that is not um not always accurate because it is dependent upon some other factors that can confound the device. But yeah,

there's a couple of different ways of doing it. Other devices that meteorologists might use might include paranometers or pyranometers if you prefer um pyra to indicate heat that measures a solar radiation actually or how much sun exposure a

place will receive over a given amount of time. These are also used not just for weather forecasts, but also when you want to figure out the best place to locate, say a solar panel farm, you might use a pyranometer to see how much solar radiation that area actually receives. Does it make sense to put a solar panel farm there or are you not going to maximize your efficiency

if you place it there. Typically, they measure sun exposure by using thermopiles, which are sensors that generate electricity as they heat up from absorbing light. So these things absorb lots of light. They're very dark, they tend to be black. Absorbed light, they generate electricity, and then by measuring the electricity you understand how much sun exposure you got. Then there are devices called celo meters which are used to measure clouds, like the ceiling spelled like that. Celo meters

they can measure cloud height and thickness. And the one I was looking at specifically does this in a pretty cool way. It shoots lasers at clouds. So the lasers hit the clouds, and then the laser lights starts to scatter as it encounters the various particles that are in clouds like water, vapor, and that kind of stuff. And then you use backscatter technology to measure that dispersal of light within the clouds. So you fire laser into the cloud.

The particles in the cloud cause the laser light to scatter at different levels depending upon the density and composition of those particles, and you measure that backscattered light to allow you to define the parameters of the cloud and it's density, and even be able to tell whether or not precipitation is likely to fall because of those clouds. You've also got visibility sensors that can measure how transparent the air is, which might sound kind of silly until

you remember that fog is totally a thing. Uh So these devices measure light attenuation and use backscatter technologies similar to the cealimeters I just talked about in order to measure visibility. So you can use optical sensors to measure of visibility as well, in other words, like cameras and stuff, and you can get a firsthand look at visibility, but these are looking at it on a more precise level than just does it look clear out there. Oh and

Doppler radar. I can't finish this episode without talking about Doppler radar. If you've watched a weather report, you've likely heard this term bandied about when it comes to measuring rainstorm systems and their movements. So Doppler radar measures not just the presence of something but it's movement either toward or away from the radar station. So your basic radar is pretty simple. You beam out a signal in a direction.

That signal encounters other stuff and bounces off of it, some of it coming back to you, and it gets to the starting location. If you look at the time between when you sent the signal out and when the signal came back, you can then use that to extrapolate how far away that thing is. And if it's stationary, then you're not going to see any difference in the frequency of the signal coming back as the one you

sent out. Right, it should be pretty much identical, and you'd say, all right, there's a stationary object that's ten miles away. Godzilla is taking a nap now. Uh. You can do this, by the way, because those radio waves, the radar waves, are traveling at the speed of light right there. It's a constant speed, so you don't have to worry about anything else. You just say, well, I know how fast light travels. I know how long it took the returning wave to get back to me. Because

I've got a timer. It tells me that this much time passed between transmission and receiving the echo. Then I can say, well, how far away, how far did the signal have to travel in order to get there and back? Um, And that will tell you how far away the object is. But you can also tell if the object is moving toward you or away from you, because the signals coming

back to you will be affected by this. Uh. It is the Doppler effect, something that is pretty easy to encounter out in the real world, just on your own. If you've ever heard a like a police car blaring a siren and the police cars coming toward you and then it passes you, you've probably noticed that the sound of the siren changed as the police car passed you. So when the cars coming towards you, it's actually compressing

those sound waves that are emitted by the siren. Uh. And because it's compressing the sound waves, UH, it means that it's increasing the frequency. It's like physically compressing, not digitally compressing. We're talking about physically compressing those waves so that the frequency is increased. That means the pitch goes up, so you hear a higher pitch noise as the cars

moving away. It's along gating those sound waves, it's stretching them out which means decreasing the frequency and us decreasing the pitch, making it a lower pitch. Well, radar has the same shifts with its signals if it hits something that's moving toward it. So you've got a radar station shoots out a little radar beam. The radar beam comes back and the frequency is shorter like it's it's the or the frequency has been increased. The wavelength is shorter.

That tells you that the objects moving towards you. It is compressed the length the wavelength of that of that signal. If the wavelength is longer, tells you that the object is moving away from you. That the frequency has hit the object, but the objects moving away and thus the returning waves have an elongated wavelength. So Doppler radar is very cool and that it can tell you where a storm system is and whether or not it's moving toward

you or away from you. Um and you use multiple observation stations with Doppler radar to get a full picture where is the system going. You can even detect things like precipitation using Doppler radar. The cool thing about Dopper radar in my mind is that it is a lazy,

lazy worker. And by that I mean in a typical hour, the Doppler radar is actively sending out signals for about seven seconds total out of the entire hour, not all at once, but if you look at an entire hour and you measure the amount of time that the Dopper radar was sending signals out, it averages to about seven seconds.

It's spending the other fifty nine minutes and fifty three seconds listening, Which seems like a great job to me to work for seven seconds and then just listen for fifty nine three seconds, But I guess it depends on what you're listening to. In this case, it's listening for those echoes, those returning frequencies for the rest of the hour. And again not all in one batch, it's spread throughout the hour, but collectively we're talking fifty nine minutes fifty

three seconds of listening. Uh. It is scanning the sky in a series of angles. So think of a Doppler radar like a think of like a little satellite dish, and you start off at say forty five degree angle pointed at the sky, and then you move it up to seven degrees for the next scan and so on and so forth, and it's also rotating, so it's taking scans of different levels of elevation. In fact, they call these elevation slices in the Doppler radar gang and UH

it gives the system a volume coverage pattern or VCP. Now, once it goes through all of its elevation slices and all of its rotations, it completes one volume scan, and it does this every five minutes or so during precipitation mode scans. The data that comes back can then be interpreted to build out a picture of weather systems where they are moving and how quickly they are moving. And that even includes active precipitation, which is pretty cool stuff.

I think now that wraps up the basic tools and sensors that meteorologists used. Not keep in mind, there are other ones I didn't talk about, satellites, which are very important. Satellites also gather a lot of data about active weather systems on the Earth, and that data gets fed into computer models or weather forecasts. UH. There are other sensors and tools as well, but I wanted to cover the ones that you would find at your typical observation station

on the ground. Keeping in mind there are others, So I hope I've given you an indication of how complex this is just from the types of information alone that have to be gathered and poured into these computer models

so that we can get accurate weather forecasts. In our next episode, I'll talk specifically about how those weather models are constructed and how they generate forecasts, and the different models that are out there, and why are there different models, and what are the computer systems that are actually running these simulations, why are we using supercomputers? And will we ever get a computer model so comprehensive that we'll never be surprised by the weather again. Well, we'll talk about

that in our next episode. Guys, if you have suggestions for future episodes of tech Stuff, let me know, would you. You can send me an email. The address is tech stuff at how stuff works dot com, or you can drop me a line using Twitter or Facebook. The handle of both of those is text stuff h s W. You can also drop by Twitch dot tv slash tech stuff, and you can watch me record these shows live like the audience I have watching me right now. I record

on Wednesdays and Friday's. Just go to Twitch dot tv slash tech Stuff and you can see the schedule there and I'll talk to you guys again really soon for more on this and thousands of other topics, because it how staff works. Dot com