An Evaluation of our WRF

We are continuously developing products to meet the needs of various industries. In doing this, we constantly are testing products to see how they handle different situations. One of the products we run is the “HazWRF3″ which is an ARW core 3km high resolution forecast model. This uses a combination of micro-physics schemes that we choose to fit each region or desired phenomena to model.

Since we build models such as these for our clients, we have decided to put together a few images showing the performance of the model for various events. There is a gallery below of some cases, that will update as more events are captured.

To speak to us about a custom modeling solution for yourself, please contact us.

 

10/1/14 KS/MO Severe Weather Event

The night before severe weather hit the Midwest, including a plethora of severe weather watches and warnings, our model had a grasp on the situation and provided ample lead time before the event.

1 2 3

7/21/14 Parkersburg, IA Tornado

Even 24 hours out, our WRF model predicted the supercell thunderstorm nearly perfectly. Actual radar is the top image with the 4 runs of the model leading up to the event, shown below,

wrf

 

 

 

5/26/14 Watford, ND Supercell

We ran the HazWRF, using a run time at 3z Monday (Sunday evening at 10pm). The **19 hour forecast** (Valid 22 UTC Monday) showed a powerful supercell passing a mere few miles north of where the actual tornado occured. The actual tornado passed through this area between 00 and 01 UTC (0046 UTC radar shown). This is an example of INCREDIBLE performance from the model and attests to the hard work and incredible talent of the Hazard Notifications LLC team!

 

wrf

 

5/26/14 West TX Supercells

Ran at 17 UTC, the 5 hour WRF forecast nearly perfectly forecasted the storms that produced severe weather.

5 hour WRF forecast vs actual radar. 5/26/14.

5 hour WRF forecast vs actual radar. 5/26/14.

 

5/25/14 West TX Supercells

At 15 hour forecast, the WRF did very well this day, capturing one supercell and hinting at the other ongoing supercell further south across Mexico.

15 hour forecast (right) vs actual (left).

15 hour forecast (right) vs actual (left).

 

5/21/14 Denver Supercell

The model at 5 hours perfect predicted the supercell that occurred later that afternoon. The model resolved the rotation path of the supercell 5 hours in advance, down to nearly perfect time and within a few miles!

5 hour forecast vs radar.. DIA supercell 5/21/14. Model was correct to within miles!

5 hour forecast vs radar.. DIA supercell 5/21/14. Model was correct to within miles!

Posted in Official.