Cloud Droplet Spectral Dispersion Research Articles

Using Machine Learning to Predict Cloud Turbulent Entrainment‐Mixing Processes

AbstractDifferent turbulent entrainment‐mixing mechanisms between clouds and environment are essential to cloud‐related processes; however, accurate representation of entrainment‐mixing in weather/climate models still poses a challenge. This study exploits the use of machine learning (ML) to address this challenge. Four ML (Light Gradient Boosting Machine [LGB], eXtreme Gradient Boosting, Random Forest, and Support Vector Regression) are examined and compared. It is found that LGB performs best, and thus is selected to understand the impact of entrainment‐mixing on microphysics using simulation data from Explicit Mixing Parcel Model. Compared with traditional parameterizations, the trained LGB provides more accurate microphysical properties (number concentration and cloud droplet spectral dispersion). The partial dependences of predicted microphysics on features exhibit a strong alignment with physical mechanisms and expectations, as determined by the interpreting method, thus overcoming the limitations of the “black box” scheme. The underlying mechanisms are that the smaller number concentration and larger spectral dispersion correspond to more inhomogeneous entrainment‐mixing. Specifically, number concentration after entrainment‐mixing is positively correlated with adiabatic number concentration and liquid water content affected by entrainment‐mixing, and inversely correlated with adiabatic volume mean radius. Spectral dispersion after entrainment‐mixing is negatively correlated with liquid water content affected by entrainment‐mixing, turbulent dissipation rate and relative humidity of entrained air. Sensitivity analysis further suggests that number concentration is mainly determined by cloud microphysical properties whereas spectral dispersion is influenced by both cloud microphysical properties and environmental variables. The results indicate that the LGB scheme has the potential to enhance the representation of entrainment‐mixing in weather/climate models.

Vertical Distributions of Aerosol and Cloud Microphysical Properties and the Aerosol Impact on a Continental Cumulus Cloud Based on Aircraft Measurements From the Loess Plateau of China

Based on aircraft measurements of aerosols and continental cumulus clouds made over the Loess Plateau of China (Xinzhou, Shanxi Province) on 30 July 2020, this study focuses on the vertical profiles of microphysical properties of aerosols and cumulus clouds, and use them to study aerosol-cloud interactions. During the study period, the boundary layer was stable with a height ∼1,500 m above sea level. Aerosols in the boundary layer mainly came from local emissions, while aerosols above this layer were mostly dust aerosols transported over long distances. Vertical profiles of aerosols and cloud condensation nuclei were obtained, and aerosol activation ratios at different supersaturation (SS) levels ranged from 0.16 to 0.32 at 0.2% SS and 0.70 to 0.85 at 0.8% SS. A thick cumulus cloud in the development stage was observed from the bottom to the top with the horizontal dimension of 10 km by 7 km, the cloud-base height of 2,450 m (15.8°C), and the cloud-top height of 5,400 m (−3°C). The maximum updraft velocity near the cloud top was 13.45 m s−1, and the maximum downdraft velocity occuring in the upper-middle part of the cloud was 4.44 ms−1. The temperature inside the cloud was higher than the outside, with their difference being positively correlated with the cloud water content. The temperature lapse rate inside the cloud was about −6.5°C km−1. The liquid water content and droplet effective radius (Re) increased with increasing height. The cloud droplet number concentration (Nc) increased first then decreased, peaking in the middle and lower part of the cloud, the average values of Nc and Re were 767.9 cm−3 and 5.17 μm, respectively. The cloud droplet spectrum had a multi-peak distribution, with the first appearing at ∼4.5 μm. SS in the cloud first increased then decreased with height. The maximum SS is ∼0.7% appearing at ∼3,800 m. The conversion rate of intra-cloud aerosols to cloud droplets was between 0.2 and 0.54, with the ratio increasing gradually with increasing height. The cloud droplet spectral dispersion and Nc were positively correlated. The aerosol indirect effect (AIE) was estimated to be 0.245 and 0.16, based on Nc and Re, respectively. The cloud droplet dispersion mainly attenuated the AIE, up to ∼34.7%.

Radiative forcing of the aerosol-cloud interaction in seriously polluted East China and East China Sea

