Small Cumulus Research Articles

Size Distributions and Dynamical Properties of Shallow Cumulus Clouds from Aircraft Observations and Satellite Data

Abstract In this paper aircraft observations of shallow cumulus over Florida during the Small Cumulus Microphysics Study (SCMS) are analyzed. Size distributions of cloud fraction, mass flux, and in-cloud buoyancy flux are derived. These distributions provide information on the specific contribution of clouds with a certain horizontal size and reveal, for example, which size has the largest effect on cloud fraction or vertical transport. The analysis of four flights shows that the mass flux and buoyancy flux are dominated by intermediate-sized clouds (horizontal dimension of about 1 km). The cloud fraction, on the other hand, is found to be dominated by the smallest clouds observed. These clouds are additionally found to have a negative contribution to the mass flux, yet a positive contribution to the buoyancy flux. About 200 flight intersections of cumuli with horizontal sizes larger than 500 m are used to obtain average horizontal cross-section profiles of vertical velocity, liquid water content, liquid ...

Shallow cumulus convection: A validation of large‐eddy simulation against aircraft and Landsat observations

AbstractLarge‐eddy simulation (LES) results of shallow cumulus convection are directly evaluated against in‐cloud aircraft measurements, as made during the Small Cumulus Microphysics Study (SCMS). To this purpose an LES case is first constructed, based on available observations (during SCMS). Then the simulations are directly compared with the in‐cloud measurements by using conditionally sampled fields. An advantage of the SCMS dataset is the combination of a range of different surface measurements, in‐cloud measurements by an aircraft at many levels in the cloud layer, and the availability of high‐resolution Landsat images.The results show that given the correct initial and boundary conditions the LES concept is capable of realistically predicting the bulk thermodynamic properties of temperature, moisture and liquid‐water content of the cumulus‐cloud ensemble as observed in SCMS. Furthermore, the vertical component of the in‐cloud turbulent kinetic energy and the cloud size distribution in LES were in agreement with the observations. These results support the credibility of cloud statistics as produced by LES in general, and encourage its use as a tool for testing hypotheses and developing parametrizations of shallow cumulus cloud processes.Several hypotheses which make use of conditionally sampled fields were tested on the SCMS data. The magnitudes and the decrease with height of the bulk entrainment rate following from the SCMS data confirm the typical values of the order of magnitude of 10−3 m−1 as reported in several other recent LES studies. An alternative formulation of the lateral entrainment rate as a function of the liquid‐water content and the mean lapse rate agrees well with the original form based on the conserved variables. Applying an often‐used simplified equation for the cloud vertical velocity to the aircraft measurements results in a reasonably closed budget. Copyright © 2003 Royal Meteorological Society

Comparison between Standard and Modified Forward Scattering Spectrometer Probes during the Small Cumulus Microphysics Study

Abstract Microphysical measurements performed during the Small Cumulus Microphysics Study (SCMS) experiment are analyzed in order to examine the instrumental limitations of forward scattering spectrometer probes (FSSPs). Complementary information collected with a modified version of the instrument, the Fast-FSSP, are used to address crucial issues such as the size calibration of the spectrometers and the effects on the measured spectra (distortion and broadening), of beam inhomogeneities, of variations of the sampling section, and of the coincidence of particles. Their impact on the calculation of liquid water content is evaluated by the comparison with measurements performed with a hot-wire probe and a particle volume monitor. In addition to the statistical approach that aims at evaluating the typical uncertainty of the measurements, special attention is given to the identification of circumstances under which some of the instrumental limitations combined are likely to affect significantly the accuracy o...

