CIRRUS CLOUD RADIATIVE AND MICROPHYSICAL PROPERTIES FROM GROUND OBSERVATIONS AND IN-SITU MEASUREMENTS DURING FIRE 1991 AND THEIR APPLICATION TO EXHIBIT PROBLEMS IN CIRRUS SOLAR RADIATIVE-TRANSFER MODELING

Measurements from the FIRE 1991 cirrus cloud field experiment in the c entral United States are presented and analyzed. The first part focuse s on cirrus microphysical properties. Aircraft 2D-probe in situ data a t different cloud altitudes were evaluated for cirrus cases on four di fferent days. Also presented are simultaneous data samples from balloo nborne videosondes. Only these balloonsondes could detect the smaller crystals. Their data suggest (at least for midlatitude altitudes below 10 km) that ice crystals smaller than 15 mu m in size are rare and th at small ice crystals not detected by 2D-probe measurements are radiat ively of minor importance, as overlooked 2D-probe crystals account for about 10% of the total extinction. The second part focuses on the lin k between cirrus cloud properties and radiation. With cloud microphysi cal properties from surface remote sensing added to the microphysical data and additional radiation measurements at the surface, testbeds fo r radiative transfer models were created. To focus on scattering proce sses, model evaluations were limited to the solar radiative transfer b y comparing calculated and measured transmissions of sunlight at the s urface. Comparisons under cloud-free conditions already reveal a model bias of about +45 W m(-2) for the hemispheric solar downward broadban d flux. This discrepancy, which is (at least in part) difficult to exp lain. has to be accounted for in comparisons involving clouds. Compari sons under cirrus cloud conditions identify as the major obstacle in c irrus solar radiative transfer modeling the inability of one-dimension al radiative transfer models to account for horizontal inhomogeneities . The successful incorporation of multidimensional radiative transfer effects will depend not only on better models but critically on the ab ility to measure and to define characteristic inhomogeneity scales of cloud fields. The relative minor error related to the microphysical tr eatment is in part a reflection of the improved understanding on solar scattering on ice crystals over the last decade and of the available wealth on ice-crystal size and shape data for this study. In absence o f this information, uncertainties from microphysical cirrus model assu mptions will remain high.