Frequently Asked Questions - Clouds Protocol

Ground based truth from humans regarding cloud locations, cloud type, and cloud height are important for validating satellite and model data used by scientists to detect and characterize clouds.

There are a number of reasons your measurements may differ but not every discrepancy implies an error on the part of the user. Satellite imaging uses a form of passive remote sensing to capture the top-down perspective of clouds in the atmosphere, whereas you, the ground observer, observes clouds from the ground-upwards. These two orientations offer complementary perspectives on the formation of clouds in the atmosphere. For example, if the satellite reports high-level cirrus clouds and the user only reports dense low and mid-level clouds, it may be the case that dense mid-level clouds obscured the observer’s view so that high level clouds were not visible from the ground.

There are a number of things to say here: 1) determining cloud height precisely as a ground observer is a very big challenge, because there are no points of reference in the sky (unless you are near a very tall building or a big mountain). This is especially true for stratus clouds, which provides very few hints from their shape about where they might be located. If you are close (off by one level), we think that's good.
2) determining cloud height from a passive remote sensor in space is also a big challenge, especially in certain cases. The figure at right (courtesy of Dr. R. Holz at the University of Wisconsin) illustrates one classic case. This is a vertical profile of the atmosphere from the CALIPSO satellite, which uses a laser to locate cloud and dust layers. Dark blue indicates clear sky; bright colors indicate cloud layers. In most of this image, CALIPSO finds more than one layer of cloud. The red dots in this image show where the MODIS instrument thinks the cloud top is (MODIS is the instrument that is used in the CERES algorithm). Notice that in areas with more than one cloud layer, the MODIS cloud height is in the clear sky region between the two cloud levels. This problem can also sometimes happen when you have a single cloud layer plus snow on the ground. MODIS can put the cloud level in between the two. (If you think you see one of these cases, please let us know!). As you can see, this is a problematic area, and your observations help us to sort it out.

Ground based truth from humans regarding cloud locations, cloud type, and cloud height are important for validating satellite and model data used by scientists to detect and characterize clouds.

Participants receive a Satellite Match comparison email when their observations falls within +/- 15 minutes of an over passing satellite. Check the Satellite Overpass Schedule online or visit the Satellite Flyovers tab in the GLOBE Observer app to check upcoming flyovers in your area.

Satellite data is typically received within 1-7 days. Once satellite data is received, it is typically processed within 1-7 days. You will receive a Satellite Match email once your Match is complete.

You should look at most of the sky that you can see. Basically, anything above the angle of the trees is within the field of view of the satellite. One way to do this: hold out your arms in a V, so that your hands are at about the level of the top of your head. That gives you the approximate angle of the sky to view.

The Satellite cuts off opacity at very definite - and arbitrary - levels. The cut-off between opaque and translucent is an optical depth of 10. In this case, the cloud had an optical depth of 6.1. That is mighty thick, and it would be very hard to tell the difference by eye. We've had some debate about whether 10 is the right cut-off. In fact, 3 might be a better value. But, either way, the human eye is never going to have that definite sensitivity.

Patterns do exist, but are dependent on local conditions. In some cases, very local conditions such as which side of a mountain you are on. So, you (and your local weather people) are the best to determine whether there really is a consistent relationship. You might be interested to explore the MODIS Rapid Response system where you can find daily (NOAA-20) and nightly satellite images. They might be helpful to see patterns, and also to see whether this is due to some regional phenomenon.

Those are places where MODIS thinks it saw fires. This Rapid Respose system is also called Rapidfire, and is used for fire management by some of the other federal agencies.