Figure 1. Example of a synthetic 6.185 micron image from 30 April 2010 at 23 UTC. The image is based on a 23-hour forecast from NSSL's 4-km WRF-ARW.
1) Product Information:
- Who is developing and distributing this product?
This product is a combined effort between the National Severe Storms Laboratory (NSSL) in Norman, Oklahoma, and The Cooperative Institute for Research in the Atmosphere (CIRA) in Fort Collins, Colorado, together with the NOAA/NESDIS RAMM Branch.
- Who is receiving this product, and how?
Daily output from NSSL's 4-km WRF-ARW is provided to CIRA, who then generate synthetic satellite imagery, which is sent to the Storm Prediction Center (SPC) via a McIDAS ADDE server.
- What is the product size?
Each image is just under 1 MB, and every day 28 images at 6 different bands are provided.
2) Product Description:
- Purpose of this product:
This product has two primary purposes: 1) Synthetic imagery from cloud model output can be used to evaluate each model run. For example, one might compare a simulated water vapor band to observed GOES imagery from 12-18 UTC to see how well the model is handling the timing and location of upper level features, such as shortwaves. 2) Since the simulated bands are based on the GOES-R ABI, looking at the imagery will prepare forecasters for how the actual GOES-R imagery will look when it becomes operational. For example, certain features may be visible at these wavelengths which are not viewable in the current GOES bands.
- Why is this a GOES-R Proving Ground Product?
The synthetic imagery is a Proving Ground Product because it replicates how actual features will appear in GOES-R ABI bands.
- How is this product created now?
Every day at 00 UTC, NSSL runs their 4-km WRF-ARW. As soon as the 12-hour forecast is completed, several variables are extracted and scp'ed to CIRA. These variables include temperature, water vapor, and other physical and microphysical parameters which are needed for the next step. When all variables have been receieved at CIRA, an observational operator is run to generate the synthetic imagery for 6 GOES-R ABI bands (3.9, 6.95, 7.34, 8.5, 10.35, and 12.3 microns) including some band differences. The simulated imagery is then converted to McIDAS AREA format and made avaiable for the SPC, who then makes the output viewable on their NAWIPS system. Hourly output between 09-12 UTC is processed daily, providing hourly forecast synthetic satellite loops out to 36 hours. The resolution of the output is 4-km to match the input resolution of the cloud model; the real GOES-R ABI bands will have 2-km resolution.
3) Product Examples and Interpretation
The synthetic imagery loops are available anytime between 10 and 15 UTC, depending on which band is used for the product. See the product description page for each individual product to know the available time. In the example below, the 7.34 micron band is compared to the observed GOES-13 Sounder data at a similar central wavelength. Note how the simulated data is at a better resolution that the sounder data (4 km v/s 10 km; GOES-R ABI will be 2 km), but the largest advantage is that the synthetic loop was available for viewing in the morning, thus providing an excellent resource for forecasters.
Figure 2. Synthetic 7.34 micron imagery from 24 April 2010 at between 12 and 00 UTC, and the observed imagery from the GOES-13 Sounder at a similar wavelength. Click for a larger view. In this case, the model does a fairly good job with the timing of the convection forming in the southeast U.S., but appears to overdo the coverage. The brightness temperatures in the clear areas match well, but the model has a cold bias with the thin cirrus clouds in the Great Lakes region.
4) Advantages and Limitations
Advantages of the synthetic ABI imagery include: 1) Satellite imagery can be viewed before the simulated time actually occurs, so forecasters know what to expect, 2) three water vapor bands allow one to view different atmospheric levels since the weighting functions peak at different levels, and 3) forecasters can use this imagery to prepare themselves for what actual GOES-R ABI imagery will look like. The biggest limitation is that the forecast is only as good as the cloud model forecast; if the model does not initiate convection, for example, then the convection will also be absent from the synthetic imagery.