Read...

Perhaps you've heard the saying, "There are three kinds of lies: lies, damned lies, and statistics [38]." There's no doubt that statistics can be used to deceive, or that some statistics can be misinterpreted, but statistics are a huge part of a meteorologist's toolbox. In fact, they're often used in weather forecasts that you probably see all the time!

For example, one statistic that you've probably seen is the probability (or chance) of precipitation. If you hear that there's a 40 percent chance of rain tomorrow, do you know what that actually means? Study after study shows that most people don't! The probability of rain doesn't tell us anything about how long it will rain, how heavy the rain will be, or what percentage of a given area will get rain (those are all common misconceptions).

A 40 percent chance of rain tomorrow means there's a four in ten chance that any point (your backyard, perhaps) in a forecast area will receive at least 0.01 inches of rain tomorrow. Alternatively, if the same forecast scenario occurred ten times, at least 0.01 inches of rain would fall on four days at any point in the forecast area, and no measurable rain would fall in the forecast area the other six days.

Some statistics that meteorologists use are a little more straightforward than probability of precipitation. For example, the observations that you learned about that go into the station model can be tracked over a long period of time to determine the long-term averages and extremes (highest and lowest values) at a location. These long-term averages and extremes constitute a location's "climate" and meteorologists often refer to them as "climatology." The National Weather Service is the keeper of many weather records in the U.S., and each day, each office of the National Weather Service issues a "climate report," which summarizes the previous day's temperature and precipitation and compares them to the climate record. Below is a sample display of temperature data in an NWS Climate Report from February 14, 2012 at Chicago's O'Hare Airport.

A portion of the official climate report issued by the National Weather Service for February 14, 2012. Feel free to investigate climate reports [39] at or close to your location.

Credit: NWS

Climatology is extremely useful for weather forecasters because it gives an expectation (based on history) of types of weather that are typical at the location. While climate information runs the gamut from temperatures to dew points, to winds, to precipitation, etc., here we'll focus our discussion on temperatures since temperature statistics are cited so commonly. We'll start with so-called "normal" temperatures.

There's Nothing Normal about Normal Temperatures

One of the statistics in the climate report above is the ”normal value“ of maximum, minimum, and (daily) average temperature [40]. What does that mean? Well, "normal" temperatures are based on perhaps the most often used statistical measure in meteorology (and in general)--the mean (or "average"). The mean is simply the sum of a set of observations, divided by the number of observations. Meteorologists average lots of things, such as temperatures over the course of a day, a week, a month, a year, or even longer.

So, how did the NWS come up with 35 degrees Fahrenheit for the normal maximum temperature in the report above? They calculated the mean of all of the high temperatures for February 14 over a 30-year period in Chicago. Similarly, the "normal minimum temperature" of 20 degrees Fahrenheit is just the mean of the minimum temperatures for February 14 over a 30-year period. So, normal temperatures are 30-year means (averages). In case you're wondering how to interpret the "normal average" temperature of 27 degrees Fahrenheit, it's the 30-year mean of the daily average temperature (the average of the daily high and low temperatures for February 14 at Chicago).

Note that in Chicago's climate report above, the 30-year means ("NORMAL VALUE") came from weather observations taken during the period from 1981 to 2010 [41]. Every ten years, the averaging period for calculating normals changes. We started using 1981-2010 in the year 2011, but in 2021 the averaging period became 1991-2020. The switch to "new normals" every ten years usually results in normal temperatures that are a little different than previous normals. Moreover, the common use of the term "normal" is unfortunate because weather seldom behaves exactly in a "normal" way. In winter, for example, a season renowned for occasionally abrupt temperature swings, it sometimes turns out that a city's normal high for the date never actually occurred as a high temperature on the date in question during the previous 30-year period! So, there's nothing "normal" about "normal" temperatures!

Still, normal temperatures are useful for meteorologists because on most days, temperatures don't deviate all that much from normal. Indeed, a forecaster should be cautious about forecasting a high temperature that's 25 degrees above the normal high for the date. Such large departures from normal are possible, but they may only occur a few times each year.