Maritime/continental microphysical contrasts in stratus

More than 15 h of in situ cloud measurements in Atlantic and Pacific stratus showed cloud droplet concentrations to be correlated with cloud condensation nuclei (CCN) concentrations measured just below cloud. Predictions of droplet concentrations based on complete CCN spectra and measured updraft velocities were better correlated with measured cloud droplet concentrations. This was especially significant for near-adiabatic parcels because these results represent a good level of closure between CCN and cloud microphysics. Over the Atlantic there was a clear dichotomy between maritime and continental cloud microphysics that was related to CCN concentrations: smaller and more numerous droplets in continental clouds with higher CCN concentrations. However, there seemed to be undercounting of continental cloud droplets because many were smaller than the 2 μm diameter threshold of the droplet spectrometer. When averaged over all altitudes, the drizzle (diameter >50 μm) liquid water content was a factor of 2 greater in the maritime clouds. However, as in previously reported small cumuli, drizzle was also a function of cloud depth. Therefore when only thin clouds (depth less than 45 mb) were considered, the maritime clouds had an order of magnitude more drizzle than continental clouds. Drizzle liquid water content was also found to depend on the mean diameter of 2–50 μm diameter cloud droplets. In the thin clouds there was a definite 15 μm mean cloud droplet diameter threshold for the onset of drizzle that was often exceeded in the maritime clouds but almost never exceeded in the continental clouds. There were also higher concentrations of 20–50 μm diameter droplets in the maritime clouds that were commensurate with the maritime/continental drizzle (50–620 μm diameter) contrast. Furthermore, when cloud parcels with equal amounts of cloud droplet liquid water were compared, there was considerably more drizzle in the maritime clouds, especially for larger amounts of cloud water in the thin clouds. All of these maritime/continental contrasts found in thin stratus clouds are very similar to previously reported maritime/continental contrasts in small cumulus clouds. These results support the second indirect aerosol effect (pluvial inhibition) for stratus clouds.

Measurements of Ultragiant Aerosol Particles in the Atmosphere from the Small Cumulus Microphysics Study

Abstract Ultragiant aerosol particles (UGA) are potentially important for warm rain formation because of their ability to initiate coalescence immediately upon entering a cloud, so it is desirable to obtain local estimates during any field campaign that studies warm rain. Estimates of UGA in clear air from a one-dimensional optical array probe averaged over long time periods from the Small Cumulus Microphysics Study have been published in the literature, but further analysis and comparisons to other probes, presented here, show that the data on which these estimates were based were probably contaminated by noise. A possible explanation for the noise in the probe is given, as are new upper limits, based on few or no particles detected by a two-dimensional optical array probe.

Early Radar Echoes from Ultragiant Aerosol in a Cumulus Congestus: Modeling and Observations

Abstract The growth of ultragiant aerosol (UGA) in a Lagrangian framework within a simulated three-dimensional cloud is analyzed and compared with radar and aircraft observations of a cumulus congestus collected during the Small Cumulus Microphysics Study (SCMS). UGA are ingested into the simulated cloud and grow by continuous collection; the resulting radar reflectivity factor and raindrop concentrations are evaluated at 1-min intervals. The calculations produce a substantial echo (>30 dBZ) within a short time (18 min), containing few raindrops (0.3 L−1). The calculated radar echo is very sensitive to the amount of UGA ingested into the modeled cloud and its liquid water content. The modeled radar echo and raindrop concentrations are consistent with the observations in that the differences fall within the modeling and measurement limitations and uncertainties.

Cloud cover analysis with METEOSAT‐5 during INDOEX

During the Indian Ocean Experiment (INDOEX), METEOSAT‐5 positioned at 63°E provided observation of the visible and infrared radiance field over the Indian Ocean. A cloud classification process using the dynamic cluster method is applied to these data. For the 3 months of the experiment (January–March 1999), daily maps of the cloud cover type are built for 0730 and 0900 UTC. The occurrence frequency of clear sky, low‐ and high‐level cloud cover is examined. These frequencies are compared to the International Satellite Cloud Climatology Program (ISCCP) D1 data set for the period 1984 to 1994. The Indian Ocean region can be classified in three zones. In the north subtropics, clear sky and small cumulus occur at least 90% of the time. Near the coast of India, clear sky is as frequent as 80 to 100%. The Intertropical Convergence Zone, characterized by the occurrence frequency of high‐level clouds greater than 30%, spreads from Indonesia to North Madagascar. Near Indonesia, high‐level cloud cover occurs more than 55% of the time. In the south subtropics, low cloud cover is the most frequent. In the eastern part the occurrence frequency reaches 80%. This percentage decreases along the western side of the ocean where low clouds break up. Between the African coast and Madagascar, high‐level clouds are frequent. The mean spatial features found are in agreement with the ISCCP climatology, except for the eastern part of the south subtropics. A regional comparison shows the difficulty of making the analysis of interannual variations of cloud cover obtained from various cloud cover retrievals applied to different satellite data sets. This difficulty arises from the nonneglectable percentage of satellite pixels which can contain some very small low clouds.

