This project has two principal objectives. One was to enable the orbiting NASA QuikSCAT scatterometer radar to detect the presence of rain over the ocean, to estimate the amount of rainfall (rainrate, in mm/hr) and to create new approaches for reducing the interference that rain causes with QuikSCAT’s ability to estimate sea surface wind vectors. The second principal objective was the integration of QuikSCAT ocean wind vector data into the NWS AWIPS (All-Weather Information Processing System). This was the major operational objective for this project. It was to enable NWS forecasters to integrate QuikSCAT ocean winds with other meteorological data sets, both observational and stochastic, in near real time. There are significant advances in both of these objectives to report.
1. PROJECT ACCOMPLISHMENTS
This project has established the feasibility of using a satellite radar scatterometer as a meteorological instrument to estimate rain over the ocean. Measurements indicate that the radar backscatter from the rain volume in the atmosphere, which can exceed the sea surface radar cross section (at 14 GHz) in many situations, depend on the rainrate, the rain column height and surface wind speed. Algorithms for inverting the RCS were developed using methods of calibration and validation of the rainrate, with winds from buoys and rainrate from coincident NEXRAD and TRMM satellite data.. These studies also determined that the dual polarization of QuikSCATs’ SeaWinds radar instrument (H-pol and V-pol) enables the application of a highly sensitive technique for measuring rainrate, differential reflectivity which is commonly used in radar meteorology, to contribute to the algorithm. Calibration studies which combined the QScat, NEXRAD and TRMM rain measurements concluded that the presence of rainrates that exceed 1 mm/hr can frequently be detected and converted in rainrate estimates. This accomplishment has significant implications and applications for future global precipitation measurements over the oceans.
The examination of rain impacts on scatterometer winds has lead to an adequate general understanding and a physically based model of this interaction. This result is expected to lead to improved scatterometer products. In particular, it is now clear that flagging scatterometer observations for rain contamination is a far from an optimal approach. In many cases, the rain has little impact on the winds. Even in the case of very heavy rainrates, if the wind speed is great enough the wind signal will dominate the rain signal, resulting in acceptably small error in the winds. Such situations often occur around Tropical Storms, indicating a far greater than expected usefulness of the remotely sensed winds in Tropical Cyclone applications.
In collaboration with the COAPS/FSU and NOAA’s office of Research and Applications the New York WFO/OKX has evaluated the operational significance of the scatterometer data wind data, for routine use and more extensively during significant weather as determined by the office. Meteorologists have been trained and assisted to use the capabilities of AWIPS to compare traditional data sets with the scatterometer observations. The integration of QuikSCAT data into AWIPS was the major operational accomplishment for this project. As a result, NWS forecasters can now integrate QuikSCAT data with other meteorological data sets, both observational and stochastic, in near real time.
2. SUMMARY OF UNIVERSITY/NWS EXCHANGES
We have worked to make NWS forecasters aware of how observations from SeaWinds on QuikSCAT (QSCAT) can be used to aid in forecast assessment. We have worked primarily through Jeffery Tongue, and participated at national meetings, a severe winds workshop at the Tropical Prediction Center, and a web page (http://www.coaps.fsu.edu/~zierden/qscat/). This effort has lead to improved awareness and improved understanding in the use of QSCAT product. As a result of this collaboration, QSCAT winds and pressures are know available on a COAPS FTP site. These winds are now ingested into AWIPS.
The ingesting of the QuikSCAT data into AWIPS was not an easy task. The collaboration among Florida State University (FSU) and NWS Eastern Region Headquarters was key to our success in this effort. While the effort took numerous months to accomplish, the QuikSCAT data has been available in the WFO’s AWIPS routinely since January 2002. The instructions for ingest of the QuikSCAT data from the FSU server have been posted to the AWIPS Local Applications Database. To the best of our knowledge, four WFO’s (San Juan, Tallahassee, Honolulu, and Corpus Christi) and the NWS’s Spaceflight Meteorology Group at the Johnson Space Center are currently accessing the data. Other offices and the Marine Prediction Center have contacted us and we will continue to work to spread the utilization of QuikSCAT data throughout the NWS.. WFO San Juan, working with Weather Service Headquarters, is currently in the process of evaluating QuikSCAT for integration into the AWIPS baseline. A significant increase in the use of QuikSCAT data will come this hurricane season, now that the data is integrated with other data sets within AWIPS.
