Abstract

The dynamics and predictability of the rapid intensification (RI) of Hurricane Harvey (2017) were examined using convection-permitting initialization, analysis, and prediction from a cycling ensemble Kalman filter (EnKF) that assimilated all-sky infrared radiances from the Advanced Baseline Imager on GOES-16. The EnKF analyses were able to evolve the various scales of the radiance fields associated with Harvey close to those observed, including those associated with scattered individual convective cells before the onset of rapid intensification (RI) and the organized vortex-scale convective system during and after RI. This was true for more than 3 days of a continuous assimilation cycling. Deterministic forecasts initialized from the EnKF analyses captured the rapidly deepening intensity of Harvey more than 24 h prior to its onset. To explore the predictability of Harvey’s intensity during RI, ensemble probabilistic forecasts and sensitivity analyses were conducted. It was found that significant ensemble spread growth was induced by initial perturbations individually in either the wind or moisture fields. The nonlinear interactions between wind and moisture perturbations further limited the predictability of the intensification process of Harvey by increasing the uncertainty in the simulated wind and moisture distributions and modifying the convective activity and its feedback on vortex flow. This study highlights both the importance of better initializing the dynamic and moisture state variables simultaneously and the potential contribution of satellite all-sky radiance assimilation on constraining them and their associated convective activity that impacts RI of tropical cyclones.

Abstract

Radiosonde, in situ, and surface-based remote sensor data from the Atlantic Stratocumulus Transition Experiment are used to study the diurnal cycle of cloud and thermodynamic structure. A cloud layer and decoupled subcloud layer separated by a stable transition layer, often observed in the vicinity of cumulus cloud base, characterizes the thermodynamic structure during the study period. The mode of cloud structure is cumulus with bases below decoupled stratus. Data are presented that support the hypothesis that diurnal variations in cumulus development are modulated by the stability in the transition layer.

The frequency of cumulus convection decreases during the afternoon, but mesoscale regions of vigorous cumulus with cloud tops overshooting the base of the trade inversion and increased surface drizzle rates are present during the late afternoon and early evening, when the transition layer is the most stable. It is postulated that mesoscale organization may be required to accumulate enough water vapor in the subcloud layer to produce the convective available potential energy needed for developing cumulus to overcome transition layer stability. The mesoscale regions appear to fit the description of cyclic cumulus convection proposed in a previous study, and this theory is expanded to account for diurnal variations in the stability of the transition layer. The occurrence of these mesoscale clusters of vigorous convection makes it difficult to determine if the latent heat flux in the cloud layer has actually decreased in the late afternoon and early evening, when the transition layer is the most stable.

Liquid water structure was examined and no pronounced diurnal signal was found. Results showed that clouds thicker than approximately 450 m tended to have subadiabatic integrated liquid water contents, presumably due to evaporation of drizzle in the subcloud layer, removal of liquid water at the surface, and the evaporation of cloud water at cloud top. A significant fraction of clouds less than 450 m thick produced liquid water contents that were greater than adiabatic, and there may be a physical mechanism that could produce such values in this cloud system (i.e., lateral detrainment of cloud water from convective elements mixing with existing liquid water in decoupled stratus or with liquid water detrained by nearby convective elements). Unfortunately, instrument limitations may have also produced these greater-than-adiabatic values and the extent of instrument artifacts in these results is unclear.

Abstract

When stratiform-cloud-integrated radiative flux divergence (heating) is dependent on liquid water path (LWP) and droplet concentration N_d , feedbacks between cloud dynamics and this heating can exist. These feedbacks can be particularly strong for low LWP stratiform clouds, in which cloud-integrated longwave cooling is sensitive to LWP and N_d . Large-eddy simulations reveal that these radiative–dynamical feedbacks can substantially modify low LWP stratiform cloud evolution when N_d is perturbed.

At night, more rapid initial evaporation of the cloud layer occurs when N_d is high, leading to more cloud breaks and lower LWP values that both result in less total cloud longwave cooling. Weakened circulations result from this reduced longwave cooling and entrainment drying is able to counteract cloud growth. When N_d is low, the cloud layer is better maintained because cloud longwave cooling is still relatively strong.

During the day, the addition of shortwave warming leads to reduced LWP for all values of N_d and, consequently, further reduced longwave cooling and weakened circulations. For high N_d , these reductions are such that the cloud layer cannot be maintained. For lower N_d , the reductions are smaller and the cloud layer thins but does not dissipate.

These results suggest that low LWP cloud layers are more tenuous when N_d is high and are more prone to dissipating during the day. Comparison with other studies suggests the modeled low LWP cloud response may be sensitive to the initial thermodynamic profile and model configuration.