Spectral Density of Cloud Liquid Water Content at High Frequencies

Abstract Aircraft measurements of liquid water content (LWC) made at sampling frequencies of 1 and 2 kHz with a particle volume monitor (PVM) probe from horizontal traverses in stratocumulus clouds during the Southern Ocean Cloud Experiment and cumulus clouds during the Small Cumulus Microphysics Study are described. The spectral density of the LWC measurements is calculated and compared to the −5/3 scaling law. The effect of PVM sampling noise is found to be small in most cases. Most measurements follow approximately the −5/3 law until cloud scales decrease below about 5 m in length. Below this length LWC variance can exceed that predicted by the −5/3 law. It is suggested that the enhanced LWC variance at small scales is related to entrainment of environmental air into the clouds, which changes primarily the droplet concentration.

Microphysical relationships in warm clouds

Relationships among cloud microphysical parameters from six aircraft projects carried out over different parts of the globe in different seasons are presented. Comparisons of project averages showed that projects with higher droplet (2–50 μm diameter) concentrations generally had smaller droplets and less drizzle (diameter >50 μm). Hence, when all projects were combined there were negative correlations between droplet concentrations and sizes. However, this was not the case for individual cloud penetrations where only 24% had negative droplet concentration–size correlations. For the stratocumulus clouds, these negative correlations were more likely to be found in clouds with greater droplet spectral broadening and drizzle. The small cumulus clouds in this study were unique in that 95% of the droplet concentration–size correlations were positive regardless of spectral broadening or drizzle. Thus, for the majority of clouds, droplets were smaller in parcels with lower droplet concentrations. Entrainment in these clouds was then consistent with homogeneous mixing that occurs locally and intermittently. The dearth of observations of larger droplets in diluted cloud parcels suggested that mixing of entrained air did not promote droplet growth.

Droplet Spectra Broadening in Cumulus Clouds. Part I: Broadening in Adiabatic Cores

Measurements of cloud droplet spectra performed with the Fast-Forward Scattering Spectrometer Probe during the Small Cumulus Microphysics Study (1995) are analyzed. Fifty cloud samples with narrow droplet spectra are selected. They are characterized by values of liquid water content slightly below the adiabatic value. Each observed spectrum is then compared to a narrow adiabatic spectrum predicted at the same level with the current theory of condensational growth in an adiabatic cloud cell, initialized with a reference spectrum measured right above the activation level, at cloud base. Broadening is characterized for each observed spectrum by the probability density function of condensational growth expressed as the Lagrangian integral of the ratio of supersaturation to vertical velocity, along the droplet trajectories. In particular it appears that the derived density functions show high probabilities of very low and very large values of condensational growth. The large values are related to a high relative density of big droplets in the measured spectra, higher than predicted by the adiabatic model. The contribution of the instrument to this feature is examined with a model of probe functioning. The simulations suggest that most of those big droplets are instrumental artifacts. The remaining broadening is parameterized by a linear relationship between the mean value and the standard deviation of the density function of condensational growth. This result will be used to examine the respective contributions to spectra broadening of microscale heterogeneities of the droplet concentration, in Part II, and of the mixing processes, in Part III of this series.