Record Temperatures and the Climate Period of Record

The climate period of record refers to the length of time that weather observations have been taken for a given city. For example, the ”CLIMATE RECORD PERIOD“ in Chicago's climate report [42] indicates that weather records in Chicago started in 1871; however, the period of record varies from city to city. Some places in the United States have reliable weather records dating back to the late 1800s, while others only date back to the 1930s to 1950s (or even more recently). Keep the idea of a "climate period of record" in mind whenever you hear someone tout an extreme, or record value (the highest or lowest temperature observed in the period of record), for a given day as "the highest (or lowest) temperature ever" for the date. What they really mean is that it's the highest or lowest temperature for the date in the period of record. So, the record high and record low for February 14 in Chicago [43] represent the highest and lowest temperatures observed on February 14 since 1871. Before records began, we have nothing reliable to compare to.

The Blue Hill Meteorological Observatory in Milton, Massachusetts is the home of the longest continuous set of weather records in the U.S.

Credit: Great Blue Hill Weather Station [44] / Jameslwoodward [45] / CC-BY- 3.0 [46]

Furthermore, if you hear about a record high (or low) temperature being broken, it's always worth asking "what's the period of record?" A record high at a station that has only been keeping records for 30 or 40 years might not be a big deal. In other words, it might be setting a record only because they've been keeping weather observations for such a short period of time. It takes more extreme weather events to break records at locations that have a period of record 100 years or longer, such as the Blue Hill Meteorological Observatory (pictured on the right) in Milton, Massachusetts. The Blue Hill Observatory has the longest single, continuous set of weather records in the U.S. (going back to 1885). The location for taking weather observations in most cities (like Chicago) has changed at least once since the late 1800s (if they've been taking observations that long).

Temperature Ranges

Another statistic that meteorologists commonly use is the range of a set of observations, which is simply the difference between the maximum observation and minimum observation. Meteorologists often look at the range of temperatures over the course of a single day (called the "diurnal range"), or the range of temperatures over the course of a month, year, etc.

For example, take this table of climatological data for State College, Pennsylvania from February, 2015 [47], which was the coldest February on record at Penn State, where weather records began in 1893. The highest temperature measured during the month was 44 degrees Fahrenheit (in red in the "Max Temperature" column). Meanwhile the lowest temperature measured during the month was -8 degrees Fahrenheit (in blue in the "Min Temperature" column). So, during February 2015 temperatures in State College ranged from 44 degrees Fahrenheit to -8 degrees Fahrenheit. If we take the difference between the two, that gives us 44 degrees Fahrenheit - (-8 degrees Fahrenheit) = 52 degrees Fahrenheit. So, February 2015 had a 52 degree Fahrenheit temperature range in State College. If you wanted to find the range of monthly temperatures for all Februaries in State College, you would take the single highest temperature on record from all Februaries and subtract the single lowest temperature on record for all Februaries (and that would give you a much larger range than the range for February 2015).

These statistics (means, normals, and ranges) are pretty straightforward, but make sure you remember what each of them is telling you. Meteorologists use them a lot, and they'll certainly come up again in this course!

Now we're going to shift gears a bit. In the past few sections, we've talked about weather observations at a single location, but meteorologists need to observe the weather over large areas, too, and that requires us to talk a bit about maps and map projections. We'll start "mapping things out" in the next section.

Read...

Meteorologists are always looking at maps! But, anyone who uses maps regularly can tell you that all maps are not created equal. For starters, (most) maps are flat, and while that's very convenient (far more convenient than carrying around a globe to simulate the spherical Earth), there's a trade-off for using flat maps. To see what I mean, think about unraveling the skin of an orange in one piece and then trying to make it lie flat without breaking or distorting it. Good luck! Trying to create a flat map of our spherical Earth brings the same problems: No matter how you do it, you'll always have distortions.

