A31E-0153
Factors controlling cloud microphysics, precipitation rate, and brightness temperature of tropical convective and stratiform clouds
This paper discusses factors controlling cloud microphysics, precipitation rate and brightness temperature of tropical convective and stratiform clouds. Tropical convective and stratiform clouds are important in radiative forcing of climates and distribution of precipitation over the ocean. The possible effects of climate change on these clouds are still not well understood. Recent studies show that the higher CCN concentration in a convective cloud can lead to more vigorous updrafts and a higher evaporation/precipitation ratio. The stronger updraft often means stronger downdraft and gust fronts, which can trigger convection nearby. This implies that increases in CCN concentration can result in an increase in area coverage and persistence of tropical cirrus and stratiform clouds. The increased cloudiness would then be expected to lower sensible and latent heat flux from the ocean by lowering sea surface temperature, affecting the future development of convective clouds. The sea surface temperature may also change in a local area due to change of ocean circulation in climate change scenarios. Satellite remote sensing is a powerful tool to study tropical and global precipitation distribution. Many physically-based passive-microwave (MW) satellite precipitation algorithms make use of cloud radiation databases (CRDs), which typically consist of microphysical profiles from cloud resolving model (CRMs) and simulated MW brightness temperature (Tb). Thus, it is important to validate Tb simulated by a CRM against the observed Tb. Also, it is important to study how any changes in the tropical clouds due to aerosols and sea surface temperature translate into the precipitation and brightness temperature. The case study chosen is KWAJEX campaign that took place from 23 July to 14 September 1999. Authors have developed microphysical physical framework (Advanced Microphysics Prediction System) to predict ice particle properties explicitly in a CRM (University of Wisconsin-Nonhydrostatic Modeling System) (Hashino and Tripoli, 2007). AMPS also predicts aerosol and liquid spectrum by explicitly resolving sizes. For this study UW-NMS AMPS is set up for 2D simulation with periodic conditions over KWAJEX campaign area with synoptic forcing. The microphysical prediction of AMPS is then validated against in-situ microphysical observations and TRMM TMI measurements. Finally, sensitivity tests to study effects of aerosol properties and sea surface temperature on precipitation rate and Tb are discussed.
A31E-0154
Sensitivity of WRF-modeled Hurricanes to the Parameterization of Microphysical Processes: Can Satellite Observations Help Determine Which Parameterizations Produce the Most Realistic Storms?
Improving our understanding and forecasting of hurricane track and intensity - particularly sudden intensification and weakening - remains a challenge for the operational and research communities, and a significant amount of work remains to be done in validating hurricane forecast models, understanding their sensitivities and improving their parameterizations. None of this can be accomplished without a comprehensive set of multiparameter observations that are relevant to both the large-scale and the storm- scale processes in the atmosphere and in the ocean. In this study we use the Weather Research and Forecasting (WRF) model to simulate hurricanes and to gain understanding of the sensitivity of the modeled storms to the representation of convective and microphysical processes. More importantly, we address the question of whether satellite observations carry enough information to help distinguish between the different simulations. To facilitate hurricane research, we have developed the JPL Tropical Cyclone Information System, which includes a comprehensive set of multi- platform and multi-sensor observations that are relevant to both the large-scale and the storm-scale processes in the atmosphere and in the ocean. In this presentation, we will illustrate how the information system can be used for hurricane research and applications. Many factors determine a tropical cyclone's intensity, such as the vertical shear of the environmental wind, upper oceanic temperature structure, and low- and mid-level environmental relative humidity. Ultimately, though, intensity and rainfall are dependent on the magnitude and distribution of the latent heating and cooling within the storm that take place during the convective process. Hence, the microphysical processes and their representation in hurricane models are of crucial importance for accurately simulating hurricane intensity and evolution since they represent the phase changes of the water and the associated hydrometeor production and latent heating/cooling. The buoyancy of the air, generated by the released latent heat, drives the vertical motion and determines the storm's intensity. The vertical distribution of the latent heat source determines the vertical structure of the storm and its interaction with the large-scale environment, thus affecting its track. The accurate model representation of the microphysical processes becomes increasingly important when running high-resolution numerical models that should properly reflect the convective processes in the hurricane eyewall. We focus on the forecast uncertainty that is created by the convective and microphysical parameterizations. We will compare and contrast high-resolution WRF ensemble model simulations of hurricanes Rita (2005), Helene (2006) and Gustav (2008). For each of the storms, each member of the ensemble will represent a realization with different microphysical/convective assumptions. We will also address the importance of model resolution. We will use satellite observations from TRMM, QuikSCAT, CloudSAT, AIRS, MLS, GPS and others to evaluate errors in WRF forecasts and to determine those parameterizations that yield a realistic forecast and those parameterizations that do not.