Initial Precipitation Formation in Warm Florida Cumulus

The microphysical processes that lead to the development of precipitation in small, warm cumulus are examined using data from the Small Cumulus Microphysics Study near Cape Canaveral, Florida. Aircraft measurements are used to determine the concentration and size distribution of giant and ultragiant nuclei in clear air as a function of relative humidity, altitude, wind speed, and wind direction. The clear-air particle distributions show that ultragiant particles (radii extending from 10 to 150 mm) exist from the surface to cloud base in concentrations that correspond to the concentrations of raindrops observed during drizzle to moderate rainfall events. A shift of the spectra toward larger size with increasing relative humidity was observed, suggesting that the spectra are composed of deliquesced particles growing by condensation. The small cumulus clouds are shown to contain cores where the observed liquid water content was nearly adiabatic. The observed evolution of the cloud droplet distribution within the near-adiabatic cores as a function of height showed an increase in the small droplet mode associated with condensation and an increase in the concentration of larger droplets associated with growth by accretion. Droplets with radii extending to nearly 100 mm were present just above cloud base. These measurements were consistent with the clear-air measurements and provided evidence that the ultragiant nuclei can immediately act as embryos for raindrop growth by accretion upon entering cloud base. Comparisons of reflectivity computed from the cumulus core composite droplet distributions with the radar-observed reflectivity data provided independent evidence that the composite spectra reasonably represented the evolving microstructure of the cores of small cumulus clouds as they grew vertically. The analyses provide strong evidence of an efficient process for the initial development of precipitation in small Florida cumuli. This process consists of raindrop embryo formation on ultragiant nuclei followed by growth by accretion as the newly formed drops proceed upward through the adiabatic cores of the cumulus clouds. These data support the conceptual model of raindrop formation in marine clouds first proposed by Woodcock a half century ago.

Observations of microphysics pertaining to the development of drizzle in warm, shallow cumulus clouds

AbstractThe evolution of drizzle in small cumulus clouds in a subtropical environment was studied. Observations were made using three instrumented aircraft, a ground‐based radar, and a 95 GHz airborne Doppler radar during the Small Cumulus Microphysics Study in east central Florida in summer 1995. Data from six clouds, from two days, are examined in this paper. Both sets of clouds were less than 2 km in depth at observation times; in‐cloud temperatures were > 10 °C.Droplet spectra observed in these clouds were generally bi‐modal. The sizes of the droplets within the large mode are consistent with growth through condensation. It remains unclear how droplets within the small mode formed, but their existence appears to be tied to entrapment and mixing.Drizzle drops (drops with diameters > 50 μm) were found at all levels in the clouds. The penetration‐averaged concentrations of drizzle drops were less than 15 L−1, except in the uppermost regions of clouds on one of the days where drizzle drop concentrations exceeded 100L−1. In general, the presence of drizzle drops in the upper regions of the clouds is consistent with models of droplet growth through condensation and stochastic collection. The existence of drizzle at low and mid‐levels may be due to collection initiated by ultra‐giant aerosols, or to the redistribution of drizzle drops from neighbouring or earlier cloud elements.

Evolution of small cumulus clouds in Florida: observations of pulsating growth

Observations have been made in six small cumulus clouds using instrumented aircraft, a ground-based radar, and a 95 GHz airborne Doppler radar. The clouds occurred on two days during the Small Cumulus Microphysics Study in east-central Florida, summer 1995. Cloud tops were below 3 km and in-cloud temperatures were warmer than 10°C. Maximum observed reflectivity factors were less than 0 dBZ. The evolution of the kinematics of the observed clouds was tracked using measurements from both radars. High-resolution cross-sections of reflectivity and vertical Doppler velocity from the airborne radar appear remarkably similar to fine-scale models of convection reported in the literature. In general, each cloud resembled a collection of individual bubbles ascending through the boundary layer. During the growth phase of a bubble, a positive correlation existed between vertical velocity and reflectivity. As bubbles penetrated further into the inversion, entrainment/detrainment led to a weakening or, in some cases, a reversal of this correlation. Growth of subsequent bubbles ascending through remnants of earlier bubbles were aided by an increase in the amount of moisture in the environment resulting from earlier detrainment of cloudy air, and thus were able to achieve higher altitudes than their predecessors.

