5. Current Research
In this section, we highlight some recent research achievements of the Division. More detailed summaries are contained in the Division's bi-annual Reports.
Operational Products Development Branch
Aircraft Icing Product Achieves High Reliability
As a result of several upgrades to the GOES aircraft icing product, the probability of detecting hazardous icing conditions is now consistently high - in the 55-70% range compared to 40-65% previously - for the Continental U.S.
Figure 9: Example of an Icing Enhanced Cloud-top Altitude Product (ICECAP) image is shown, valid at 1700 UTC, on February 17, 2004. Areas of potential icing are color-coded in intervals of 6,000 ft to show maximum cloud top altitude. Pilot reports of icing are superimposed showing: numerical icing intensity (0 to 5), aircraft type, and altitude in feet. Severe icing (code 5) at 8,000 ft was reported in eastern Tennessee within two hours of the GOES product. Some icing (such as that shown in northwest U. S.) is obscured by high cloud layers and cannot be detected.
Significant Advance in Satellite Wind Measurements
Winds derived from tracking cloud and water vapor features in geostationary satellite observations have been produced for decades. However, because of the limitations of geostationary satellite viewing, these wind retrievals are not available for Polar Regions. To overcome this problem, a new capability has been developed that takes advantage of the frequent observations of Polar Regions by the MODIS on the NASA Terra and Aqua satellites. First developed at the Cooperative Institute for Meteorological Satellite Studies (CIMSS), it is based upon established methodologies and algorithms used to derive wind observations from the GOES series of satellites. MODIS cloud-drift and water vapor wind observations provide unprecedented coverage in the polar regions of the globe, areas where wind observations are sorely lacking. Figure 10 shows an example of the MODIS water vapor motion wind products in the Northern Hemisphere polar region. In 2004, the Operational Products Development Branch (OPDB) integrated the MODIS winds capability within the existing operational NESDIS winds processing system. Recent work by the JCSDA shows that these polar region winds have a positive impact on weather forecasts.
Figure 10: MODIS water vapor motion winds over the Northern Hemisphere polar region.
AIRS Data Significantly Improve Weather Forecasts
Experimental weather forecasts at the JCSDA using AIRS radiance observations indicate significant improvements in global forecast skill compared to the operational system without AIRS data. The improvement in forecast skill at 6 days is equivalent to gaining an extension of forecast capability of several hours. This magnitude of improvement is quite significant when compared to the rate of general forecast improvement over the last decade. A several hour increase in forecast range at 5 or 6 days normally takes several years to achieve at operational weather centers.
The AIRS impact study consists of two parallel series of 27 consecutive daily weather forecasts, each extending out to 10 days during the month of January 2004. To specify the initial conditions for each forecast, the control series assimilates all conventional and satellite observations except for AIRS observations; the experimental series assimilates all the data used in the control run plus the AIRS observations.
The skill of the forecasts was evaluated by comparing the forecasts with the verifying analyses of the observations using anomaly correlations. Anomaly correlation is a statistical measure for evaluating large-scale/medium-range forecast skill and provides a reliable indication of overall model skill. The anomaly correlation is the correlation between observed (verifying analysis) and predicted deviations from climatological conditions. It is clear from the accompanying figure that AIRS data have a consistent and significant effect on forecast skill.
Figure 11: 500hPa Z Anomaly Correlations with (Ops.+AIRS) and without (Ops.) AIRS data, Northern hemisphere, January 2004
Data, algorithms, and images presented on STAR websites are intended for experimental use only and are not supported on an operational basis. More information