3. PRESENTATIONS AND PUBLICATIONS
Weissman, D.E., M.A. Bourassa and J. Tongue, 2002; Effect of rainrate and wind
magntiude on SeaWinds scatterometer wind speed errors, Journal of Atmospheric
and Oceanic Technology,
Vol. 19, No. 5, May 2002, pp 738-746
Weissman, D.E., M.A. Bourassa and J. Tongue, 2001; Effects of rain-rate and
wind magnitude on SeaWinds scatterometer wind speed errors over the ocean, IEEE
2001 International Geoscience and Remote Sensing Symposium, 9-13 July, University
of New South Wales,
Weissman, D.E., J. Tongue and M. Bourassa, 2001: Utilization of satellite scatterometer wind measurements and NEXRAD precipitation data to improve regional ocean forecasts. 18th AMS Conference on Weather Analysis and Forecasting, 30 July – 3 August 2001, Fort Lauderdale, Florida. pp 278-282
Zierden, D.F., M.A. Bourassa, J. Tongue, D.E. Weissman and J.J. O’Brien, 2001; Near-realtime winds and surface pressures from SeaWinds, AMS 11th Conference on Satellite Meteorology and Oceanography, 15-18 October, Madison, WI.
Weissman, D.E., M.A. Bourassa, J.J. O’Brien and J. Tongue, 2001; SeaWinds radar cross section measurements infer rainrate over the ocean, Specialist Meeting on Microwave Remote Sensing, 5-9 November 2001, Boulder, CO
Weissman, D.E., M.A. Bourassa, J.J. O’Brien and J. Tongue, 2001; Estimates of the rainrate over the ocean using SeaWinds radar cross section measurements. Fall Meeting American Geophysical Union, 10-14 December, 2001, San Francisco, CA
Tongue, J.S, J. Watson and M. Tauber, 2002: Integration of New Data Sets into AWIPS for Use in an Operational Forecast Environment. AMS AWIPS Symposium, 14-17 January 2002, Orlando, Florida.
4. SUMMARY OF BENEFITS AND PROBLEMS ENCOUNTERED
4.1 Benefits to NWS and Hofstra University
Lastly, it must be noted that the diverse team involved in this project has developed a long-term relationship that has and will continue to impacted many operational meteorologists through our presentations, papers and seminars. Working together, we have discussed the QuikSCAT data set with hundreds of individuals. This has increased knowledge on remote sensing and the use of polarized radar data.
Hofstra University gained from this collaborative project because a current research program, supported by the National Aeronautics and Space Administration to study the application of the QuikSCAT scatterometer, will benefit from having improved measurements of the sea surface radar cross section in the presence of rain. This project will be expanded with an additional inquiry into the physical processes that occur as the radar beam passes through a rain layer and interacts with the roughened surface, wherein the rain data can now be extracted from NEXRAD base and composite reflectivity files.
An Honor student at a local high school has decided to work with Dr. Weissman, using the subject of this COMET project and a NASA research investigation, to develop his own scientific studies for participation the Intel National Science Talent Search competitions, for CY 2003.
4.2 Operational Difficulties Encountered
The reliability and timeliness of the QuikSCAT data was a drawback to its use operationally. The most significant problem was data cut-offs that lost a portion of the pass. FSU has developed techniques to correct this and data are currently relatively reliable (approximately 80%), though additional work is still needed.
Forecasters require training on how to integrate QuikSCAT information into the forecast process. Forecasters generally relate QuikSCAT data to buoy data, which is available on an hourly basis – while QuikSCAT has a temporal resolution of 12 hours – similar to that of the rawinsonde network. This results in forecasters not making routine use of the data. With the launch of the next satellite – temporal resolution will improve to as much as 6 hours.
4.3 Future Considerations:
There is much work to complete in the operation arena related to our work. The correction for significant rainfall, part of the goal of this project, will assist with major cyclone analysis and prediction.
Improving on the ingest of QuikSCAT data into AWIPS is an effort we plan to continue to explore. Currently, AWIPS is only able to plot the wind barbs – we hope to ingest u and v derived grids of the QuikSCAT wind field so that additional diagnostics can be performed. Also, we hope to ingest the FSU produced derived pressure fields in the near future. A much greater interest has been expressed in the pressure product; however, the NWS support required put this product into the AWIPS system was not available. This product is easily compared to forecasts of surface pressure, with the goal of locating significant errors in the forecast. This approach is particularly useful during severe weather; however, without the data being in the AWIPS system the forecasters are not expected to have the time to make such comparison.
The application of QuikSCAT over land shows great promise. Remote samples of soil moisture and rainfall may prove extremely valuable to many research efforts including numerical weather prediction and climate change.