Cloud cover observed simultaneously from FOLDER and METEOSAT

The POLDER instrument that was aboard the Japanese ADEOS platform between August 1996 and June 1997. is designed to the global observations of the polarisation and directionality of the sun-light reflected by the Earth-atmosphere system. The cloud detection from POLDER takes advantage of the original capabilities of the instrument (spectral polarisation and directionality). This cloud detection scheme uses 5 threshold tests based on pressure, reflectance, polarised reflectance and spectral variability. The results of the POLDER cloud detection scheme are compared to those of the LMD dynamical clustering method applied to visible and infrared METEOSAT data and local spatial variability of these two parameters. Special focus is given to the detection capabilities of the two kind of measurements for cloud situations such as small cumulus, thin cirrus and multilayered cloud cover. Results of this comparison would give some insight on the behaviour of the International Satellite Cloud Climatology Project (ISCCP) cloud detection scheme built mainly from visible and infrared measurements.

Validation of droplet spectra and liquid water content measurements

Abstract Various airborne devices have been recently developed to improve measurements of cloud microphysics. Data from the Small Cumulus Microphysics Study (SCMS) provide a unique data set for evaluation of the probe performances. In this paper we present an analysis of data collected during two SCMS flights where five instruments were tested: the Forward Scattering Spectrometer Probe (FSSP), the Fast-FSSP, the Cloud Drop Spectrometer (CDS), the Particulate Volume Monitor (PVM) and the CSIRO hot wire probe.

Comments on `A comparison of optical measurements of liquid water content and drop size distribution in adiabatic regions of Florida cumuli'

A recent paper by Lawson and Blyth (1998) [Lawson, R.P., Blyth, A.M., 1998. A comparison of optical measurements of liquid water content and drop size distribution in adiabatic regions of Florida cumuli. Atmos. Res. 47–48, 671–690.] is critiqued for its conclusions dealing with the measurement with several different aircraft microphysics probes of liquid water content (LWC) in `adiabatic cores' of isolated and relatively small cumulus clouds found during the 1995 Small Cumulus Microphysics Study (SCMS) experiment in Florida. The criteria used in that paper to identify the cores were found to be unreliable, leading to identifications of adiabatic cloud cores that contained sub-adiabatic LWC. The analysis in the present paper of 1000-Hz particulate volume monitor (PVM) LWC data for SCMS Cu leads to the following conclusions: (1) A new probe, the cloud droplet spectrometer (CDS), located on the same aircraft and operated at 1 Hz, overestimates maximum LWC in the Cu, and the claim of Lawson and Blyth (1998) that the 1-Hz CDS data show often precisely LWC equivalent to the predicted adiabatic LWC in those Cu is not supportable. (2) The high-frequency PVM data show adiabatic LWC in aircraft passes close to cloud base, and maximum LWC that is slightly sub-adiabatic in small cloud parcels in passes through the active updraft regions of taller Cu. (3) The taller Cu show internally sharp LWC gradients that are interpreted as nonuniform local mixtures of LWC and interfaces resulting from the entrainment process. (4) The understanding of the observed high-resolution LWC structure requires a better understanding of Cu dynamics associated with cloud growth, and entrainment, and mixing processes.

Characteristics of small tropical cumulus clouds and their impact on the environment

This study uses a number of data sets (Moderate Resolution Imaging Spectroradiometer (MODIS) Airborne Simulator data, space shuttle photography, Radiation Measurement System data, aircraft data, and shipboard soundings) to investigate the characteristics of small tropical cumulus clouds and their impact on their environment. The goal is to uncover useful information with application to radiative transfer simulation and satellite remote sensing. In fields of small cumulus clouds, size distributions are found to decrease in number with increasing diameter according to a double power law relation, often with a clear break diameter. Fractal dimensions corresponding to the horizontal area and perimeter of the clouds are greater for the larger clouds than for the smaller clouds, with the same break diameter as the size distributions, meaning that the larger clouds have more ragged perimeters. These two results suggest a characteristic horizontal length scale dividing larger and smaller boundary layer cumuli. Spatial distributions show a clear tendency toward clustering. Smaller cumuli appear to grow upward more quickly with increasing horizontal size than do larger cumuli. Albedo is found to increase with greater cloud fraction and higher solar zenith angle. Even sparse fields of small cumulus cause significant shortwave forcing at the ocean surface. Simulation suggests that small cumulus may introduce significant errors into sea surface temperature retrievals and that such clouds can be difficult to remove with operational cloud‐filtering schemes. Clouds smaller than about 1 km in diameter are not seen to precipitate.

