| RAMMB CIRA 4th Quarter Report |
|---|
July August September 2008 |
There was a plenary meeting to discuss various options for setting up a polar data server at CIRA. RAMMB has plans to begin pulling MIRS, IASI, SSMI and ATOVS data to support hurricane research, and to support a polar data server at CIRA. A centralized, near-realtime polar data server, with http, ftp, McIDAS server and LDM (for AWIPS feeds) access, and automated data subset archive, will be highly beneficial to research at CIRA. This data will also be a valuable resource for blended product development, Satellite Algorithm Testbed work, and for AWIPS Proving Ground product evaluation. The eventual goal will be to provide a robust server, RAID storage (with fail-safe backup) and a set of access routines that would be available to all CIRA and NOAA researchers and clients. Phase I will involve the set up of a RAMMB server and RAID array to create the DDS MIRS, IASI, SSMI and ATOVS data archive. This will get us ready for NPP and NPOES. (J. Knaff)
Three consecutive daylight periods (10-12 September; 12 UTC- 00 UTC) of SRSO imagery were collected over Hurricane Ike using the operational GOES-W satellite. The purpose was to attempt to capture either or both a period of rapid intensification or of secondary eyewall formation. This unique dataset does show the slow demise of the small remnant eye over the three day period and the strengthening and reorganization of the storm near landfall. Loops of the visible albedo product are available from ftp://rammftp.cira.colostate.edu/knaff/IKE/vis/avi. (J. Knaff)
Work continues on calibrating the World Wide Lightning Locator Network (WWLLN) dataset for use in the GOES-R Risk Reduction tropical cyclone project. We have successfully converted the WWLLN count data to flashes/km²/year for comparison to the TRMM dataset. We discovered that over the entire WWLLN dataset period (August 2003 – December 2007) for latitudes between 50ºS and 50ºN, the average flash density is 57 times smaller than the TRMM dataset (April 1995 – December 2005) over the same area. When this scaling is applied, the general patterns between the TRMM and WWLLN data are similar as shown in Figures 1 and 2 below, providing some confidence in the use of the WWLLN data. (R. DeMaria)
Figure 1: A plot of the TRMM flash density dataset for April 1995 - December 2005
Figure 2: A scaled plot of the WWLLN flash density dataset for August 2003 – December 2007 (Scale Factor: 57)
We have found that the quality of the WWLLN dataset is highly dependent upon the activation of WWLLN sensors. More lightning strikes are recorded near active WWLLN sensors. These sensors have been activated over the course of the WWLLN dataset’s lifetime. For example, Figures. 3 and 4 compare the scaled lightning density over the Atlantic for Jan of 2004 and Feb 2007 that were generated as part of a study of the influence of the Gulfstream on the atmosphere. The large differences in the distributions are due to the limited sensors available in 2004.
Figure 3: WWLLN Jan 2004 flash density data subset scaled by 7340 to match average TRMM flash density
Figure 4: WWLLN Feb 2007 flash density data subset scaled by 57 to match average TRMM flash density
The fact that the WWLLN dataset improves over time leads us to conclude that it is necessary to compute monthly calibrations for the WWLLN dataset. The generation of these monthly calibrations will be the next step on this project. Calibrations associated with a monthly time period as well as a spatial subset of the WWLLN dataset may also be useful. Future work will be devoted to generating calibrations that vary with time and space. (R. DeMaria)