Convective Clouds and Interactions with Aerosols: A Perspective from Satellite Observations
Convective clouds are a dominant type of clouds that are particular susceptible to local influences of the
boundary condition including aerosol. Taking advantage of extensive satellite observations made by a suite of
state-of-the-art sensors (MODIS, CLOUDSAT/CALIPSO), we systematically investigate the behaviors and
statistics of both shallow and deep convective clouds (DCC) and their interactions with aerosols. For shallow
clouds, cloud particle size may increase or decrease with aerosol loading pending on cloud liquid path,
atmospheric and boundary conditions, and aerosol properties. For DCC, we find that the distribution of the
frequency of occurrence for cloud optical depth show a remarkably inter-annual stability for a particular region,
but it shows more variability from one region to another. Ice particle size generally depends on temperature.
Surface elevation also has important impact. The variability in the dependence of ice particle size on
temperature can be explained to the first order by a simple thermodynamic model. Both the freezing level and
particle size of the DCC cloud particles show significant dependence on aerosol loading that have not been
accounted in most, if not all models.
Tropical High Clouds and Their Relationships with Precipitation and Radiative Feedback
This study uses measurements from multiple Tropical Rainfall Measuring Mission (TRMM) sensors, namely Clouds and the Earth's Radiant Energy System (CERES), TRMM Microwave Imager (TMI) and Visible and InfraRed Scanner (VIRS), to estimate basic properties of tropical high clouds that affect radiative fields. Generally, there are two kinds of the high clouds: individual cirrus clouds (ICC) and deep convective systems (DCS). For DCS, both precipitation and rainfall efficiency increase with SST, indicating that DCSs more effectively remove the moisture transported into the upper troposphere when the temperature gets warmer. Despite increasing rainfall efficiency, the cloud area coverage rises with SST at a rate of ~7%/K in the warm tropical seas. There is little evidence that greater rainfall rates caused by increasing SST would reduce the cloud size and detrainment of DCS as suggested by some previously hypothesized dehydration scenarios for warmer climates. Besides the area coverage, cloud IWP also increases with SST. Instantaneous ICC has little or even negative correlation with DCS in small areas. When spatial and temporal domains are increased, ICCs become more dependent on DCSs due to the origination of many ICCs from DCSs and moisture supply from the DCS in the upper troposphere for the ICCs to grow, resulting in significant positive correlation between the two types of clouds. The estimated radiative feedback due to the change in tropical high cloud area coverage with sea surface temperature appears small and about -0.14 Wm-2K-1, which would not cancel out the estimated anthropogenic forcing of doubled atmospheric CO2.
Evaluation of NASA GISS SCM simulated Deep Convective Clouds using the integrated surface and satellite observations at the ARM SGP site
A total of 101 convective cases has been selected from 1999 to 2001 using collocated ARM MMCR and WSR- 88D reflectivity measurements and GISS Single Column Model (SCM) simulations over the ARM SGP site. By collocating the WSR-88D volumetric scan with the MMCR, we reconstruct the WSR-88D data into a time-height series directly over the SGP that match the MMCR data format to provide a complete time-height series of vertical reflectivity profiles. Due to the highly variable spatial and temporal nature of deep convection, data from two WSR-88D radars are mapped onto the GOES retrieved cloud-top properties in the domain to provide three dimensional observations of cloud and precipitation. MMCR and GOES retrievals are used to quantify the representation of vertical and areal cloud fraction of the WSR-88D. Using this integrated 3-D radar and satellite dataset and the NASA GISS SCM model simulation, we will address the following two questions: 1) Can the GISS SCM statistically simulate these convective systems during the 3-yr period? 2) Can we quantitatively evaluate the SCM-simulated convective and stratiform cloud fractions (CF), as well as their associated water and energy budgets?