A31E-0155
Structure of the Meiyu Frontal System on 7-8 July 2007: Comparison between Cloud- resolving Simulations and Observations
Meiyu frontal system that formed and moved over the Huai River basin of China on 7-8 July 2007 is investigated using weather radar observations, multiple satellite products, and surface meteorological and re- analysis data. Results from this analysis are used to evaluate cloud-resolving simulations with a focus on the vertical structure of hydrometeors. The convective systems consisted of generally west-east oriented leading convective lines with stratiform cloud regions trailing off to the north and east. Newer convection occurred on the western edge of the line. The convective centers progressed through a period of rapid growth, with echo tops penetrating to maximum heights of 16-km, then decreasing to height of 13-km, which corresponds to the height of the stratiform clouds with which the convective elements merged at the end of their lifetimes. The CONTROL simulation reasonably reproduced the spatial distribution of accumulated surface precipitation as well as the leading convective lines and trailing stratiform precipitation regions revealed by the observations. However, compared to the observations, the simulated convective cells are bigger and more intense extending higher into the troposphere while the stratiform regions are narrower. Significant differences are revealed between the vertical structure of the observed and simulated hydrometeors. Possible reasons for this discrepancy are presented. Sensitivity experiments are conducted to explore the impacts of microphysical and boundary layer processes on the simulated hydrometeor structure.
A31E-0156
Evaluations of microphysics schemes from the microwave rainfall measurement perspective
Passive microwave remote sensing of precipitation has been successfully used to monitor the global hydrologic cycle and rainfall retrieval algorithms continue to improve for accurate measurement. In the framework of current retrieval algorithms, the cloud resolving model (CRM) and its microphysical processes play an important role together with the radiative transfer model and their inversion technique. In this study, we construct various a-priori databases using the Weather Research and Forecasting (WRF) model with four different microphysics schemes. Due to different characteristics of microphysical processes especially in their frozen hydrometeors, retrieval results of precipitation fields and rainfall amounts are found to be different. This study discusses a sensitivity of microwave rainfall retrievals to microphysical parameterization in the CRM.
A31E-0157
Improving Microphysical Processes in the GCE Model Using TRMM and Other Observations
Advances in computing power allow atmospheric prediction models to be run at progressively finer scales of resolution, using increasingly more sophisticated physical parameterizations and numerical methods. The representation of cloud microphysical processes is a key component of these models. Over the past decade both research and operational numerical weather prediction models [i.e., the Fifth-generation National Center for Atmospheric Research (NCAR)/Penn State University Mesoscale Model (MM5), the National Centers for Environmental Prediction (NCEP) Eta, and the Weather Research and Forecasting Model (WRF)] have started using more complex microphysical schemes that were originally developed for high-resolution cloud- resolving models (CRMs). Recently, satellite observations in conjunction with a satellite simulator are used to identify the strengths and weaknesses of CRMs' bulk microphysical schemes based results from more detailed spectral bin microphysics. Consequently, these microphysics could be improved through detailed comparisons with observations. In this paper, we will present several improvement to the bulk microphysics, including (1) incorporating the effects of ice nuclei (IN) and ice crystal concentration, (2) replacing the Fletcher curve with the Meyers curve for the specification of ice crystal number concentration in the parameterization for the Bergeron conversion of cloud ice to snow, and (3) incorporating a new temperature-dependent DSD (TeDD).