A number of imperfect techniques (called map projections) exist for making flat maps of Earth. Each projection has its own strengths and weaknesses, and it's important for you to know what those are in order to analzye weather maps that you see. We're going to focus on two common projections here: polar stereographic projections and mercator projections.

Polar Stereographic Projections

On a polar stereographic projection, the observer's perspective comes from looking down at either the north or south pole. This unique vantage point allows operational weather forecasters to watch, for example, the movements of weather system over long distances. But, there's a catch (of course). Latitude lines, which run west-east (paralleling the equator) and mark the distance from the equator (0 degrees latitude), are not straight on polar stereographic projections. Instead, latitude lines appear as large circles (or arcs, if you're not looking at the entire hemisphere). Meanwhile, longitude lines (also called "meridians"), which run north-south and mark the distance from the Prime Meridian (0 degrees longitude), are straight, and they all converge at the poles. To see what I mean, check out the polar stereographic projection below.

A polar stereographic projection focused on North America. Latitude lines appear as arcs of large circles, while longitude lines are straight, and converge at the North Pole. The arrow off the west coast of North America represents a westerly wind (the wind direction is 270 degrees).

Credit: David Babb @ Penn State is licensed under CC-BY-NC-4.0

So, on polar stereographic projections, finding east and west isn't simply a matter of looking to the right or the left. Similarly, finding north and south isn't as easy as looking up or down on the image. Those directions will change orientation on the map depending on what area of the map you're looking at. For example, the arrow off the West Coast of the United States in the image above might look like it's blowing from the northwest, but if you look closely, it's blowing parallel to the nearest latitude arc. That means the wind is actually blowing from due west (270 degrees).

For a real-life example, take a look at this map including station models near Alaska from 06Z on December 1, 2011 [48]. I've outlined the weather station on tiny St. Paul's Island. Knowing that this is a polar stereographic projection, can you determine the wind direction? It might look like it's from the southeast at first, but it's really from due east (90 degrees). If you look carefully, the wind is blowing parallel to the nearest latitude arc. The bottom line is simply that you must orient yourself to the local latitude arcs and longitude lines on polar stereographic projections. Latitude arcs run west-east and longitude lines run north-south (and they intersect at right angles). Without orienting yourself, you can't properly determine things like wind direction. To further illustrate this issue, I created a short video below (2:37) that shows how the same wind direction can appear very differently on station models in different areas of a polar stereographic maps.

There's one other big issue with polar stereographic maps--size and distance distortion. Note in the idealized polar stereographic map shown above that Alaska and Texas look to be about the same size. But, in reality, Alaska is more than twice the size of Texas! In fact, the farther you get away from the pole on a stereographic projection, the more signifcant the size and distance distortion, and that causes some issues when analyzing weather features closer to the equator. For analyzing weather features closer to the equator, meteorologists often turn to mercator projections.

Mercator Projections

Mercator projections are routinely used for weather maps and other imagery that focus on low latitudes (near the equator). This type of projection minimizes distortion near the equator, which makes it better for tracking weather systems at low latitudes. The closer you are to the equator on a mercator projection, the more accurate the distance representation.

Another benefit of the mercator projection is that compass directions are preserved. That's a huge benefit when tracking weather features over some distance. For example, north is always oriented in the same direction when looking at the image, regardless of where you are on the map. So, typically, north, south, east, and west are easy to find on mercator projections. You should still find latitude and longitude lines to orient yourself, but because each is a straight line, the directions are more intuitive, as shown in the sample mercator map below. North is at the top of the image, south is at the bottom, west is on the left, and east is on the right (regardless of location on the image).

Latitude and longitude lines are straight on mercator projections, preserving compass directions as weather systems move. However, while size distortion is minimized at low latitudes, sizes are greatly exaggerated closer to the poles.

