With the addition of CloudSat and CALIPSO, the A-Train constellation of satellites will provide a unique cloud observing capability through a variety of passive and active instruments. We propose to use our experience in developing MODIS operational cloud products to compare the various A-Train cloud products and explore their inter-sensor consistency. The comparisons will be used to evaluate and improve existing MODIS algorithms and thereby extend information from the A-Train active observations to the imager swath. We will also pursue strategies for combining measurements from different sensors, specifically the MODIS, CALIPSO lidar and AIRS, to improve the retrievals of existing cloud products as well as retrieve new cloud information (e.g. asymmetry parameter of cirrus clouds).
While improved understanding of the biases and uncertainties inherent to passive cloud property retrievals is important for a broad range of science investigations, our focus will be on the radiative and physical properties of cirrus clouds, particularly thin cirrus (defined as an optical depth of less then 2).
Specifically, the investigation will include:
- Product comparisons: Collocation and comparison of operational MODIS Aqua cloud products with those from AIRS, POLDER (PARASOL), and the active instruments on CloudSat and CALIPSO. Products of primary interest include cloud detection, thermodynamic phase, cloud-top properties, multilayer cloud detection, liquid and ice water path, optical thickness, and effective particle size. Analysis will include a variety of cloud types, regions, and seasons. The analysis will be used to better understand the limitations of passive retrievals as well as provide a means for improving passive techniques, and extending these improvements to historical satellite data sets.
- Multi-instrument algorithms: In addition to comparing and characterizing the relative consistency of A-Train cloud products, we will initiate several studies to develop improved cloud products by combining passive and active observations. These studies will benefit from our experience in the development of research-level cloud retrieval algorithms. Cloud boundary information from the active sensors, particularly CALIPSO, will be used as a constraint in AIRS and MODIS retrievals. As a next step, we will develop fundamentally new products by combining the multi-instrument A-Train observations into a single retrieval using Bayesian or optimal estimation approaches. In addition to retrieving a spectrally unified optical depth, the retrieval approach estimates the asymmetry parameter of cirrus clouds.
In the coming decades, passive satellite sensors will continue to be a staple for cloud climatology studies, global cloud process investigations, as well as forecast and climate model validation, parameterization development, and assimilation. The proposed investigation will provide an improved understanding of past, current and future operational passive cloud property retrievals and provide a pathway for algorithm improvements.