A31E-0158
Numerical Simulations of a Snow Storm Using the Goddard Cloud Microphysics Scheme with the WRF Model and the Comparison with Ground and Satellite Radars
One of the grand challenges of the Global Precipitation Measurement (GPM) mission is to improve precipitation measurements in mid- and high-latitudes during cold seasons through the use of high-frequency passive microwave radiometry. For this, the Weather Research Forecast (WRF) model with the Goddard microphysics scheme is coupled with the Satellite Data Simulation Unit (WRF-SDSU) that has been developed to facilitate the over- snowfall retrieval algorithm by providing virtual cloud library and microwave brightness temperature (Tb) measurements consistent to the GPM Microwave Imager (GMI). This study tested the Goddard cloud microphysics scheme in WRF in snowstorm events (January 20-22, 2007) over the Canadian CloudSat/CALIPSO Validation Project (C3VP) site in Ontario, Canada. In this meeting, we will present the performance of the Goddard cloud microphysics scheme both with 2ice (ice and snow) and 3ice (ice, snow and graupel) as well as other WRF microphysics schemes. Results will be compared with the King Radar data. We will also use the WRF model outputs to drive the Goddard SDSU to calculate radiances and backscattering signals consistent to satellite direct observations. These simulated radiance are evaluated against the measurement from A-Train satellites. Note that the Goddard cloud microphysics scheme is now officially included in the WRF V3.
A31E-0159
Goddard Satellite Data Simulation Unit: A Comprehensive Multi-sensor Satellite Simulators for Supporting Satellite Missions
Variability of cloud-precipitation processes affect earth's energy and water budgets and atmospheric circulation. Different aspects of cloud-precipitation structures and properties are being observed from a variety of satellite instruments at present, including A-train constellation satellites (e.g., Aqua, Aura, CloudSAT, CALIPSO, and PARASOL) as well as other single-platform multi-sensor satellites (e.g., TRMM and Terra). Therefore, a combination of multi-platform and multi-frequency satellite observations can provide a more complete view of cloud-precipitation processes, ranging from cloud formulations, to coalescence processes, to the onset of precipitation. On the other hand, it becomes great challenge for scientists to analyze a large number of different satellite observations or products that usually assume different retrieval assumptions and complexities. In order to facilitate this effort, a comprehensive unified satellite simulator, Goddard Satellite Data Simulation Unit (SDSU), is being developed at the NASA Goddard Mesoscale Dynamics and Modeling Group. The Goddard SDSU is the end-to-end multi satellite simulator unit that can compute satellite-consistent radiance or backscattering signals from visible to microwave spectrum ranges based upon the simulated atmosphere and condensates consistent to the microphysics from a variety of cloud models. The Goddard SDSU currently includes passive microwave, radar, lidar, passive visible- infrared, broadband, and ISCCP-like simulators. These simulated radiances and backscattering can be directly compared with the high-resolution satellite direct observations in order i) to establish the satellite- based cloud-parameterization (or microphysics) evaluation framework, ii) to support radiance-based data assimilation, and iii) to provide a priori data files to satellite-based algorithm developers.
A31E-0160
Evaluating Goddard Multi-scale Modeling Framework Through the TRMM Triple-sensor Three-Step Evaluation Framework
This study presents a methodology known as the Tropical Rainfall Measuring Mission (TRMM) Triple-Sensor Three-step Evaluation Framework (T3EF) for the systematic evaluation of precipitating cloud types and microphysics in a cloud-resolving model (CRM). T3EF utilizes multi-sensor satellite simulators and novel statistics of multi-sensor radiance and backscattering signals observed from the TRMM satellite. Specifically, T3EF compares CRM and satellite observations in the form of combined probability distributions of precipitation radar (PR) reflectivity, polarization-corrected microwave brightness temperature (Tb), and infrared Tb to evaluate the candidate CRM. T3EF is used to evaluate the Goddard Multi-scale Modeling Framework (MMF), which is the NASA GEOS finite-volume GCM coupled with two-dimensional Goddard Cumulus Ensemble (GCE) models that explicitly resolves convective eddies and condensates at each GCM grid. Evaluation revealed that Goddard MMF captured the global distributions of shallow raining system well, while it over-predicted the occurrence of cirrus-overlapped cumulus congestus systems. T3EF also reveals that Goddard MMF over-predict the PR reflectivity and microwave Tb depression in the tropical warm pool. Unveiling the detailed errors in the Goddard MMF performance provides the better direction for model improvements.
A31E-0161
Improving a Bin Spectral Microphysical Scheme Using Long-Term TRMM Satellite Observations
9 years' TRMM squall line observations over the Southern Great Plain during late spring, early summer are compiled to provide validations for a cloud-resolving model. The Goddard Cumulus Ensemble (GCE) model coupled with the bin spectral microphysical scheme from the Hebrew University Cloud Model (HUCM) are used to simulate a squall line with leading convection/trailing stratiform type. Both ground radar and TRMM PR radar, as well as TMI 85G brightness temperature show consistent features in squall lines, especially in the stratiform region. Using forward radiative transfer calculations, the GCE model simulated radar and brightness temperature are calculated using particle size distributions simulated by the bin spectral microphysical scheme. Comparisons between the TRMM data and model simulation reveal discrepancies in the stratiform region. Improvements in ice particle collisional efficiencies and empirical ice particle partitioning in the bin microphysical scheme significantly improve the comparisons between the model and observation. This study shows the significance of multi-sensor satellite observations in validating model physics.
