Microphysical modes of precipitation growth determined by S‐band vertically pointing radar in orographic precipitation during MAP

Abstract

AbstractHigh‐resolution vertically pointing S‐band Doppler radar data obtained within orographic rain by the University of Washington Orographic Precipitation Radar (OPRA) at Locarno‐Monti, Switzerland during the Mesoscale Alpine Programme (MAP) Intensive Observing Periods (IOPs) 2b and 8 are examined to characterize the fine‐scale precipitation structure. The relative roles of collection of cloud water by raindrops versus collection of cloud water by ice particles (riming) in the same column are assessed with a one‐dimensional (1D) Kessler water‐continuity model using the observed data characteristics as input.The OPRA data reveal nearly ubiquitous small‐scale fallstreaks each <5 minutes in duration occurring within a wide range of precipitation intensities. Periods of higher rain rates >10 minutes in duration associated with convective cells evident in coarser‐resolution data manifested as increased frequency and intensity of fallstreaks in the OPRA data. A subset of fallstreaks had sufficient magnitude in reflectivity to indicate a clear connection from the snow region, across the melting layer and into the rain region. The small‐scale inhomogeneities in microphysical structure associated with fallstreaks are likely to be associated with convective overturning in the unstable IOP 2b case. In the more stable IOP 8 case, shear‐driven turbulence near the melting layer, evident in truck‐borne and airborne X‐band Doppler‐radar data shown in companion studies, is a likely contributor to fallstreak formation.The vertical profile of vertical air velocity (w) is critical to understanding the relationship between collection and riming, and their contributions to precipitation efficiency. Characteristics of the w profile within precipitating cloud were estimated in both rain and snow using observed radar parameters from the OPRA data. In rain, reflectivity‐weighted fall speed was derived from observed reflectivity and assumptions about the raindrop size distribution. The derived fall velocity was subtracted from the observed Doppler radial velocity to estimate w in rain regions to within ∼1.5 m s−1. In snow regions, a lower bound on peak updraught velocity was estimated from the observed maximum Doppler velocity. For the data examined from MAP IOP 2b, typical peak updraught velocities at Locarno‐Monti were between 2 and 5 m s−1.These estimated peak updraught velocities were used to construct plausible parabolic profiles of w that were in turn used as input to a 1D Kessler water‐continuity model. Model output shows local maxima in the riming rate and mixing ratio of ice within 2 km of the freezing level, indicating a favourable environment for graupel formation. The location of this layer in the model is consistent with the observed occurrence of graupel in National Center for Atmospheric Research S‐Pol radar data in the IOP 2b storm in a companion study. The modelled rates of collection and riming are comparable, and both make significant contributions to the growth by accretion of precipitation within the storm updraughts. Within the rain layer, liquid‐water coalescence (autoconversion and collection) in updraughts with peak velocities of 5 m s−1 can account for up to ∼40% of column‐integrated cloud water.Our analysis of the observational data and the 1D model results confirms and extends previous work indicating that a combination of liquid‐water coalescence and ice‐phase processes is needed to obtain high precipitation efficiencies >40% in orographic precipitation. Copyright © 2003 Royal Meteorological Society.