Abstract

A database of stratus cloud droplet (diameter <50 μm) size distribution parameters, derived from in situ data reported in the existing literature, was created, facilitating intercomparison among datasets and quantifying typical values and their variability. From the datasets, which were divided into marine and continental groups, several parameters are presented, including the total number concentration, effective diameter, mean diameter, standard deviation of the droplet diameters about the mean diameter, and liquid water content, as well as the parameters of modified gamma and lognormal distributions. In light of these results, the appropriateness of common assumptions used in remote sensing of cloud droplet size distributions is discussed. For example, vertical profiles of mean diameter, effective diameter, and liquid water content agreed qualitatively with expectations based on the current paradigm of cloud formation. Whereas parcel theory predicts that the standard deviation about the mean diameter should decrease with height, the results illustrated that the standard deviation generally increases with height. A feature common to all marine clouds was their approximately constant total number concentration profiles; however, the total number concentration profiles of continental clouds were highly variable. Without cloud condensation nuclei spectra, classification of clouds into marine and continental groups is based on indirect methods. After reclassification of four sets of measurements in the database, there was a fairly clear dichotomy between marine and continental clouds, but a great deal of variability within each classification.

The relevant applications of this study lie in radiative transfer and climate issues, rather than in cloud formation and dynamics. Techniques that invert remotely sensed measurements into cloud droplet size distributions frequently rely on a priori assumptions, such as constant number concentration profiles and constant spectral width. The results of this paper provide a basis for evaluating the sensitivity of these techniques. In particular, there were large enough differences in observed droplet spectral widths to significantly affect remotely sensed determinations of cloud microphysics.

Abstract

Testing the often-made assumption that ice particle aggregates (snowflakes) are well represented by oblate spheroids, ellipsoid fits are applied to aggregate images. An algorithm to retrieve both the ellipsoidal parameters and the orientations of the fitted ellipsoids is applied to Multi-Angle Snowflake Camera measurements of ice particle aggregates observed in Alaska. The resulting ellipsoids have shapes closer to prolate spheroids than the oft-assumed oblate spheroids. A robust linear relationship exists between the two characteristic aspect ratios of the ellipsoids. The most probable orientation of the maximum dimension of the retrieved ellipsoids is not in the horizontal plane, and the rotational angles of the maximum dimensions in the horizontal plane are not uniform, but instead display some correlation with the wind direction at the times of the measurements. The retrieval results can be used to improve the representation of aggregates in microphysics and/or electromagnetic radiation scattering models applicable to radar and satellite measurements.

Abstract

An 8-mm wavelength radar, 3-mm wavelength radar, and 10.6-µm wavelength lidar operated side by side in vertically pointing mode during the First ISCCP Regional Experiment (FIRE II). This data collection mode yielded detailed information on distribution of cloud and cloud boundaries as a function of altitude. Statistics on the location of cloud boundaries during the FIRE II experiment indicate that cloud bases tended to form at two discrete levels centered around 2.5 and 7.5 km, cirrus cloud tops formed most frequently at 9.5 km and cloud thickness were usually 2 km or less. The atmosphere had the highest incidence of cloudiness at 8.5 km AGL, with a secondary maximum at an altitude of 3.5 km AGL. The incidence of cloudiness fell off rapidly between 8 and 11 km; there was also a distinct minimum in cloudiness at 2 km AGL. The diurnal variation of upper-level cloud base and top heights was about 1.0 km AGL, with the highest bases and tops occurring at 0500 UTC and the lowest bases and tops occurring at 1500 UTC. Co-occurring cloud layers (two or more simultaneous layers) were common, with the condition of a single cloud layer accounting for only 40% of the observation period.

Abstract

A cloud particle size retrieval algorithm that uses radar reflectivity factor and Doppler velocity obtained by a 35-GHz Doppler radar and liquid water path estimated from microwave radiometer radiance measurements is developed to infer the size distribution of stratus cloud particles. Assuming a constant, but unknown, number concentration with height, the algorithm retrieves the number concentration and vertical profiles of liquid water content and particle effective radius. A novel aspect of the retrieval is that it depends upon an estimated particle median radius vertical profile that is derived from a statistical model that relates size to variations in particle vertical velocity; the model posits that the median particle radius is proportional to the fourth root of the particle velocity variance if the radii of particles in a parcel of zero vertical velocity is neglected. The performance of the retrieval is evaluated using data from two stratus case study days 1.5 and 8.0 h in temporal extent. Aircraft in situ microphysical measurements were available on one of the two days and the retrieved number concentrations and effective radii are consistent with them. The retrieved liquid water content and effective radius increase with height for both stratus cases, which agree with earlier studies. Error analyses suggest that the error in the liquid water content vanishes and the magnitudes of the fractional error in the effective radius and shortwave extinction coefficient computed from retrieved cloud particle size distributions are half of the magnitudes of the fractional error in the estimated cloud particle median radius if the fractional error in the median radius is constant with height.