A31E-0162
Using TRMM to study Severity of Deep Convection
The Tropical Rainfall Measuring Mission (TRMM, Kumerow, et al. 1998) precipitation radar (PR) detected precipitation feature level-2 database (Liu et al. 2008) provides, among various other data, complete radar, microwave temperature, lightning flash rate and NCEP reanalysis thermodynamic summary data for over 2.1 million precipitating storms between the beginning of 1998 and the end of 2007. Here these data are used to further the studies of Brooks et al. 2003 and Brooks et al.2007, who propose using the region of thermodynamic space which a storm occupies to forecast the storm's severity. Here, we show that though more and less severe storms occupy different regions of thermodynamic space, more intense storms occupy a subregion of the less intense storms' region. As such, we believe we can expand their model by examining troposphere depth and orography as leading additional causes. In doing so, we examine multiple TRMM- based indices of storm severity to objectively identify qualitative differences among severe storms and to attribute these different indices to their environmental factors.
A31E-0163
A Radar Climatology of Tropical Anvil
The vertical structure of reflectivity observed by the Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR) and CloudSat Cloud Profiling Radar (CPR) is used to investigate the geographical distribution and temporal variability of tropical anvil (i.e., thick, non-precipitating cloud associated with deep convection). Based on 10 years of TRMM measurements from 1998 to 2007, anvil observable by the PR occurs 0.2% of the time across the tropics and has an average echo top about 7.7 km and an average thickness about 2 km. In order to quantify the amount of anvil that the PR is missing, coincident PR-CPR overpasses were examined. Statistics show that the TRMM PR underestimates anvil top height by 1 to 10 km with an average of around 5 km. A similar calculation concerning areal coverage is underway. Anvil occurs more often and is higher and thicker over land than over the ocean. Some tropical land regions, especially those affected by monsoon circulations, experience significant annual variability in anvil properties. Strong interannual anvil variability occurs over the central Pacific due to ENSO. Geographical and temporal anvil variations can be related to precipitation characteristics, as well as features of the large-scale environment (e.g., shear, humidity, and aerosols). For example, the anvil to rain area ratio is highest over the Atlantic ITCZ, an area that is affected by Saharan dust transport. The ability to observe and model the occurrence and vertical extent of anvil is essential in capturing the energy and water budget of the tropics.
A31E-0164
A-train Data Depot (ATDD) Providing users with convenient display and download services for collocated A-train instrument data.
The Goddard Earth Sciences DISC (Data and Information Services Center) actively supports A-train mission
researchers by providing display and data download access to a substantial number of cloud/aerosol,
temperature and pressure parameters measured by multiple sensors for platforms in the Atrain satellite
constellation.
Instruments supported include Cloudsat, CALIPSO, MODIS, AIRS, OMI, MLS, and POLDER together with
model data from GDAS and ECWMF with temporal coverage June 2006 through present. Our Giovanni tool
provides users with the capability of accessing, displaying and downloading subsetted multi-parameter data
which has been automatically collocated both spatially and temporally with the Cloudsat instrument's sub-
orbital track.
Image inter-comparison products are provided for both vertical profiles and narrow horizontal data swaths.
This subsetted data may be downloaded in HDF4, PNG or Google Earth KMZ file format. Users may also
download time series collocated data from an FTP site. Sample cloud precipitation products measured by
multiple A-train instruments will be presented.
http://disc.gsfc.nasa.gov/atdd/index.shtml
A31E-0165
Analysis of High Temporal Resolution Satellite Data for Sub-Tropical Boundary Layer Clouds Along a Pacific Ocean Crossection
The representation of clouds in numerical weather and climate prediction models is one of the most
fundamental unresolved issues facing the atmospheric modeling community. In particular, cloud
parameterizations are unable to simulate the hydrological cycle in a thermodynamically consistent manner.
This is an unfortunate situation since clouds are known to have a profound influence on the physics and
dynamics of climate. In particular, clouds associated with the boundary layer and deep convection play a
significant role in determining the tropical and sub-tropical atmospheric circulation.