Credit: David Babb @ Penn State is licensed under CC-BY-NC-4.0

Since compass directions are preserved and distances at low latitudes are depicted with reasonable accuracy, why don't meteorologists just use mercator projections all the time? Well, size and distance distortion are significant farther away from the equator. As weather systems move north or south, that can be a problem because they might appear to change size, when in reality the changes are just an artifact of the map projection. On the mercator map above, for example, Alaska absolutely dwarfs Texas (far more than in reality). At even higher latitudes, Greenland looks to be about the size of Africa, but in reality, Africa is more than 13 times larger than Greenland. At the extreme, the North and South Poles (single points, in reality) appear as straight lines at the top and bottom of Mercator maps. Now that's distortion!

Ultimately, no map projection is perfect, and meteorologists use the strengths of each type to their advantage. But, they must be familiar with the quirks of each type of projection, or else they could easily draw an incorrect conclusion about the direction, distance moved, or the size of a weather feature. Most often, it seems that students are a bit less familiar with polar stereographic projections, so to help you get comfortable with working them, check out the Key Skill and Quiz Yourself boxes below so that you can further explore the interpretation of these maps, and practice with determining wind direction on them.

Read...

Now that we've covered some basics about maps, let's start investigating some ways that meteorologists use maps! To begin with, if you are an avid hiker or skier, you've probably consulted a topographical map [49] for insights into elevation and the "grade" (steepness) of slopes and hiking trails. Believe it or not, deciphering topographical maps actually lends itself to the process of interpreting weather maps.

To get you started, let's look at the Big Island of Hawaii.  Just in case you're not familiar with the topography of Hawaii, let's take a virtual fly-by of the Big Island [50] to get a sense of the dramatically changing elevation on the Big Island. As you might have guessed, elevation changes rapidly, varying from sea level to the volcanic summits of Mauna Loa and Mauna Kea at 13,452 feet and 13,796 feet (respectively) in a relatively short distance.

Now that you have a sense for the wildly varying elevations on Hawaii, focus your attention on a topographical map of the Big Island (below). Such a two-dimensional map represents a "plan" or "top-down" view."

A topographical map of Hawaii (right) gives you a representation of elevation. The more contour lines, the greater the change in elevation. Note the two volcanic peaks on the contour map and compare them with the image of Hawaii taken from space (left).

Credits: (left) MODIS-NASA; (right) David Babb @ Penn State is licensed under CC-BY-NC-4.0

Each contour on the map is an isopleth, which is a contour connecting points of equal value ("iso" translates to "equal" and "pleth" means "value"). In this case, the isopleths connect points of equal elevation above sea level, and are drawn for every 1,000 feet of elevation above sea level. I point out, however, that such a choice is strictly arbitrary. Isopleths could have been drawn for every 250 feet above sea level, meaning that there would be four times the number of isopleths on the map, probably making it look a little cluttered. The arbitrary choice of 1000 feet for the "contour interval" was a trade-off between the look of the map and its detail.

To explain what I mean by "detail," consider the southernmost tip of the island. Suppose I asked you to locate the point precisely due north of the southernmost tip that has an elevation of 400 feet. With isopleths drawn every 1000 feet, you must visually estimate the distance from the southernmost tip to the 1000-foot contour and then pick a reasonable intermediate spot to place the point. Had contours been drawn every 250 feet instead of every 1,000 feet (the cluttered look), your task would have been a little easier and your answer would likely have been more accurate. Just in case you're having difficulty making the connection between a three-dimensional mountain and a two-dimensional topographical map, I've created a short video (2:00) showing a virtual representation of Hawaii's topography [51] (video transcript [52]) to help you to better visualize what I'm talking about.

Zooming in on a topographical map of Mauna Kea.

Credit: David Babb @ Penn State is licensed under CC-BY-NC-4.0

Looking at this virtual representation of Hawaii in the video, you might get the mistaken notion that all isopleths are closed curves or circles. Depending on the breadth of the map, however, some contours could conceivably extend from one side of the map to another without making a closed loop. Indeed, zooming in on the topographical map of the Big Island and restricting the focus just to Mauna Kea (see image on the right), some elevation contours now simply end at a boundary. We'll see this quite often when we start looking at contoured maps of weather variables.

