One of the exciting new products that will be available on GOES-R is the split window difference (SWD) which is simply the difference between the 10.35 micron and 12.3 micrometer channels. This channel difference has been shown to provide information about atmospheric column water vapor. Higher SWD values (larger positive difference) can correspond to deepener low-level moisture in a cloud-free environment. This signature can be utilized to anticipate where and when convective initiation will occur in cloud-free conditions away from complex terrain (such as the Great Plains). Although similar bands were available on some previous GOES instruments, their coarse resolution and poor signal-to-noise ratio made them less useful for identifying subtle small-scale features in the low level moisture field.
In order to demonstrate this product (since the 12.3 micron channel is not available on the current GOES imager), we use synthetic imagery from the 4-km NSSL WRF-ARW model. Here is an example of the SWD on a day with a dryline across Texas:
The larger (positive difference) values of SWD are shown in warm colors, while the location of the cross section (shown below) is illustrated by the east-west oriented black line. Next, we will look at output from the NSSL WRF-ARW model along the cross section line:
The white line indicates SWD values (scale on the right) while the colors are specific humidity. SWD values are greatest along the dryline where the depth of the moisture is greatest. The low-level temperature lapse rate also plays a role in the SWD, but as can be seen in the cross section, the depth of the moisture is the dominating factor.
A loop of the synthetic SWD from the NSSL WRF-ARW:
shows the animation from 1500 – 0000 UTC at hourly intervals from the model. On the left is the synthetic IR (10.35 micron) band and on the right is the synthetic SWD product (larger SWD values are shown in warmer colors).
The first thing to note is the skies are clear before convective initiation across Texas which is necessary to make use of the product in this way. The larger SWD values develop along the dryline prior to convective initiation. Keep in mind this synthetic data is at hourly intervals, but once GOES-R becomes available, the data will be displayed at 5 (or even 1) minute intervals.
We can preview how this data may appear on GOES-R by looking at an example from the MSG (Meteosat Second Generation) SEVERI instrument over Europe. An event occurred on 6 July 2012 where convection developed along a convergence boundary under clear skies prior to initiation. Also, this event occurred over flat terrain (Poland) which is important since complex terrain complicates this signature.
Here is the zoomed in visible imagery (over Poland) from the MSG satellite from 0845 – 1500 UTC 6 July 2012:
The key to note is the clear skies prior to convective initiation.
Here is the zoomed in SWD imagery (over Poland) from the MSG satellite over the same time period:
Focus on the clear area (that was shown in the visible image) from the center of the scene southeastward. SWD values gradually increase (going toward warmer colors) indicating deepening moisture along this convergence boundary, followed by convective initiation (expanding regions of blue/purple later in the loop).
A local maximum in SWD developed over a convergence boundary (under clear skies) about 2 hours prior to convective initiation. Forecasters can make use of this information when attempting to predict where / when convective initiation will occur. As looking at this imagery becomes routine with GOES-R for diagnosing convective initiation (under clear skies beforehand), experience with this product will lead to greater forecaster confidence in timing and location of convective initiation.
For more detailed information on this product, see this article:
Lindsey, D.T., Grasso, L., Dostalek, J.F., and J. Kerkmann, 2014: Use of the GOES-R Split-Window Difference to Diagnose Deepening Low-Level Water Vapor. J. Appl. Meteor. Climatol., 53, 2005–2016. http://dx.doi.org/10.1175/JAMC-D-14-0010.1