Next Generation Science Standards:
- MS-ESS3-5: Ask questions to clarify evidence of the factors that have caused the rise in global temperatures over the past century
Key Vocabulary: temperature, average, data, weather, atmosphere, climate, precipitation, humidity, trend
In our first post on climate change myths, we wrote about why we should expect colder temperatures during winter months. The tilt of the Earth’s axis causes the hemisphere angled towards the Sun to change as the Earth travels along its orbit. When the Northern Hemisphere is angled away from the Sun, the Sun’s energy hits that part of the Earth much less directly, resulting in (generally) colder temperatures in that hemisphere. Just because it’s cold in February, does not mean that the Earth’s climate isn’t changing.
But what does it mean for the climate when there are a few warm days in January? Is that proof that the world is heating up? To really figure out what is happening to our Earth, we need to understand the important differences between weather and climate.
Myth 2: Weather and climate are the same thing
When scientists talk about weather, they are referring to the conditions of the atmosphere at any given moment. If it’s snowing, cloudy, hot, cold, sunny, windy, etc. When we look at the weather report, it’s a forecast of what the atmospheric conditions will be for the next few days. Weather is a snapshot of what is going on in the air around you at that time.
Climate refers to the long-term patterns of precipitation, humidity, atmospheric pressure, and temperature, measurements usually taken over a 30 to 40 year period. Decades worth of data is used to describe a region’s climate. Differences in these patterns result in a multitude of different climates. For example, the Southeastern United States has a humid subtropical climate, usually having warm, humid summers and dry, cool, but not cold, winters.
These differences are subtle but important. Weather is what is happening on any given day in the atmosphere, one singular data point. Climate is modeled around long-term trends based on years’ worth of data collected. Just because a region is experiencing a few days of atypical weather, doesn’t mean that its climate is changing. There are warm days in Alaska and it occasionally rains in the Sahara Desert. However, those data points aren’t enough to show a change in a region’s climate. What is needed is more data points to give us confidence in our conclusions. This is exactly what scientists do when researching the climate trends.
Modeling sample size
The data collected by researchers can be hard to digest; most people do not have an advanced degree in atmospheric science. To illustrate the differences between weather and climate, let’s use an analogy that famed data geek Nate Silver would appreciate: baseball.
Modern baseball has been played every year since 1903. Thirty teams play 162 games a year, with starting position players getting about 450 at-bats a season. This gives us a lot of data to work with, making baseball a good model for looking at the difference between an instantaneous data point and a long-term trend. In fact, general managers look at similar data to make projections on player performance and to determine rosters. Let’s do just that.
Say we are the general manager of a team, and we are looking to bring in a big name player on for the new season. We will simplify this scenario by ignoring the unique positions on a baseball team, MLB salary rules, and player age. And while there are a lot of advanced metrics that could be used to determine a player’s value (OBP, OPS, WAR), we are going to stick to batting averages.
We start our search by looking at the free agency list and immediately see two big names: Mike Trout and Alex Rodriguez. But which one should we try to sign? We want a player of all-star caliber, someone who can hit for around 0.300 average or higher. Let’s go back through the box scores from last year and take look at a 5 game stretch in June and compare the two players.
5 Game Average
|5 Game Batting Average||0.210||0.409|
Certainly, a great week for Rodriquez. With an average above 0.400, he would not only be an all-star, but very likely the best hitter in the league (the last player to hit over 0.400 for the season was Hall of Famer Ted Williams in 1941). Trout on the other hand, really struggled, hitting barely above 0.200, the dreaded Mendoza line, the point in which most teams send players back to the minor leagues. In fact, by only looking at these 5 days, he would be lucky to be in the Majors at all. If this was the only data that we looked it, Alex Rodriguez would be the player to sign to a big contract, no brainer.
