Recent aircraft field campaigns in the Arctic have deployed microwave
radiometers suitable for cloud retrievals. The Beaufort Arctic Storms
Experiment (BASE, September - October 1994) included 15 research
flights in the Beaufort Sea region observing clouds over open water,
the marginal ice zone, and the permanent ice pack. The Surface Heat
Budget of the Arctic (SHEBA) and FIRE Arctic Cloud Experiment
(FIRE ACE) fielded two aircraft platforms with passive microwave
sensors during May 1998, and one platform during July 1998. The
SHEBA/FIRE-ACE flights were conducted in the vicinity of the SHEBA
ice camp, so surface based sensors such as a cloud radar and
lidar are also available for comparison with passive microwave
retrievals. The sensor of primary interest for this work is the
Airborne Imaging Microwave Radiometer (AIMR) with channels at 37 and
90 GHz. The advantages of an airborne radiometer compared to a
satellite sensor are vastly improved resolution and the availability
of coincident in situ measurements for comparison with retrievals.
A method for retrieving cloud liquid water path (LWP) from SSM/I data
has been adapted for use with data from AIMR. In this modified
algorithm, brightness temperatures at 37 and 90 GHz are used together
with aircraft based measurements of sea surface temperature, surface
air temperature, and average cloud temperature to estimate LWP in
liquid phase clouds over open ocean. Results of several cases from
BASE are compared with in situ measurements of integrated liquid water
content from a hot wire probe and a forward scattering cloud probe.
Typical LWP values for fairly homogeneous stratus clouds over the
Beaufort Sea are on the order of 100 g m^-2 or less. Isolated
convective clouds have estimated LWP as high as 300 g m^-2. Within
the estimated uncertainty of the algorithm and the limitations of the
cloud sensors, good agreement between the retrieved values and the in
situ estimates is obtained.
Retrieval of cloud properties over sea ice depends on the ability to
distinguish between clouds and the surface. Lower frequencies (i.e.,
37 GHz) penetrate clouds fairly efficiently while the signal at higher
frequencies is affected by the presence of clouds, so use of
multi-spectral data provides an indication of the presence of clouds.
Quantitative measurement of bulk cloud properties requires an estimate
of surface emissivity at the pertinent frequencies. Data from SHEBA
flights during clear conditions will be used to estimate surface
emissivity and characterize its spatial and temporal variability.