Geostationary satellite-based observational datasets are capable of capturing the space and time evolution
of clouds on scales similar to current cloud resolving models (5-km and 15 minutes). We examine a two year
(2007-2008) collection of Geostationary Operational Environmental Satellite (GOES-11) data along a cross
section within the northern Pacific Ocean, using a multispectral 3-channel cloud properties retrieval
technique. The retrieved cloud variables of interest are cloud top temperature, cloud top effective radius and
cloud optical depth, from which liquid water path, emissivity, and cloud top height are derived. The
algorithm is designed to work in an operational setting, enabling comparison versus model produced cloud
fields every three hours. The task of retrieving these cloud properties is complicated by the nature of
passive-sensing imagers and therefore biased toward cloud top, in comparison with cloud profiling
capabilities of active-sensing (radar) systems such as CloudSat.
http://www.nrlmry.navy.mil/NEXSAT.html
A31E-0166
Precipitation from CloudSat: New Discoveries and Comparisons with Other Sensors
The CloudSat precipitation algorithm has yielded new information about where rain occurs over the planetary oceans and the distribution of light rainfall, particularly for rain less intense than about 5 mm h-1. CloudSat has the ability to actively-sense rainfall in the high latitudes, and is therefore a new addition to our suite of global precipitation observations outside the tropics. CloudSat measurements show that the high latitude oceans are amongst the rainest areas on the planet, particularly during the southern hemisphere winter. Rainfall distributions obtained from CloudSat are compared with measurements made by the Tropical Rainfall Measurement Mission (TRMM) Precipitation Radar (PR). Both direct matchups through orbital crossovers and regional, seasonal comparisons of the two sensors are considered. Implications for our understanding of the distribution of light rainfall are discussed. The CloudSat precipitation product is also compared with near-simultaneous measurements from the AMSR-E instrument, also part of the A-Train constellation, and ship-based rainfall measurements from the Comprehensive Ocean-Atmosphere Dataset (COADS). Future validation efforts planned for the CloudSat precipitation product are also discussed.
A31E-0167
CloudSat Tropical Cyclone Database
Since June 2006, CloudSat has made over 3000 overpasses of tropical cyclones providing a unique opportunity to explore the inner views of the core clouds and precipitation structure in tropical systems. In conjunction with Naval Research Laboratory and multi-sensor satellite observations, data extraction is performed along track from each storm specific CloudSat overpass. This presentation will describe the data assimilation techniques and data parameters contained in the CloudSat tropical cyclone database and provide a composite look of cloud and precipitation structure as a function of r, z for intensifying versus weak storms. Individual cases and groups will focus on low vs high vertical wind shear environment and how cloud fields are affected. Storms will be grouped and analyzed by sea surface temperatures, vertical wind shear, basin and year. These observations will be analyzed within the context of other observations obtained from A-train sensors.
A31E-0168
Climatologies of cloud water content distributions using CloudSat data
A crude treatment of cloud processes on sub-grid scales in current climate models is widely recognized as a major limitation in predictions of global climate change. Cloud parameterizations represent the effects of the sub-grid cloud processes on large-scale climate processes. The development of realistic cloud parameterizations requires accurate characterizations of sub-grid distributions of moist conserved thermodynamic variables. In an effort to improve cloud parameterizations, we characterized the sub-grid distributions of cloud water content (CWC) by applying a multivariate statistical method to CWC retrieved from CloudSat. We estimated the probability density function (PDF) of CWC using the Maximum Likelihood Estimation. The PDF of the liquid water content (LWC) follows a Gamma distribution while the PDF of the ice water content (IWC) seems to follow a superposition of two Gamma distributions. The best-fit parameters of the Gamma distributions to the PDF of LWC are obtained as a function of region, altitude, cloud type and season. The results show that the variations of the best-fit Gamma parameters among different cloud types are much larger than the seasonal variations. These results raise interesting science questions: Why do LWC distributions follow the Gamma function but IWC does not? What are the underlying physical mechanisms that cause the distribution differences between liquid and ice?
