UAH CI "Strength of Signal" Probability Forecasts Up and Running at 2012 GOES-R Proving Ground

With the EWP forecasters stationed in southwest Texas today, we were noticing some of the newer instances of convective initiation flaring up just across the Rio Grande River in Mexico.  Provided are a couple of the first views of the new UAH CI  SATCAST Strength of Signal (SS) output, as seen in the AWIPS-II display.

Figure 1.  AWIPS-II screenshot of the UAH CI Strength of Signal output from 1732 UTC (top-left pane) and corresponding satellite IR image (top-right pane), from the same satellite scan.

Figure 2.  AWIPS-II screenshot of the UAH CI Strength of Signal output from 1845 UTC (top-left pane) and corresponding satellite IR image (top-right pane), from the same satellite scan.

This new version of the SATCAST algorithm provides a continuous scale from 1-100 of the strength of the satellite-derived signal for near-future CI.  Per forecaster feedback from last year's 2011 SPC Proving Ground, this new product has taken a step away from the binary "Yes/No" output CI forecasts and now combines all of the available satellite spectral information into a single numerical value.  Higher values correspond to faster rates of cloud top cooling rates, colder/higher cloud tops, and wider updraft widths.

The Strength of Signal values are currently derived from the early beginnings of an ever-growing neural network in which the algorithm is trained from an empirical database with the correct answers of what immature forms of convective clouds go on to produce CI and which ones do not.  As this database continues to grow over the coming weeks, months, and years, the algorithm will "grow smarter".

Furthermore, this new paradigm is paving the way to soon begin training the algorithm with input from Rapid Refresh Numerical Weather Prediction model data. As this transition occurs in the near-future, the output will look the same, but it will become increasingly more meaningful as a true probabilistic product.

This work ties into ongoing research with NOAA's Earch System Research Laboratory (ESRL) in which the SATCAST data is being assimilated into developmental versions of the Rapid Refresh (RAP) and High Resolution Rapid Refresh (HRRR) models in order to "put the convection in the right place" for hot-starting the models, inevitably leading to more accurate short-term model output.  To that end, we envision a truly integrated and data-fused product in which a positive feedback effect occurs as short-term model data characterizing the broad-scale atmospheric environment is ingested by the CI algorithm in order to produce more accurate CI forecasts, and these more accurate CI forecasts are assimilated into the short-term prediction models in order to significantly improve how and where the models are simulating CI.  If the models can get the placement of early convection correct, then their longer term forecasts will take a giant leap forward in terms of accuracy in all aspects of their output.

EPZ Warning Desk- Dona Ana Co. update

Experimental forecaster discussion from EWP blog...

"UAH CI shows new convection developing over eastern Dona Ana county and expect continued development through 1930Z.  EM and other decision makers should continue to monitor this area for possible warnings and statement."

Simulated satellite imagery in AWIPS II

 

AWIPS II 4-panel display of the NSSL-WRF simulated satellite imagery, showing standard IR window (top left), WV (top right) and 10.35-12.3 micron channel difference.

CIRA and UW-CIMSS are providing the HWT and SPC with synthetic satellite data that is generated from the 4km NSSL-WRF.  This data is available for all GOES-R ABI IR bands and arrives locally by 7am.  This year we made a first effort at bringing the data into AWIPS II, which was successful thanks to Darrel Kingfield at NSSL.  The above image is a 4-panel display of the synthetic IR, WV and 10.35-12.3 micron channel difference, which was described in a previous blog post here.  I spoke with one of the forecasters in detail about the 10.25-12.3 micron channel difference and explained that GOES-R will be able to provide this, as well as other unique combinations of imagery such as RGBs, every 5 minutes over CONUS.  I discussed with her how we expect to combine imagery into unique products to help the forecasters use the ABI channels rather than simply sending all the bands to each forecast office.  The forecaster mentioned that she definitely wants to bring this simulated satellite imagery back to her office and I had her coordinate with the UW-CIMSS representative this week.

In addition, based on forecaster feedback from last year, CIRA provided us with a new color table to better distinguish areas where convective initiation might occur in the 10.35-12.3 micron channel difference imagery.  Forecasters have expressed approval for the new color table and have mentioned that it is easy to detect threat areas.

EWP: UW-CTC captures early afternoon convection

An unstable and weakly capped atmosphere has resulted in widespread convective development over southern Texas, northern Mexico into southern New Mexico late this morning and early this afternoon.  The UW-CTC product captured the vertical growth of these storms as inferred by the cloud-top cooling rates in the images below.  As was shown yesterday, the UW-CTC rates are providing good lead-time ahead of strong radar reflectivity.  The top panels are the GOES IR-window brightness temperatures (top left) and GOES visible reflectance (top right).  The bottom panels are the UW-CTC rates (bottom left) and El Paso, Texas 0.5 degree reflectivity (bottom right).  Unfortunately the timing of initiation was prior to the start of EWP so these hits have not been extensively evaluated by the forecasters on shift.  Perhaps later development in areas of thin cirrus clouds will provide more examples of the results of including the GOES optical depth retrievals in the UW-CTC algorithm.

 Figure 1. Valid 1715 UTC 08 May 2012.

 Figure 2. Valid 1745 UTC 08 May 2012.