Abstract

The full-spectrum correlated k-distribution (FSCK) method, originally developed for applications in combustion systems, is adapted for use in shortwave atmospheric radiative transfer. By weighting k distributions by the solar source function, the FSCK method eliminates the requirement that the Planck function be constant over a spectral interval. As a consequence, integration may be carried out across the full spectrum as long as the assumption of correlation from one atmospheric level to the next remains valid. Problems with the lack of correlation across the full spectrum are removed by partitioning the spectrum at a wavelength of 0.68 μm into two bands. The resulting two-band approach in the FSCK formalism produces broadband rms clear-sky flux and heating rate errors less than 1% and 6%, respectively, relative to monochromatic calculations and requires only 15 quadrature points per layer, which represents a 60%–90% reduction in computation time relative to other models currently in use.

An evaluation of fluxes calculated by the FSCK method in cases with idealized clouds demonstrates that gray cloud scattering in two spectral bands is sufficient to reproduce line-by-line generated fluxes. Two different approaches for modeling absorption by cloud drops were also examined. Explicitly including nongray cloud absorption in solar source function-weighted k distributions results in realistic in-cloud heating rates, although in-cloud heating rates were underpredicted by approximately 8%–12% as compared to line-by-line results. A gray cloud absorption parameter chosen to fit line-by-line results optimally for one cloud or atmospheric profile but applied to different cloud combinations or profiles, also closely approximated line-by-line heating rates.

Abstract

The Tropically Excited Arctic Warming (TEAM) mechanism ascribes warming of the Arctic surface to tropical convection, which excites poleward-propagating Rossby wave trains that transport water vapor and heat into the Arctic. A crucial component of the TEAM mechanism is the increase in downward infrared radiation (IR) that precedes the Arctic warming. Previous studies have examined the downward IR associated with the TEAM mechanism using reanalysis data. To corroborate previous findings, this study examines the linkage between tropical convection, Rossby wave trains, and downward IR with Baseline Surface Radiation Network (BSRN) downward IR station data. The physical processes that drive changes in the downward IR are also investigated by regressing 300-hPa geopotential height, outgoing longwave radiation, water vapor flux, ERA-Interim downward IR, and other key variables against the BSRN downward IR at Barrow, Alaska, and Ny-Ålesund, Spitsbergen.

Both the Barrow and the Ny-Ålesund station downward IR anomalies are preceded by anomalous tropical convection and poleward-propagating Rossby wave trains. The wave train associated with Barrow resembles the Pacific–North America teleconnection pattern, and that for Ny-Ålesund corresponds to a northwestern Atlantic wave train. It is found that both wave trains promote warm and moist advection from the midlatitudes into the Arctic. The resulting water vapor flux convergence, multiplied by the latent heat of vaporization, closely resembles the regressed ERA-Interim downward IR. These results suggest that the combination of warm advection, latent heat release, and increased cloudiness all contribute toward an increase in downward IR.

Abstract

Cumulus clouds can become tilted or elongated in the presence of wind shear. Nevertheless, most studies of the interaction of cumulus clouds and radiation have assumed these clouds to be isotropic. This paper describes an investigation of the effect of fair-weather cumulus cloud field anisotropy on domain-averaged solar fluxes and atmospheric heating rate profiles. A stochastic field generation algorithm was used to produce 20 three-dimensional liquid water content fields based on the statistical properties of cloud scenes from a large eddy simulation. Progressively greater degrees of x–z plane tilting and horizontal stretching were imposed on each of these scenes, so that an ensemble of scenes was produced for each level of distortion. The resulting scenes were used as input to a three-dimensional Monte Carlo radiative transfer model. Domain-averaged transmission, reflection, and absorption of broadband solar radiation were computed for each scene along with the average heating rate profile. Both tilt and horizontal stretching were found to significantly affect calculated fluxes, with the amount and sign of flux differences depending strongly on sun position relative to cloud distortion geometry. The mechanisms by which anisotropy interacts with solar fluxes were investigated by comparisons to independent pixel approximation and tilted independent pixel approximation computations for the same scenes. Cumulus anisotropy was found to most strongly impact solar radiative transfer by changing the effective cloud fraction (i.e., the cloud fraction with respect to the solar beam direction).