The great thing about contour maps like our elevation map of Hawaii is that it allows us to see the elevation pattern relatively easily (much more easily than if we just plotted individual elevation numbers at random points). And, that's why meteorologists use contour maps regularly: contour maps help meteorologists easily find patterns in weather data. But, meteorologists must also be able to look at a contour map and, using the pattern of contours, estimate a value at any point. That's our next task at hand. Read on.

Read...

Topographic map of Hawaii with point "P" labeled. Contours are drawn every 1,000 feet.

Credit: David Babb @ Penn State is licensed under CC-BY-NC-4.0

Meteorologists regularly use contour maps to see how weather variables (temperature or pressure, for example) change over large areas, but they also use them to estimate values of weather variables at individual points. To get us started on interpreting contour maps, we're first going to revisit our topographic map of Hawaii (right).

How would we go about figuring out the elevation of the point marked "P"? First, we must identify the two contours that lie on either side of "P." In some cases the contours that we need are clearly labeled; however, in other instances, you will need to use the contour interval (1,000 feet, in this case) to "count" up or down from a labeled contour. On our map, point "P" lies between 3000 feet and 4000 feet contours, so the elevation at point "P" is greater than (but not equal to) 3000 feet but less than (but again, not equal to) 4000 feet. We know that point "P" is not equal to either of these values because it does not lie on one of the contour lines.

Technically, this is all that we can say for sure about the elevation at point "P" -- it can be anywhere between 3,000 and 4,000 feet. However, often instead of giving a range of elevation, we would prefer a single number. In such cases, we can interpolate (make an estimate by assuming the elevation changes linearly) between the two known contours. In this case, we can see that point "P" is about half-way between the two contours and thus has an elevation of approximately 3,500 feet. Using interpolation usually provides us with a sufficient approximation, but we must always realize that it's only an estimate.

What about closed contours that have no other contours drawn within them? Look again at the contour map and focus your attention on the southern-most peak, Mauna Loa. How might we estimate the height of this peak? Clearly, we do not have two contours to use... or do we? The last drawn contour is 13,000 feet. This means that the peak is higher than 13,000 feet. However, we know that the summit of Mauna Loa is not higher than 14,000 feet. Why? Well, if it were, then the 14,000 feet contour would have been drawn around the summit. So, for an inner-most closed contour, the range is always between the last drawn contour and the next undrawn contour. You should also note that interpolation cannot be performed in such cases because we have no way of knowing how far away the undrawn contour is. The summit of Mauna Loa is 13,452 ft (within our deduced range of 13,000-14,000 feet), but we really couldn't have known that just from our topographic map (unless the peak elevation was labeled).

Two variables that are commonly contoured by meteorologists are temperature and air pressure. Isopleths of temperature are called isotherms (contours of constant temperature), and isopleths of pressure are called isobars (contours of constant pressure). Don't confuse the two! It would be incorrect to look at a map that shows only contours of constant temperature and call them "isobars," for example.

Maps with isobars of sea-level pressure help meteorologists recognize areas of high and low pressure. Why is this important? Well, generally speaking, areas of high pressure at the surface are often areas of "fair" weather (relatively calm, with at least some sunshine). Meanwhile, regions of low pressure are often somewhat unsettled (cloudy with precipitation). To see a map with isobars of sea-level pressure, check out the image below, from 15Z on May 1, 2017.

Analysis of sea-level isobars (contours of constant pressure) at 15Z on May 1, 2017.

Credit: University of Illinois

The H's mark centers of high pressure (highest pressure in the region), while the L's mark centers of low pressure (lowest pressure in the region). The feature that would immediately jump out at a weather forecaster as a region of "active" weather is the large area of low pressure centered over the Midwest (near the Iowa / Wisconsin border).