  Figure 3. Valid 1815 UTC 08 May 2012.

EWP daily debrief 5/8

Today we had our first daily debrief with the EWP forecasters.  The forecasters were focused on the San Angelo and Austin, TX areas issuing severe thunderstorm warnings throughout the evening.  We discussed the previous day's activities and asked them a little about what they thought about the product performance during their warning operations...


SATCAST / UW-CTC
- The forecasters created a procedure called "Awesome CI" to complement last year's "Ultimate CI" (which we have made available to the forecasters in AWIPS II) that included the SATCAST product with the UW cloud-top cooling, along with OUN-WRF hourly theta-e and wind fields to combine both observed CI products with some model-based analysis fields... "Helped increase our situational awareness, which was a big plus."
- NWS Charleston forecaster mentioned both CI products back to home office and said they were very excited to look at them.
- "When the (SATCAST) data was coming in there was some positive responses to it... there were consecutive images where there were strong signals and it definitely increased my confidence."
- "With these products, even with training, experience is the key thing to get forecasters to use it."
- "It was nice having both the SATCAST and the CTC, where one is more sensitive and helping identify the area, the other helped key in on those really strong storms that I should be paying attention to."
- Would it have been helpful to see the feedback questions before going into operations to get an idea of what to look for?  "It might have helped direct us."
- "With the CTC, most of the signals were out in W TX and we weren't really looking at the radar, and those strong signals helped."

Nearcast
- Theta-e difference used extensively when storms were already mature... monitored what environment the storms were moving into... helpful in identifying intensification.
- "Could clearly see the boundary between the more stable air and unstable air."
- Couldn't get all the individual field levels in AWIPS II... was only able to get the difference.

Overall / Training
- Would be helpful to get some of the WES procedures built into AWIPS II.
- All these products would be helpful if someone was assisting the warning forecaster looking at these (and other) derived products.
-  Do you feel comfortable with the current training setup, do you want more time with the PIs or material on Monday?  "Getting us into that mindset prior to arrival with the WES case was very beneficial."

Nearcast in a warning environment

EWP forecasters are currently monitoring severe weather over southern TX and issuing experimental warnings for storms that have been producing severe hail.  One of the forecaster groups localized for the forecast office in Austin, TX has been monitoring GOES-13 IR with the UW-CIMSS differential theta-e product overlaid.  Their goal is to identify the near-storm environment that the currently severe storm is moving into to aid in the warning decision making process.  This provided a very good example how the Nearcast product maintains previous observations where new observations cannot be made due to cloud cover.  This can be useful in providing high-resolution observational data key to identifying localized regions of instability that cannot be obtained through other means.  In this case the forecasters decided to issue an experimental severe thunderstorm warning at 2228 UTC based on the current representation of the storm on radar and the Nearcast differential theta-e fields.

EWX Meso Update 2320 UTC

Forecaster comments from the EWP blog...

The first image below shows cloud top cooling rates at 1715Z.  Note the values of almost -35 K in the light blue over west central TX.    The second image below of IR satellite imagery at 1745Z, depicts the beginning phases of strong convection in the same region as the cloud top cooling noted 30 min before strong convection initiated.  The CIMSS cloud top cooling product correctly predicted the formation of a  complex of strong storms that later became severe during the late afternoon/early evening hours.

EWP: UW-CTC in areas of thin cirrus

The montage below demonstrates the improvement to the UW-CTC products to operate in areas of thin cirrus cloud cover during daytime hours.  This improvement was one of the largest deficiencies identified by forecasters in previous testbeds.  Diagnosing cloud-top cooling rates in presence of thin cirrus clouds is possible by inclusion of GOES daytime optical depth retrievals.   The UW-CTC rate at 2145 UTC 07 May 2012 was approximately -6 K / 15 min (lower left, figure 1); with further cooling of ~ -18 K / 15 min at 2215 UTC 07 May 2012 (lower left, figure 2).  By 2245 UTC the Dallas-Fort Worth radar indicated 50-55 dBZ composite reflectivity.  Another cell further west was not captured by the algorithm, possibly due to its position closer to colder anvil temperatures.

Figure 1. 

 Figure 2.

Figure 3.

EWX Meso Update 2230UTC

Forecaster comments from the EWP blog...

Issued a (experimental) warning on a cell for hail and wind based on radar trends and the fact that this storm was moving into a good environment.  The above image of radar reflectivity shows the storm moving southeast into a good theta-e environment depicted by the yellow axis on the above image.  The theta-e imagery is CIMSS-NRE Layer GOES Vertical Theta-e Diff Low-Mid Img

EM Discussion @ 21Z

Experimental Area Forecast Discussion (AFD) from EWP blog...

Location: Latest guidance supports continued development of strong to severe thunderstorms moving across McColloch, San Saba, Llano, San Saba and Mason counties.

Impacts:  primary impact will be large hail and damaging winds.  there is still potential for isolated tornadoes.

Met discussion:  OUNWRF showed evidence of moisture convergence on the h8/h7 wind and theta E H8-7 layer; MRMS Max hail size estimated 1.5″ hail;  UAH CI strength of signal showed 80-100 values over the areas of favorable moisture convergence; CI-CTC projected new development over Llano county.