A31E-0169
Understanding Differences Between Co-Incident CloudSat, Aqua/MODIS and NOAA18 MHS Ice water Path Retrievals Over the Tropical Oceans
Accurate measurement of the physical and radiative properties of clouds and their representation in climate models continues to be a challenge. Model parameterizations are still subject to a large number of tunable parameters; furthermore, accurate and representative in situ observations are very sparse, and satellite observations historically have significant quantitative uncertainties, particularly with respect to particle size distribution (PSD) and cloud phase. Ice Water Path (IWP), or amount of ice present in a cloud column, is an important cloud property to accurately quantify, because it is an integral measure of the microphysical properties of clouds and the cloud feedback processes in the climate system. This paper investigates near co-incident retrievals of IWP over tropical oceans using three diverse measurement systems: radar from CloudSat, Vis/IR from Aqua/MODIS, and microwave from NOAA-18/MHS. CloudSat 94 GHz radar measurements provide high resolution vertical and along-orbit structure of cloud reflectivity and enable IWP (and IWC) retrievals. Overlapping MODIS measurements of cloud optical thickness and phase allow estimates of IWP when cloud tops are identified as being ice. Periodically, NOAA18 becomes co-incident in space / time to enable comparison of A-Train measurements to IWP inferred from the 157 and 89 GHz channel radiances. This latter measurement is effective only for thick convective anvil systems. We stratify these co-incident data (less than 4 minutes separation) into cirrus only, cirrus overlying liquid water clouds, and precipitating deep convective clouds. Substantial biases in IWP and ice effective radius are found. Systematic differences in these retrievals are considered in light of the uncertainties in a priori assumptions of PSDs, spectral sensitivity and algorithm strategies, which have a direct impact on the IWP product.
A31E-0170
Assessing the Value Added to Radar-Radiometer Cirrus Retrievals by Ice Water Path Derived From Sub-millimeter Radiometry
During the CRYSTAL-FACE and TC4 campaigns, observations of several cirrus cloud events were made. Several retrieval algorithms designed for use with A-train instruments are applied to these events for the computation of ice water content and of effective particle size profiles. These retrievals are further constrained by values of cloud ice water path obtained from a sub-millimeter wavelength retrieval in order to study the efficacy of such a constraint. To these ends, data from the Cloud Radar System (CRS), the Cloud Physics Lidar (CPL), the MODIS Airborne Simulator (MAS), the MODIS/ASTER Airborne Simulator (MASTER), and from the Compact Scanning Submillimeter-wave Imaging Radiometer (CoSSIR) aboard the NASA ER-2 are used; and the results of these retrievals are compared with coordinated in situ measurements.
A31E-0171
Sensitivity of Tropical Clouds to SST, RR, and Large-Scale Circulation
A number of studies have investigated the relationship between tropical deep convection and sea surface temperature (SST) and the implications of these relationships for climate and climate change. Deep convection and SST are somewhat indirectly related to each other through their individual relationships with the large-scale circulation, however, because regions of high SST are also regions of large-scale low-level convergence which encourages deep convection. Thus the intrinsic sensitivity of convection to SST is somewhat less apparent. Many previous studies use outgoing longwave radiation as a proxy for deep convection, but the ubiquity of thin cirrus disconnected from active deep convection in the tropics makes interpretation of observed relationships difficult. In this study we use rainfall data from the TRMM Multisatellite Precipitation Analysis as direct measures of deep convective events, SST from AMSR-E, cloud fraction, cloud top temperature, and cloud optical thickness from MODIS, longwave and shortwave TOA fluxes from CERES and vertical velocity from NCEP/NCAR reanalysis to address a number of questions: Within a given circulation regime (as determined by the magnitude and vertical shape of the vertical motion), how does the ensemble of cloud properties vary with SST, SST gradient, and precipitation rate? How do the observed changes in cloud properties and their respective radiative forcings affect the TOA and surface energy budgets? Likewise, how does the large- scale circulation respond to a change in SST gradient, and how does this circulation change impact the cloud properties? The sensitivity of convective evolution to the underlying SST will also be assessed by using a compositing technique centered on intense rain events. Implications of these results for tropical climate will be discussed.
