College of Engineering | Apply to CSU | Disclaimer | Equal Opportunity Statement | Privacy | Search CSU

Using Total Precipitable Water Anomaly as a Forecast Aid for Heavy Precipitation Events

You must be on the CSU network—either physically or using VPN—to watch this or any of the videos on this site.

October 18, 2012
Lance Vanden Boogart
Hosted by Tom Vonder Haar (advisor), Russ Schumacher, Stanley Kidder, Jorge Ramirez (Civil and Environmental Engineering)

Abstract

Heavy precipitation events are of interest to weather forecasters, local government officials, and the Department of Defense. These events can cause flooding which endangers lives and property. Military concerns include decreased trafficability for military vehicles which hinder both war- and peace-time missions. Even in data-rich areas such as the United States, it is difficult to determine when and where a heavy rain event will occur. The challenges are compounded in data-denied regions. The hypothesis that total precipitable water anomaly (TPWA) will be positive and increasing in time preceding heavy precipitation events is tested in order to establish an understanding of TPWA evolution. Results are then used to create a simple precipitation forecast aid.

The operational, 16 km, 6-hourly TPWA product developed at the Cooperative Institute for Research in the Atmosphere (CIRA) compares a blended TPW product with a TPW climatology to give a percent of normal value. TPWA evolution is statistically described for 84 heavy precipitation events which occurred between August 2010 and November 2011. An algorithm which uses various TPWA thresholds is then developed and tested to determine the extent to which TPWA might be used to aid in forecasting precipitation over mesoscale domains using only satellite data inputs.

The hypothesis of positive and increasing TPWA preceding heavy precipitation events is confirmed. Event- average TPW values rise for 36 hours and peak at 154% of normal at the event time. Using the statistics derived from the heavy precipitation events, a viable forecast aid algorithm was created. Events detected by the algorithm were not of sufficient magnitude to be termed “heavy” precipitation events, which had a 12-hour area-mean accumulation of 25 mm in the events analyzed. Yet precipitation values of greater than 5 mm were achieved for multiple thresholds. False alarms were not quantified, thus qualifying the algorithm’s use as an aid, not a deterministic tool. The algorithm’s ability to be easily modified, while only relying on satellite data inputs, gives it potential for future use in the field.