AN OBSERVING SYSTEM SIMULATION EXPERIMENT FOR THE LASER ATMOSPHERIC WIND SOUNDER (LAWS)

Citation
Gd. Rohaly et Tn. Krishnamurti, AN OBSERVING SYSTEM SIMULATION EXPERIMENT FOR THE LASER ATMOSPHERIC WIND SOUNDER (LAWS), Journal of applied meteorology, 32(9), 1993, pp. 1453-1471
Citations number
46
Categorie Soggetti
Metereology & Atmospheric Sciences
ISSN journal
08948763
Volume
32
Issue
9
Year of publication
1993
Pages
1453 - 1471
Database
ISI
SICI code
0894-8763(1993)32:9<1453:AOSSEF>2.0.ZU;2-V
Abstract
A series of observing system simulation experiments (OSSEs) was conduc ted to assess the potential impact of the Laser Atmospheric Wind Sound er (LAWS) instrument on a 5-day forecast using the Florida State Unive rsity (FSU) primitive equation multilevel spectral global circulation model. This proposed Earth Observing System satelliteborne instrument is a CO2 Doppler lidar wind sounding system. The instrument's requirem ent for usable measurements is that clouds or high concentrations of t ropospheric aerosols must exist within the sample volume. Two differen t orbits, a 55-degrees inclined and a 98-degrees sun synchronous, were examined by adding simulated LAWS wind profiles into a global four-di mensional data assimilation system and comparing the analyses and fore casts to a control experiment. Also, two different sets of simulations were examined for the 55-degrees inclined orbit. The first set follow ed the assumption of other previous Doppler lidar wind sounding OSSEs; a global concentration of aerosols exists such that observations will be usable at each pulse location. The second set of simulations incor porated the effects of subvisual cirrus, aerosols, molecular attenuati on, and sampling-scale turbulence. All simulations of LAWS wind observ ations are degraded when the lidar pulse encounters the earth's topogr aphy and when the downward-integrated cloud amount reaches a critical threshold. The four-dimensional data assimilation system consists of a multivariate optimum interpolation analysis and a nonlinear normal-mo de initialization using the aforementioned FSU global circulation mode l. In this set of assimilations only upper-air data was used with the exclusion of temperature sounder data, which may reduce the overall sk ill of the forecasts in the largely data-void Southern Hemisphere, whi ch has been seen in temperature sounder OSSEs. The inclusion of LAWS w ind observations exhibits an overall improvement of the forecast skill for this study. The greatest increase in skill is in the Southern Hem isphere, as can be seen in both the motion and mass fields. The 98-deg rees sun-synchronous orbit resolved the polar meteorological features much better than the 55-degrees inclined orbit. Otherwise, the two dif ferent orbits were very similar, with the 55-degrees inclined showing a better analysis of the tropics. A comparison of the two different se ts of simulations for the 55-degrees inclined orbit is also revealing. The incorporation of physical effects into the LAWS simulations shows a reduced forecast skill as compared with the other set of simulated LAWS observations. However, a global increase in forecast skill can st ill be seen over that of the control experiment. This is an attempt at creating more ''realistic'' simulations, and it is difficult to draw robust conclusions without completing a large number of forecasts, exa mining many different meteorological phenomena. In the future, the atm ospheric physical effects will be much better incorporated into the LA WS simulated observations. The analysis of Global Backscatter Experime nt data will aid in a better definition of global concentrations of ae rosols and will attempt to relate specific concentrations of aerosols to chemical composition and associated large-scale meteorological feat ures.