|
|
How Do Models Make Clouds?
for more advanced readers
CONVECTIVE ADJUSTMENT
Pro's and Con's
This is a simple scheme and does not require a lot of computer power. However, in just one time step (usually less than 10 minutes) the scheme represents the vertical mixing effects of a cloud that would normally occur throughout the clouds entire lifetime (a cumulus cloud has a typical lifetime of 45 minutes). Also, it only takes into account the local atmospheric temperature profile. In reality, cloud processes also are affected by the surface and by horizontal transport by large-scale motions.
What is a Time Step?
We know what the "state" of the atmosphere is in the running model at any one time. We predict how quickly that state is changing. For example, perhaps the temperature is 40deg F and is changing at -2deg F per hour. We now use this rate of change to predict what the state of the atmosphere will be in a set amount of time called a time step. Let's say our time step is one hour. That means after one time step with the state, above, our temperature will be 40deg - 2deg = 38deg F. We then calculate all the variables in the model, like temperature, knowing their rates of change and recompute new rates of change based on how the model is programmed. In one of the Colorado State University GCMs, a typical time step is 200 seconds. Next page -> how do models make clouds, continued
Links and resources
for more advanced readers
CONVECTIVE ADJUSTMENT
Pro's and Con's
This is a simple scheme and does not require a lot of computer power. However, in just one time step (usually less than 10 minutes) the scheme represents the vertical mixing effects of a cloud that would normally occur throughout the clouds entire lifetime (a cumulus cloud has a typical lifetime of 45 minutes). Also, it only takes into account the local atmospheric temperature profile. In reality, cloud processes also are affected by the surface and by horizontal transport by large-scale motions.
What is a Time Step?We know what the "state" of the atmosphere is in the running model at any one time. We predict how quickly that state is changing. For example, perhaps the temperature is 40deg F and is changing at -2deg F per hour. We now use this rate of change to predict what the state of the atmosphere will be in a set amount of time called a time step. Let's say our time step is one hour. That means after one time step with the state, above, our temperature will be 40deg - 2deg = 38deg F. We then calculate all the variables in the model, like temperature, knowing their rates of change and recompute new rates of change based on how the model is programmed. In one of the Colorado State University GCMs, a typical time step is 200 seconds. Next page -> how do models make clouds, continued
Links and resources