A31E-0172
Analysis of Stratocumulus Droplet Size Distribution Variability From Multispectral and Multiangle Polarized Reflectances Observations
We are revisiting here the use of multidirectionnal observations of cloud polarized reflectances in the backscatter direction for liquid clouds microphysical properties retrieval. Cloud glories and cloud bows observations have a long record and have been theoretically and thoroughly investigated by numerous authors. Multiangle polarization measurements over liquid clouds have demonstrated a clear potential for inferring cloud effective radius and have been used successfully for this purpose. Comparisons have been made recently between cloud effective radius retrieved from POLDER (Polarization and Directionality of the Earth's Reflectances) data based on polarization measurements and more traditional retrievals performed using bispectral technique in the shortwave and near infrared from MODIS (MODerate resolution Imaging Spectroradiometer) data. Results of this study showed an apparent systematic discrepancy between the two techniques and conclusions were directed toward problems in the microphysical assumption used in the MODIS retrievals. We remind in this study some potential artifact in the use of polarization for cloud droplet size retrieval and present evidence that MODIS and POLDER retrievals, though different, can be reconciliated using simple single scattering considerations. In a second step, using full radiative transfer simulations based on a Monte Carlo polarized code, we examine how the vertical and horizontal variability of cloud droplet size distribution can impact total and polarized reflectances. From these results we draw perspective to characterize liquid clouds droplet size distributions from a combination of multispectral and multiangle polarized reflectances measurements provided by POLDER and MODIS observations.
A31E-0173
An Examination of AIRS data during overshooting deep convection events observed from MODIS, CloudSat and CALIPSO data
Instruments onboard A-Train satellites provide a unique dataset for examining tropical deep convection that penetrates the stratosphere (referred to as overshooting events), toward clarifying stratospheric- tropospheric exchange of water vapor. The tropical deep convection events are detected using the MODIS/CloudSat joint data browser. Overshoots are determined by examining data from CloudSat and CALIPSO cloud top heights and comparing these values to the height of the tropical tropopause level. CloudSat and CALIPSO transects are then collocated with MODIS data. The MODIS brightness temperature differences between the 6.715 micrometer water vapor absorption line spectrum and the 11.08 micrometer atmospheric window are determined for all cases, revealing positive ?TB values for each event with a maximum ?TB of 5.16 K. The overshooting events are then examined using radiances within the water vapor absorption band measured by the high spectral resolution infrared sounder, AIRS to examine the detailed spectral signature that produces the positive brightness temperature differences. Results of this evaluation will be presented and the implications of this data on upper tropospheric/lower stratospheric humidity will be discussed.
A31E-0174
Properties of Infrared Radiances at 3.7, 8.5, 11.0, and 12.0 μm Over Cirrus Clouds From MODIS Measurements
Infrared radiances over cirrus clouds are affected by many factors, e.g., cloud microphysical properties, optical properties, cloud top heights, cloud physical geometrical thicknesses, atmospheric profiles, and surface emission. These can result in uncertainties in retrieving cirrus cloud properties and simulating radiances over cirrus clouds at the infrared wavelengths. In this study, cloud properties from MODIS, CALIPSO, and CloudSat measurements, and atmospheric profiles from AIRS measurements are used to investigate the sensitivity of infrared radiances at 3.7, 8.5, 11.0, and 12.0 μm over cirrus clouds to various parameters including micro- and macro-physical, optical properties, and atmospheric and surface properties. The intent of this sensitivity study is to provide new insight for developing retrieval algorithms to estimate the microphysical and optical properties of cirrus clouds using the MODIS 3.7, 8.5, 11.0, and 12.0 μm bands.
A31E-0175
Comparison Between POLDER/PARASOL and MODIS/AQUA Operational Cloud Products.
Clouds play a very important role in climate and forecast models, but several uncertainties concerning their microphysical or macrophysical description are still persistent. Cloud amount and cloud properties are key parameters of the climate system and they need to be adequately monitored. Among the recent generation of Earth-orbiting instruments designed for Earth's observation, POLDER (Polarization and Directionality of the Earth's Reflectances) and MODIS (MODerate resolution Imaging Spectroradiometer) offer original capabilities that would allow improving the existing long term cloud climatology. The PARASOL mission was launched on December 18, 2004 and is orbiting in conjunction with other platforms of the A-Train constellation. This mission has been providing data from the third version of the POLDER instrument since March 2005. Almost four years of continuous data are now available from a POLDER type instrument providing quantitative and global measurements of total as well as polarized solar radiation reflected by the Earth-atmosphere system. The original measurement capabilities of POLDER (polarization, multi-viewing, multi-spectral) have been used to develop an improved set of algorithms in order to provide clouds physical and optical properties, together with water vapor and radiation budget parameters at global scale. The so- called "Earth Radiation Budget (ERB) and Clouds" processing line is now implemented at the French ICARE (Cloud-Aerosol-Water-Radiation Interactions) Center in Lille and the "ERB and Clouds" operational products are available to the scientific community. Here we present an overview of operational "ERB and Clouds" products and analyze monthly means of main cloud properties (cloud amount, cloud pressure, cloud optical thickness, etc ...) derived from POLDER and compare them to equivalent MODIS cloud parameters.