The first step in interpreting values from this image is to figure out the contour interval, which is four millibars (millibars is a standard unit of pressure). Knowing that, we can estimate the pressure near the center of the low over the Midwest. The inner-most closed isobar is 996 millibars (it's unlabeled, but it's inside of the 1000 millibar isobar and values are decreasing toward the center of the low), so the pressure must be less than 996 millibars. However, there's no 992 millibar isobar drawn, so the pressure must be greater than 992 millibars.

Similarly, meteorologists contour maps of temperature because such maps allow them to easily spot areas of cold and warm air. But, maps of isotherms are often color-coded to help make the interpretation easier (for what it's worth, colors are sometimes used on topographic maps [53], too). I caution you that color schemes and contour intervals can vary widely on images you may see on television or online, and determining which color is which can sometimes be a real challenge. But, if you understand how to interpret contour maps, you can keep your wits about you and use the isotherms (in conjunction with the colors, if you'd like) to gather useful information.

For example, check out the map of isotherms at 15Z on May 1, 2017, below. The contour interval on the image is 5 degrees Fahrenheit. Given that, what would you say is an appropriate range in temperature for somewhere in Indiana? Given that the 60 degree Fahrenheit isotherm and 55 degree Fahrenheit isotherm nearly match the eastern and western borders of the state (respectively), it's fair to say that temperatures in Indiana were greater than 55 degrees Fahrenheit, but less than 60 degrees Fahrenheit. And, if you look at the color key on the right of the image, the shade of yellow-green (or is it green-yellow?) over Indiana matches the color used for temperatures between 55 and 60 degrees Fahrenheit (although some might find it hard to tell).

Analysis of surface isotherms (contours of constant temperature) at 15Z on May 1, 2017.

Credit: University of Illinois

If we wanted to pick a specific point, "P," on the map, near Indianapolis, Indiana [54], and estimate a specific temperature, we could do that, too. Point "P" is approximately half-way between the 55 and 60 degree Fahrenheit isotherms, so the temperature at "P" is about half-way between 55 and 60 degrees Fahrenheit. We'll call it 57 or 58 degrees Fahrenheit, knowing that we're just making our best estimate based on the isotherms.

I should also point out that sometimes you won't even see isotherms plotted, and instead, a temperature map will just be shaded. In reality, every boundary where colors change is an isotherm (even if it's not drawn or labeled), and sometimes very small contour intervals are used. On such maps, like this analysis of temperatures from 17Z on May 24, 2017 [55]. Using the color key is the only way to go, and even then (because colors change every degree in this case), estimating a specific temperature at a point can be difficult when the shadings are so similar.

To see a few more examples of interpreting values from a contour map, watch the short video (3:28) below.

Finally, before you continue on, make sure to check the Key Skill and Quiz Yourself sections below. They'll help you make sure that you can interpret contour maps on your own, which is a useful skill outside of meteorology, too (elevation and many other variables can be isoplethed, and the principle for interpreting the values is the same, regardless of what's being plotted).

Read...

Besides allowing meteorologists to estimate values of weather variables at specific points, contour maps are also useful tools that help meteorologists to see patterns in the data. In other words, contour maps make it easy for meteorologists to see how a weather variable (like temperature or pressure) is changing over a large area. Maps of isotherms allow meteorologists to identify regions of warm and cold air, as well as how temperatures change over certain areas. Similarly, maps of isobars allow meteorologists to find areas of high and low pressure, and see how pressure changes over certain areas.

The change in a variable over a certain distance is called the gradient, and gradients are very important in meteorology. Zones where weather variables have large changes are often zones of active weather, so meteorologists like to keep tabs on areas with so-called "large gradients." Tuck this idea away, as the importance of gradients will come back again and again in our studies of meteorology.

Signs warn motorists of steep grades on mountain roads. For a road to have a steep grade (in other words, a large gradient), the elevation of the road must change relatively rapidly over a short traveling distance.

