Principal Investigator (PI): Tianle Yuan, University of Maryland Baltimore County (UMBC)
To build a novel and science-based Earth system data record of marine low-level and deep convective cloud objects and their environmental conditions. The unique elements of the record will include:
- A cloud objective-oriented approach with the best possible characterization of the environmental conditions by integrating measurements from multiple sensors and reanalysis/observation data at the native level-2 resolution;
- A long-term (30+ year) data record of marine low-level cloud objects with mesoscale organization classification based on a machine-learning method;
- A long-term (10+ year) data record of deep convective cloud objects based on a detect-and-spread method;
- An intensive observation period (IOP) product using A-Train passive and active sensors that allows process level analysis and more complete characterization of cloudy scenes.
Marine low-level clouds and deep convective clouds sit at the nexus of several Grand Challenges outlined by the World Climate Research Programme: Clouds, Circulation and Climate Sensitivity, Understanding and Predicting Weather and Climate Extremes, Near-Term Climate Prediction. They are also highlighted in reports produced by the U.S. Global Change Research Program. Their cloud feedback is at the heart of overall climate feedback uncertainty that is closely related to the magnitude of future warming. They are important for model biases in both near-term predictions and long-term projections. Their coupling with circulations (e.g. Madden-Julian Oscillation/MJO) are sources of predictability in intra-seasonal forecasting. Their interactions with aerosols are thought to be large and of opposite signs in terms of indirect forcing, and dominate current uncertainty in the overall anthropogenic forcing estimate. It is therefore imperative to improve understanding of two cloud regimes.
Motivated by this science focus, the project will create an integrated product that is customized for the science need. The goal is to produce an easy to use object-oriented cloud database that allows for interpreting key physical processes governing two cloud regimes. It will serve the community as a data infrastructure for in-depth analyses.
This project will go beyond individual data pixels to organize data around cloud objects that are organically generated by nature. Cloud objects are first classified into different categories of mesoscale organizations in the case of marine low-level clouds and detected in the case of deep convective clouds. Gross features of a cloud object such as cloud horizontal size, overall probability distribution of cloud variables, location and time will be summarized. The environment of a cloud object is then characterized to the best possible extent using multi-sensor multi-parameter collocated data at native resolution.
For example, in the IOP period, i.e. the A-Train era, relevant observations include liquid water path retrievals from Moderate Resolution Imaging Spectroradiometer (MODIS) and microwave sensors, aerosol concentration from MODIS and Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), cloud height retrieval from active and passive sensors, radiative fluxes from Clouds and the Earth's Radiant Energy System (CERES), cloud microphysics and optical depth from MODIS, key moisture and temperature variables from various sensors, aerosol retrievals from MODIS, radar reflectivity profile and precipitation retrievals from CloudSat, as well as meteorological variables derived from reanalysis dataset and sea surface temperature (SST).
These parameters are important for understanding cloud behaviors in these two regimes. They will provide the context to understand observed clouds and offer a database for model-observation comparisons. The long timespan also allows for trend analyses of cloud properties together with their environment. Therefore, we are creating an Earth system data record (ESDR) of marine low-level and deep convective cloud objects for community users to systematically study these two important cloud regimes.