However, we want to be confident that our new player is going to perform at a high level. The more data we use, the more accurate our prediction of how both Trout and Rodriguez will perform in the upcoming season. Instead of a few days’ worth of baseball, let’s look at their batting averages over the past several years. From 2012-2016 Mike Trout and Alex Rodriguez had 2847 and 1367 at-bats respectively. Statisticians call the amount of data you are looking at sample size and the larger your sample size, the more confident you can be in your conclusions.
5 Year Average
|Mike Trout||Alex Rodriguez|
|Date:||Batting Average||Batting Average|
|5 Year Batting Average||0.309||0.229|
We can see a big difference in player performance. Mike Trout has consistently hit at or above 0.300 in each of the last 5 years, all-star numbers. Alex Rodriguez, however, did not live up to his June 2016 hot streak. While having a solid year in 2012, his average dropped quite a bit since then, spending a year out of baseball and several more hitting closer to 0.200. The likelihood of Rodriquez having another good season is much lower than him performing poorly. By looking at thousands of at bats over several years instead of a handful of at-bats over a couple days gives us a much better picture of each player. thousands of at-bats instead of a handful over We sign Mike Trout to a big contract, and our team goes on to have a great season.
Weather vs. Climate
Just as the performance of a baseball player in any given game doesn’t show their worth, a few days of weather cannot give us a clear picture of the climate. Taking a larger sample size gives us more confidence in what we are looking for. A small set of data could “disprove” the idea that global climate is warming. In 2014, the Polar Jet Stream weakened, spilling cold Arctic air over the entire continent of North America. This brought record-breaking low temperatures over many American cities, including where I live, Cincinnati, Ohio. Below is a data set of 5 days from the Polar Vortex week of February 2014.
5 Day Weather Sample
|Date:||High Temp F° (C°)|
|Average Temperature||16° F (-9° C)|
Even for February in Cincinnati, where the average daily high for the month is 44° F (7° C), those are some very cold days. But even though the high temperatures that week was nearly 30° F (16ׄ°C) below average, it’s not enough data to give us any information about the climate. We could just as easily go back through the record and find several days in a row well above the historical average. The first week of 1961, Cincinnati had quite the heat wave, with most highs at or above 48° F (9° C), several degrees above the January average of 39° F (4° C). But again, the sample size is too small to justify a claim on long-term trends. The weather on any given day is not necessarily indicative of the climate, just as a single game doesn’t show a baseball player’s true value. A few data point does not prove anything; you must take a larger sample size.
So what is happening with climate data?
Luckily, we don’t have to limit our study to the weather patterns in one Midwestern city. Instead, we can look at the aggregated data from the last 130 years of weather across the entire world. NASA’s Earth Observatory, as well as several other organizations, have done just that, and have compiled the data in the chart you see below.
The trend shows that temperatures have increased, by about 1.8° F (0.75° C) over the last 60-70 years. While that may not seem like a lot, we need to put that in perspective. During the last glacial maximum approximately 24,000 years ago, global temperatures were about 10.4° F (5.8° C) below the 20th-century average. At that time most of Canada and most of Europe was covered in a layer of ice several miles thick. 2016 was also the hottest year on record, the third year in a row that has happened. In fact, 11 of the hottest 12 years to be recorded have occurred in the 21st century. The reality is that the long-term trends show the Earth’s temperature is increasing, and quickly. It’s not in dispute, it’s what the data is telling us.
So the data shows that the Earth is warming. But what is causing the warmth? In Part 3 of our series we put some of this change into perspective on a geologic scale and try to come up with an understanding of why our climate seems to be changing.
- Elizabeth Kolbert’s book, The 6th Extinction, is an extremely informative, if not outright depressing, read on how humans activity have been causing the next great culling of biodiversity in our planet’s history.
- Ezra Klein interview Kolbert on his podcast, The Ezra Klein Show. A great conversation between intelligent people.
- Randell Munroe of the XKCD has an illustrated comic to put into perspective on how the climate has changed over the last 20,000 years, that is at the same time easily understandable, aesthetically pleasing, and disconcerting.