Evaluation of GISS SCM Simulated Cloud and Radiative Properties Using Both Surface and Satellite Observations
To evaluate the GISS SCM simulated cloud fractions, three years of surface and GOES satellite data have been collected at DOE ARM Southern Great Plains (SGP) site during 1999-2001. The GOES derived total and high cloud fractions from both 0.5° and 2.5° grid boxes are in excellent agreement with surface observations, suggesting that the ARM point observations can represent large areal observations. Compared to the ARM radar-lidar observed cloud fractions, the SCM simulated most mid-level clouds, overestimated low clouds, and underestimated total and high clouds with additional missed during the summer season. Further studies have revealed that the model simulated cloud fractions are strongly dependent on the large-scale synoptic pattern and its associated variables such as vertical motion and relative humidity. Because a significant amount of clouds over ARM SGP occur during synoptically quiescent conditions, the model has issues producing enough high cloud cover. This work suggests that alterations need to be made to the stratiform cloud scheme to better represent the sub-grid scale cloud variability in this case. The model simulated radiation budget is also evaluated with two years of collocated ARM surface radiation and CERES and GOES TOA radiation over the SGP site during March 2000-Dec. 2001. For this comparison, the model simulated surface and TOA radiation budgets agree well with surface and satellite observations (∼10 W m-2). Model simulated cloud optical depth, however, is about an order of magnitude higher than CERES/GOES retrievals, which may explain why the radiation budget is reasonable and yet total cloud fraction has a negative bias compared to observations. Further study is warranted to better understand how this impacts cloud radiative forcing.
Cloud radiative effect on tropical troposphere to stratosphere transport represented in a large-scale model
GFDL AM2 model simulations are analyzed to assess the simulated radiative effect of tropical tropopause layer (TTL) cirrus on tropical troposphere-to-stratosphere transport (TST). The strongest upward motion in the model's TTL is generally driven by dynamics instead of radiation, occurring in those TTL cloudy regions that overlap with optically thick clouds in the upper troposphere (UT). However, the occurrence frequency of such strong ascent is about one order of magnitude smaller than that of moderate ascent related to the radiative effect of TTL cirrus. The mean upward velocity of moderate ascent in the cloudy regions (~ -2.5— -3.5 hPa/day) is one order of magnitude larger that that induced by TTL clear-sky radiative heating (~ -0.18 hPa/day). This supports the hypothesis that cirrus radiative heating contributes substantially to the average tropical TST rates. The implication for future model-satellite comparisons is discussed.
The Interaction between Clouds and Radiation Processes according to the NASA GEWEX SRB Release 3.0 Dataset
The NASA Global Energy and Water-cycle Experiment (GEWEX) Surface Radiation Budget (SRB) project data Release 3.0 covers a continuous 24-year period from July 1983 to June 2007. In addition to shortwave/longwave downward/upward fluxes at the Earth's surface and the top of atmosphere (TOA), this satellite-based dataset also provides cloud fraction, cloud optical depth and aerosol optical depth. The satellite covers the entire globe every 3 hours, and the highest temporal resolution of the dataset is thus 3hours. Three-hourly-monthly, daily and monthly means are derived therefrom. The spatial resolution of the dataset is 1 degree by 1degree. The dataset has been extensively validated against the datasets of the Baseline Surface Radiation Network (BSRN), the World Radiation Data Centre (WRDC) as well as the Global Energy Balance Archive (GEBA). Fairly good agreement has been achieved. This dataset provides a wealth of opportunities to study how the clouds and radiation processes in the atmosphere interact with each other and with other factors on regional and global scales. Such studies may enable us to better understand the dynamics of the global climate system. In this study, we first present SRB-BSRN comparisons under various cloud conditions, including clear-sky and all-sky conditions, as a way to validate the GEWEX SRB dataset. We then present some climatological statistics of shortwave/longwave radiation fluxes at the Earth's surface and TOA, cloud fraction, and cloud optical depth. We will attempt to show how these variables affect each other dynamically according to the NASA GEWEX SRB Release 3.0 data.