Global Climate Models (GCMs)

0:05 Hello, my name is Derek Rosendahl and I am a post-doctoral research scientist with the 0:09 South Central Climate Science Center. 0:11 Today, I will be discussing what a global climate model is and 0:14 how it works. 0:20 Scientists use models to represent a number of things such as objects, concepts, or systems 0:24 in order to gain a better understanding of them. 0:27 But the term “model” can mean many different things, from a physical model of a watershed 0:30 to examine the movement of sediment in a river, to 0:33 a statistical model of the relationship between consumers’ interest and the type of vehicle 0:37 they buy. 0:38 In the case of a global climate model, we are referring to a numerical model that runs 0:41 on a powerful supercomputer. 0:43 The lines of the computer code describe what scientists know about 0:46 the physical processes that occur in the atmosphere, ocean, biosphere, and land surface. 0:52 Fundamentally, these processes represent how heat, moisture, momentum, and chemicals move 0:57 across space and through time. 0:59 Imagine trying to simulate the continuous Earth system in a computer model. 1:03 Global climate models work by dividing up the Earth’s atmosphere, 1:06 oceans, land, and sea ice into lots of 3 dimensional boxes, or grid cells. 1:11 Modern global climate models can have grid cells that are 50 miles 1:14 square in the horizontal and that can go up to 90 levels in the atmosphere and down to 1:18 60 levels in the ocean. 1:20 Most common global climate model simulations begin at an initial state resembling the 1:24 beginning of the 20th century and then respond over time to changes in the Earth’s energy 1:28 budget by processes such as the amount of incoming 1:31 solar radiation or atmospheric concentrations of 1:34 greenhouse gases and aerosols. 1:37 Using equations for the basic laws of physics, fluid motion, and 1:40 chemistry, variables such as temperature, precipitation, wind, moisture, and pressure 1:45 are solved at each of the grid cells and are able to 1:47 interact with each other across space and time. 1:50 For example, the winds may be blowing colder air 1:52 into a region, so, over several time steps, the 1:55 temperature at a given grid point will decrease because the computer calculates where the 1:59 wind field moves that colder air mass. 2:01 When we visualize the results, or output, of these global climate models, it is sometimes 2:05 difficult to tell the difference between the model output 2:07 and actual measurements from satellites, weather stations, or other observing systems. 2:12 That’s how good today’s global climate models are! 2:14 The current generation of global climate models includes the physics that represent how 2:17 radiation is transmitted, reflected, and absorbed within the atmosphere and at the earth’s 2:22 surface. 2:23 They represent clouds, rain, and snow; evaporation of water from the oceans; transpiration from 2:29 plants; the movement of water into the soil; and ocean circulation around the globe. 2:34 They can generate tropical storms, cold fronts, jet 2:36 streams, and trade winds. 2:38 They can also release Sulphur dioxide through volcanic eruptions or absorb 2:42 carbon dioxide in the oceans and forests. 2:45 Essentially all of the major drivers of our climate system are represented in the current 2:48 generation of global climate models. 2:51 Some processes are either too small in size or too short in duration to be modeled well. 2:55 In these cases, simplified representations of these 2:58 processes called parameterizations are used. 3:01 Individual thunderstorms are one example of these small-scale phenomena. 3:05 In these cases, modelers represent the physical impacts of 3:08 these processes, such as the exchange of heat and 3:10 moisture between the small scale and the large scale. 3:14 Global climate models typically include parameterizations for processes such as small-scale 3:18 vertical motions – or convection; turbulence near the earth’s surface; development of 3:22 cloud droplets, raindrops, and snowflakes; and 3:25 molecular interactions with solar and terrestrial radiation. 3:29 Although global climate models have seen significant improvements over the last few decades, 3:33 they still are not perfect. 3:34 No model can exactly reproduce our complex climate system. 3:37 There are still physical processes in our climate 3:40 system that scientists are trying to better understand, 3:43 and as they do so, that knowledge will continue to be incorporated into climate models. 3:47 When you think about the increasing complexity of the represented processes and the millions 3:51 of grid points needed to cover the earth’s 3:53 surface, atmosphere, and oceans, you realize that the 3:56 computer power needed to calculate thousands of equations over hundreds of years is immense. 4:01 That is why global climate models need to run on supercomputers. 4:04 In fact, we use some of the fastest computers in the world to run our 4:07 global climate models. 4:09 With all of that computing power needed, why do we even model the climate in this way? 4:13 The main reason is that we are trying to understand 4:15 how atmospheric processes & feedbacks impact our climate system, especially in areas where 4:20 there are few measurements, such as in South America or Asia. 4:24 We also can learn about our climate system by making changes in the model 4:27 and seeing how those changes affect the climate. 4:30 Since we only have one planet, we can’t learn 4:32 about how our atmosphere works by conducting lab experiments on it. 4:35 We must model it instead. 4:37 We also use climate models to measure the influence humans have on our climate. 4:41 We do this by running the models with and without manmade 4:43 greenhouse gas emissions in what are called “fingerprint studies” and then comparing 4:47 the similarities and differences in the results. 4:49 Finally, and perhaps most importantly for decision makers, is that we can use climate 4:53 models to project what our future climate may be. 4:56 Global climate models are the most important tool for 4:58 showing us that our future climate depends on the decisions we make today. 5:02 Without global climate models, we would not be able to manage 5:04 for a changing climate.