Observations of MODIS Fog Product

Note:  This entry is provided by Henry Steigerwaldt, the SOO at NWS WFO Nashville, TN.

Early on this particular morning, high pressure covered the area. Winds were light and skies were clear most of the night and into the early morning hours. And there was enough moisture to allow fog formation in a few areas across Middle TN.  Scattered low and mid clouds were moving NW and then N into the southern portions of Middle TN.  I’m going to focus on just two areas where fog formed, one along the Cumberland River and the other along the Duck River.


This first image above is the MODIS IR product for 0801Z 26 AUG 2009. I’ve outlined most of the low to middle cloud area. Note that you can see some of the larger lakes and rivers stand out somewhat, with their darker (warmer) temperatures from the cooler land areas in this 1 km resolution product. But diagnosing where fog is occurring would be hard to do using only this one product in this case. Note that several 0800Z observations have been included, but they are next to impossible to read on this particular image. The lowest visibility was 3 miles.

The second image is the GOES Low Cloud Base product at 0815Z, with some of the same observations included. This 4 km resolution product only was able to detect two pixels of red (white circles), one each in the areas in which we will briefly investigate,which indicated cloud basesless than 1000 feet.

The third image is the MODIS Fog product. With its excellent 1 km resolution, and its color curve enhancement, it certainly helps to pinpoint areas where fog is occurring. An area of fog is easily observed along a portion of the Cumberland River. This portion of the Cumberland is wider than some other areas because it’s located upstream of the Cheatham Dam. The river in this area is called Cheatham Lake. The Duck River is not as wide as the Cumberland and is not anywhere near as long. An area of fog had also formed along a portion of the Duck. Although the Duck River is not as wide as the Cumberland can be in places, a longer, wider and thus more extensive area of fog had developed.

The fourth image in this blog is a High-Resolution topography. The areas along the Cumberland and Duck Rivers where fog developed have been outlined. The black dotted lines have been added to help show other portions of the rivers’ channels. I’ve included one 08Z observation of 3 miles in fog located south of the Duck River. Unfortunately, no other observations are available anywhere close to the locations where the fog formed along these two rivers.

The picture above is taken along the Duck River in the area where the fog formed. Notice how much narrower the Duck River is when compared to the Cumberland.

Besides a clear night with light winds, fog formation is also dependent on many other factors.These include topography, moisture profile in the lowest levels of the atmosphere, boundary layer turbulence, soil type, soil moisture, ground temperature, total precipitable water in the atmosphere, and the difference between water temperature and air temperature if the location is in a stream or river valley. Although it’s rare to have all this information available to us,forecasters can still make relatively good judgments as to whether fog formation is favored or not on any given night.

My thanks to Google Maps for being able to capture the river images for use in this blog.

NWS WFO, Nashville, TN

3 thoughts on “Observations of MODIS Fog Product

  1. Nice post. It provides a lot of detail on the use of the GOES and MODIS “fog” product. These products are actually “fog and low cloud” products in that the difference image methodolgy can not always uniquely differentiate between fog near the ground and low clouds that may extend from near the ground to some higher level in the atmosphere. This post demonstrates this limitation. However, the 11 micrometer channel on MODIS (GOES) can be used to estimate the cloud or fog top temperature. In fog cases, this temperature will be much warmer that for low clouds (indicating the top of the fog bank is closer to the ground than the cloud top). Thus using the 11 micrometer channel along with the “fog and low cloud” product can be more effective in detecting fog than either one alone. SPoRT is exploring an “enhanced fog” product that takes this into account.

  2. Gary:

    Is the MODIS Cloud Top Pressure (4 km) product in our AWIPS the same thing as
    the 11um brightness temperature product you mentioned?

    If not, we don’t have that 11um product in our system to look at.

    Thanks.
    Henry Steigerwaldt
    Nashville, TN

  3. Henry. Very good question. The MODIS 11 micrometer channel (could be called longwave infrared) is not the same as the MODIS Cloud Top Pressure product. The MODIS 11 micrometer (lonwave infrared) channel (1km resolution) is very much like the GOES infrared window channel (4km resolution) at the similar wavelength. It measures energy emitted by the surface or cloud top (with just a little attenuation by the atmosphere above it). The Cloud Top Product is derived from the 11 micrometer imagery AND a cloud mask to better quantify the pressure of the top of the clouds. Unfortunately, the cloud mask is not perfect (often misses the low/warm clouds/ fog or over determines the presence of the clouds) and leads to false determinations. It would be interesting to monitor the CTP product in fog situations to evaluate its performance. I will also check on why your office may not be getting the 11 micrometer channel.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s