HR: 0830h
AN: A21E-1034    [PDF]
TI: Scale Invariance of Precipitable Water Vapor in the Arctic From Ground-Based Radiometric Measurements
AU: * Cadeddu, M
EM: maria@sensor2.gsfc.nasa.gov
AF: JCET/UMBC, Greenbelt Rd, Greenbelt, 20771
AU: Racette, P
EM: per@priam.gsfc.nasa.gov
AF: NASA GSFC, Greenbelt Rd, Greenbelt, 20771
AU: Wang, J
EM: wang@sensor2.gsfc.nasa.gov
AF: NASA GSFC, Greenbelt Rd, Greenbelt, 20771
AB: This study gives a first assessment of the usefulness of Millimeter-wave Radiometer (MIR) brightness temperature measurements for studying the scale invariance in atmospheric Precipitable Water Vapor (PWV) and Liquid Water Path (LWP) distributions in the Arctic. It is shown that MIR data display well defined scaling properties at frequencies close to 183.3 and 89 GHz during both clear-sky and cloudy conditions for horizontal scales between 350 m and 350 km. The turbulent variability of PWV and LWP, in the extremely dry arctic environment, is characterized using three mathematical techniques involving increasingly higher statistical moments. The first technique is the Detrended Fluctuation Analysis from which one can determine the existence of long-range correlations and the Hurst exponent H for the time series. Then, spectral analysis, that is a second order statistics and relies on the assumption of a Gaussian distribution, is performed on the time series data to explore scaling properties through the spectral exponent b. Lastly, a multiplicative cascade model is applied to millimeter-wave measurements to describe intermittency features characteristic of non-homogeneous turbulent fields. The results for LWP during cloudy days are in excellent agreement with previous studies conducted in different environmental conditions giving H = 0.33 and b = 1.61. PWV data during clear sky days have a slightly higher average Hurst exponent and spectral exponent (H=0.57, b = 1.89). Both PWV and LWP have similar intermittency parameter: C1 ~ 0.1 and C1 ~ 0.06 respectively. The analysis shows that intermittency is an important feature of arctic water vapor variability that can not be captured by a second order statistics such as spectral analysis. Therefore water vapor should not be regarded as a passive scalar subject to homogeneous turbulence. Instead, it should be treated as a randomly advected tracer that presents a multifractal (anomalous) scaling. In the present experiment the data could be well fitted using a universal log-Levy cascade generator. These experimental results can be used to validate theoretical models describing the transport of water vapor at higher latitudes.
DE: 0300 ATMOSPHERIC COMPOSITION AND STRUCTURE
DE: 0365 Troposphere--composition and chemistry
DE: 0368 Troposphere--constituent transport and chemistry
DE: 0394 Instruments and techniques
SC: Atmospheric Sciences [A]
MN: 2003 Fall Meeting