Credit: Public Domain

But, to start us off in our discussions of gradients, let's return to the example of elevation. Driving to the summit of the beautiful mountains in central Pennsylvania, motorists are greeted by signs that warn of "steep grades" ahead, which refers to large gradients. What does that mean in practical terms? For the mountain road to have a large gradient, the elevation of the road must change relatively rapidly over a short traveling distance. In other words, you're driving on a steep hill. Similarly, a small gradient means that the elevation of the road changes very little with distance (the road is relatively flat).

How do we distinguish between a large and small gradient on a topographical map? Let's return to the video showing the virtual topography of Hawaii [51] (video transcript [52]) you saw earlier. Remember that the contours of constant elevation are packed close together in areas where the elevation changes rapidly over short distances (like near the mountain summits). Thus, a large gradient on a topographical map, which marks a large change in elevation over a relatively short horizontal distance (steep terrain), corresponds to a tight packing of contours.

Meanwhile, the "packing" of contours of constant elevation is rather loose where the terrain is flatter. In other words, the gradient is small. Thus, a small gradient, which marks little change in elevation over a relatively short horizontal distance (fairly flat terrain), corresponds to a loose packing of contours.

Now let's take what we've learned from topographical maps and generalize it to weather maps. Check out the 06Z analysis of sea-level pressure on March 15, 2017 (below), which shows a powerful low-pressure system along the New England Coast. Note the tight packing of isobars over the northeastern U.S. and eastern Canada, indicating a large pressure gradient over this region (relatively large pressure changes over a short distance). As you'll learn in later in the course, tight packings of isobars (large pressure gradients) correspond to strong surface winds. This case was no exception! This area of low pressure was dubbed "The Blizzard of 2017 [59]" and winds along the coast in Massachusetts gusted to more than 70 miles per hour.

The 06Z analysis of sea-level pressure on March 15, 2017 showed a powerful low-pressure system along the New England Coast, surrounded by a large pressure gradient. The large pressure gradient helped make it quite windy during the so-called "Blizzard of 2017."

Credit: University of Illinois

On the flip side, for an example of a small pressure gradient, look at the western United States on the map above. There's not a single isobar in California or Nevada, and overall isobars are packed very loosely in the West, so pressure changes very little over the entire region.

Turning our attention to temperature, tightly packed isotherms represent large horizontal changes in temperature over a relatively short distance (that is, a large temperature gradient). For example, the analysis of surface isotherms from 19Z on May 18, 2017 [60] shows a very large temperature gradient (tight packing of isotherms) stretching all the way from the Southwestern U.S. to the Great Lakes and into Canada. Such a large temperature gradient marks a contrast between very warm air and colder air. For example, check out the change in temperature just in the state of Kansas (outlined in the black box [61]). In the southeast corner of the state, temperatures were in the 80s. In the northwest corner, temperatures were in the 40s! As we will learn in later lessons, large temperature gradients often indicate the presence of cold and warm fronts, which you've probably heard weathercasters talk about before (and that was certainly the case here)

On the other hand, if you want to see an example of a small temperature gradient, look no further than the Gulf Coast States. The packing if isotherms was very loose, and temperatures didn't change very much with distance. In fact, temperatures over a several state area were mainly between 85 and 90 degrees.

To see another example of interpreting temperature gradients, check out the short video (1:40) showing some large and small temperature gradients in the winter time!

One quick note: When talking about gradients, be careful that you use the proper terminology. For example, if temperature changes rapidly over a short distance, you might refer to this as a "strong," "tight," "steep," or "large" gradient. You wouldn't, however, say that the "temperature gradient is changing rapidly." Doing so means that the change in temperature is changing. While this is certainly possible, it's not how we talk about a single gradient, because the gradient itself IS the change over a certain distance.

The bottom line is that meteorologists really pay attention to gradients in temperature and pressure (among other variables), mainly because large gradients can be zones of very "interesting" weather! Before wrapping up the lesson, make sure you try your hand at the Key Skill section below, to make sure you can "size up" gradients.