A comparison of optical measurements of liquid water content and drop size distribution in adiabatic regions of Florida cumuli

Airborne measurements of liquid water content (LWC) and drop size distribution were made in adiabatic regions of small, growing cumulus clouds during the Small Cumulus Microphysics Study (SCMS). A new instrument, the cloud drop spectrometer (CDS), which measures LWC and also drop size from an ensemble of drops, was flown for the first time in the field. Measurements from other sensors, including a Particle Measuring Systems (PMS) forward scattering spectrometer probe (FSSP), the `fast' FSSP (FFSSP) developed by the Centre National De Recherches Meteorologiques (CNRM), and a Gerber Scientific airborne particulate volume monitor (PVM-100A), are compared with the CDS data collected in adiabatic and other regions. The CDS appeared to reliably measure very close to the predicted value of LWC in regions identified as being adiabatic. In addition, the drop size distribution measured by the CDS compared very well with the FSSP and FFSSP measurements, except where the 3–200 μm range of the CDS allowed it to measure larger drops than the nominal 3–45 μm range of the FSSP, and the 2.7 to 38.4 μm range of the FFSSP.

Organization and Representation of Boundary Layer Clouds

Abstract This study is intended to summarize and to simplify the complicated processes in boundary layer cloud regimes using a single parameter, Q1, the normalized saturation deficit. With the aid of large eddy simulation (LES) data from different boundary layer cloud regimes it is illustrated i) that the in-cloud buoyancy flux is maximized when the fractional cloudiness approaches zero; ii) that the ensemble average buoyancy flux possesses two maxima, one for the trade wind cumulus case and one for the stratocumulus case; and iii) that the preferred mode for boundary layer clouds is either small cumuli or high values of cloudiness, and that cloudiness transitions from one regime to the other are difficult to represent numerically as in the transition regime the cloud-water-related variables are very parameter sensitive. In addition, the importance of the contribution of the liquid water flux to the in-cloud and total buoyancy flux is outlined, and simple analytical and empirical methods are presented to ...

Diurnal Evolution of the Cloud-Topped Marine Boundary Layer. Part I: Nocturnal Stratocumulus Development

The structure and evolution of the extratropical marine atmosphere boundary layer (MABL) depend largely on the variability of stratus and stratocumulus clouds. Stratus clouds are generally associated with a well-mixed MABL, whereas daytime observations of stratocumulus-topped boundary layers generally indicate that the cloud and subcloud layers are decoupled. In the Atlantic Stratocumulus Transition Experiment, aircraft measurements show a surface-based mixed layer separated from the base of the stratocumulus by a layer that is stable to dry turbulent mixing. This layer forms due to shortwave heating of the stratocumulus clouds. Cumulus clouds often develop in this transition layer and they play a fundamental role in the redistribution of heat in the decoupled stratcumulus-capped boundary layer. They are, however, very sensitive to small changes in the heat and moisture in the boundary layer and are generally transient features that depend directly on the surface sensible and latent heat fluxes. The cumulus contribute a bimodal drop-size distribution to the stratocumulus layer skewed to the smallest sizes but may contain many large drops. Clouds increase at night in response to the combined effect of convection, which can transport drops to the top of the MABL, and outgoing longwave radiation, which cools the boundary layer. The relationship between the cumulus clouds and the latent heat flux is complex. Small cumulus may enhance the flux, but as more water vapor is redistributed vertically by an increase in convective activity the latent heat flux decreases. This study illustrates the need for boundary-layer models to properly handle the occurrence of intermittent cumulus to predict the diurnal evolution of the stratocumulus-capped MABL.