 A common mistake people make when it comes to climate change is confusing weather with climate. So how is weather different from climate? Weather is the state of the atmosphere at a given point in time. What's the current temperature, cloud cover, air pressure, wind direction and speed? Is it raining or snowing? Climate, on the other hand, is the average weather over a long period of time. It represents long-term factors like the average high and low temperature for a given date at a given location, the record highs and lows, precipitation amounts and types, and the seasonal variation. As a meteorology professor who teaches and trains future weather forecasters, I sometimes get comments like, oh, so you get paid to be wrong? I laugh a little, but then I list the many reasons why weather forecasting is difficult. I'll explain that long-term weather forecasts past five days should be taken with a grain of salt. But overall, weather forecasting has gotten quite good. Short-term weather forecasts are extremely accurate and have improved dramatically. On the other hand, I live in Boulder, Colorado, where the weather changes very quickly. If a forecast is just slightly off, I might unexpectedly find myself walking to the bus in a snowstorm wearing a t-shirt. Why? Because the models had forecasts that the temperature at that time of day would be warm. This can be frustrating. Nailing the timing of weather events for a given location can be difficult. So the goal of a weather forecast is to tell you the temperature, precipitation and cloud cover forecast for your exact position at an exact time in the near future. This is done using a variety of data from weather balloons, weather stations and satellites that are put into a weather model. The weather model consists of computer code that does calculations to get the forecast. The model divides the world up into blocks and the resolution or the size of the blocks is extremely important. You don't want a nearby mountain range to be represented as a few blocks on a grid. You want it to resemble a mountain range. And you also don't want your storm forecast to be relegated to a low resolution like a day long. You'd rather have hourly or three-hourly forecasts to get a better idea of when the temperature will drop. Meteorologists analyze the results from several weather models to make the short-term forecast that is sent to you in the form of a web page, television or radio broadcast, or maybe an app on your phone. Climate models, on the other hand, have completely different goals. Yes, climate models simulate weather systems and all of those useful variables that come out of a weather forecast. But the goal isn't to tell you that the weather in Denver, Colorado on March 30th of the year 2050 will be a high of zero degrees, a low of negative five, and 10 centimeters of snow should fall between noon and four o'clock. The goal is to tell you that the average weather from the year 2080 to 2100 for spring over the central United States will be an average temperature and range of such and such, this much average precipitation, and so on. This is called the climate, the average of a bunch of weather over a long period of time. Using climate simulations, we can then say that the springtime temperatures at the end of the 21st century will be this much warmer than the last 20 years, this much wetter or drier, and have this many more droughts, for example. To get these long-range projections, climate models are built differently than weather models. The climate is influenced by processes that happen over much longer time scales and bigger areas than weather. The climate models used in the IPCC reports, called atmosphere-ocean general circulation models, include many important earth systems. They take into account cycles such as the carbon cycle. They also include feedbacks, including interactions between the atmosphere, ocean, sea ice, and land that happen on a global scale. Climate models tell us the factors that control our everyday weather will be different in the future. So that's how weather and climate models differ. However, one myth distorts what models are in order to cast doubt on their usefulness. The myth goes like this. Since modern computer models can't predict the weather two weeks from now with any certainty, how can we rely on computer models to predict what the earth's climate might be a hundred years from now? This myth suffers from a fallacy of impossible expectations. It confuses weather with climate and tries to use the emotion of disappointment in a field weather forecast to have you fall victim to this fallacy. Now, imagine if I asked you to predict the result of a coin flip, heads or tails. You would have a 50-50 chance of getting it right. Now, what if I asked you to predict the results of a million coin tosses? You would tell me about half would be heads and half would be tails, and you would be almost 100% correct. This is a good analogy for the difference between weather and climate predictions. Although, luckily for us, your chances of predicting the weather tomorrow are usually better than 50-50. Using your confidence in the ability for forecasts of the weather to then judge the validity of climate modeling just isn't fair, because as you know, there are two completely different challenges.