A31E-0176
Remote Sensing of Ice Cloud Microphysical Properties From Thermal Infrared Radiometry During the CIRCLE Experiment
Previous studies have reported that the two channels at 10.6 and 12 μm allow a determination of the optical thickness and effective size of ice cloud, for particle size less than 100 μm. In the context of the CALIPSO mission, a sensitivity study based on radiative transfer calculations has shown that the shape of non spherical ice particles and to a lesser extent, the size distribution, have noticeable effects on top-of- atmosphere brightness temperature. Using the split window technique (with channels at 10.6 and 12 μm) and assuming cloud structure knowledge, the influence of the cloud model leads to a theoretical accuracy of about 30 to 50 % on the retrieved effective size of the particle for small (< 20 μm) or large particle (> 20 μm), respectively. The accuracy on the retrieved optical thickness is rather good, of the order of +/- 10 %. This would however lead to an accuracy ranging from 10 to 50 % on the derived Ice Water Path. From a theoretical point of view, the use of the third channel (8.7 μm) should allow better constraining the cloud model, improving the determination of cloud properties. In this study, a method for ice cloud characterization from thermal infrared imagery is applied and evaluated. Infrared Imaging Radiometer (IIR) measurements at 8.7, 10.6 and 12 μm are used for reference scenes acquired during the CIRCLE experiment (summer 2007 over North / West of Europe). Indeed, validation flights have been performed with airborne measurements simultaneous with A-train ones. Especially, infrared radiances have been collected with the radiometer CLIMAT, which has similar infrared channel characteristics to IIR. Such a case study is now available including microphysics measurements as well as a detailed set of atmospheric data. Additional information about cloud structure is also available from airborne and space lidar (LNA and CALIOP) and radar (RASTA and CLOUDSAT) measurements. This database is of great interest to better constrain radiative transfer calculations. Comparisons between optical and microphysical properties of ice particles deduced from IIR or in situ measurements are presented. Contribution of the 8.7 μm channel to the cloud characterization is discussed.
A31E-0177
An Assessment of Cloud and Aerosol Radiative Forcing from CERES with Independent Passive (Ground Networks) and Active (Calipso Lidar) Measurements
Much of the interest in clouds and aerosols relates to the direct radiative forcing of these atmospheric
constituents. Satellite data are the main tools for estimating both cloud and aerosol forcing on the global
scale. What are the strengths and weaknesses of the satellite record for these forcings? Here we test the
CERES Terra (March 2000 - December 2006) and Aqua (July 2002 - December 2006) production records of
CRS, which includes computations of cloud and aerosol forcing at surface and TOA matched to broadband
satellite SSF observations, footprint by footprint. The ground-based data for this test spans 60 sites with
broadband measurements (ARM, BSRN, and SURFRAD) for the entire record. We use collocated AERONET
and MFRSR measurements of AOT at 8 of these sites. Our application of Calipso lidar covers only June-
December 2006.
CRS computes the Surface and Atmosphere Radiation Budget (SARB) with the Langley Fu-Liou code.
Essential inputs are MODIS-based clouds (Minnis et al.) and aerosols (Remer-Kaufman retrievals, buttressed
by the MATCH aerosol assimilation provided by Fillmore); and GEOS4 temperature and humidity. The first
generation of computed values for reflected SW and LW at TOA, when monthly and globally averaged, have
more stability than do the CERES observations; revised observations are more consistent with the original
calculations reported here. There are large errors in long-term mean computed cloud forcing to SW and LW
on a regional basis. The Calipso lidar shows that adjustments to retrieved cloud top altitude would improve
both SW and LW calculations.
As ground-based measurements cover only a point, short-term comparisons with computations based on
area-averaged satellite data for cloud forcing yield more noise than insight. But the interannual variability
(IAV) of SW insolation (and by inference, cloud forcing) computed by CRS and observed at ground sites is
very consistent. Confidence in the direct aerosol forcing to SW insolation is weak at most sites; but
surprisingly satisfactory for the seasonal forcing to downwelling LW over the few desert sites with
measurements.
http://snowdog.larc.nasa.gov/cave/