The radiative forcing caused by aerosol-cloud interaction (ACI) is one of the most critical factors that lead to climate research uncertainty. In East China and the adjacent sea areas, the severe air pollution makes the ACI effect stronger than in other regions, but few observational studies focus on the effect of different aerosol components. This study estimated the shortwave radiation effect at the top of the atmosphere (TOA) caused by the interaction between the increased aerosol and the warm liquid cloud in East China and the East China Sea by applying the multiple linear regression into two ACI effect calculating methods proposed by Quaas et al. (2008; Method 1) and Chen et al. (2014; Method 2). Four aerosol components, black carbon (BC), dust (DU), organic carbon (OC), and sulfate (SU), are included in this research. The total ACI radiation effect of the four aerosol components is −12.08 ± 4.63 W m−2 in Method 1 and − 9.25 ± 10.44 W m−2 in Method 2, respectively, indicating a cooling effect to the planet. The divergence in the two methods is probably because the higher aerosol loading increases the cloud property retrieval deviation, and the cloud droplet spectral dispersion is neglected in Method 1. The effect of different aerosol components has significant diversity, with OC and SU have a cooling effect, DU has a heating effect, and BC shows an obvious geographical distinction. Further analysis suggests that under a similar aerosol optical depth (AOD), the areas with high relative humidity (RH) and low-tropospheric stability (LTS) have a more significant cooling effect due to the suppression of droplet evaporation and entrainment. Regions with higher average cloud top height also have a stronger cooling effect, especially for BC and SU. The possible reason is that high cloud altitude reduces the aerosol concentration within the cloud, further alleviates the heating effect of absorbing aerosols, and reduces the cloud droplet evaporation caused by the over much cloud condensation nuclei. This research contributes to a better understanding of the aerosol-cloud radiative effect and its mechanism in East China and the East China Sea.

Role of droplet size classes on the cloud droplet spectral dispersion as observed over the Western Ghats

The aerosol indirect effects (AIE) on cloud microphysical properties have been studied using ground-based in-situ measurements. Aerosol and cloud microphysical properties were collected from a high altitude site in the Western Ghats region of India, where clouds pass over the surface during South West monsoon. The AIE was estimated using cloud droplet number concentration (AIEn) and cloud droplet effective radius (AIEs) at different fixed liquid water content (LWC) bins. The AIE varied in the range 0.01–0.13 for LWC between 0.04 g m−3 and 0.26 g m−3. The maximum AIEn and AIEs (0.125 and 0.119) were observed at the lower LWC bin (0.06–0.07 g m−3). An overestimation in AIEn is due to the dispersion effect. The offset was reduced and AIEn became close to AIEs after dispersion correction. Hence for the correct estimation of AIE using cloud droplet number concentration (Nc), dispersion correction should be considered. At the same time, the AIE estimate obtained using Reff does not need correction as the dispersion effect is implicitly included. Relative dispersion can be viewed not only as the combined effects of σ and Rm, but also as the combined effects of cloud droplets of different sizes. Further, for the first time in India, the relative contribution by the smaller and medium size droplets to the total dispersion of cloud droplet size distribution (CDSD) at lower and higher LWC bins were studied. In lower LWC, the smaller size droplets are contributing maximum (71%) to the total Nc, whereas the medium size droplets are contributing maximum (61%) to the total dispersion. At higher LWC, the contribution to total Nc by the smaller size droplets was reduced by 20%, however contribution by the medium size droplets increased by 17%, compared to the lower LWC case. Hence at higher LWC, both smaller (42%) and medium (55%) size droplets have a significant role in the total dispersion of CDSD.

Improved Aerosol Processes and Effective Radiative Forcing in HadGEM3 and UKESM1

AbstractAerosol processes and, in particular, aerosol‐cloud interactions cut across the traditional physical‐Earth system boundary of coupled Earth system models and remain one of the key uncertainties in estimating anthropogenic radiative forcing of climate. Here we calculate the historical aerosol effective radiative forcing (ERF) in the HadGEM3‐GA7 climate model in order to assess the suitability of this model for inclusion in the UK Earth system model, UKESM1. The aerosol ERF, calculated for the year 2000 relative to 1850, is large and negative in the standard GA7 model leading to an unrealistic negative total anthropogenic forcing over the twentieth century. We show how underlying assumptions and missing processes in both the physical model and aerosol parameterizations lead to this large aerosol ERF. A number of model improvements are investigated to assess their impact on the aerosol ERF. These include an improved representation of cloud droplet spectral dispersion, updates to the aerosol activation scheme, and black carbon optical properties. One of the largest contributors to the aerosol forcing uncertainty is insufficient knowledge of the preindustrial aerosol climate. We evaluate the contribution of uncertainties in the natural marine emissions of dimethyl sulfide and organic aerosol to the ERF. The combination of model improvements derived from these studies weakens the aerosol ERF by up to 50% of the original value and leads to a total anthropogenic historical forcing more in line with assessed values.

Analytical studies of the cloud droplet spectral dispersion influence on the first indirect aerosol effect

Atmospheric aerosols (acting as cloud condensation nuclei) can enhance the cloud droplet number concentration and reduce the cloud droplet size, and in turn affect the cloud optical depth, as well as the cloud albedo, and thereby exert a radiative influence on climate (the first indirect aerosol effect). In this paper, based on various relationships between cloud droplet spectral dispersion (ɛ) and cloud droplet number concentration (Nc), we analytically derive the corresponding expressions of the cloud radiative forcing induced by changes in the cloud droplet number concentration. Further quantitative evaluation indicates that the cloud radiative forcing induced by aerosols for the different ɛ−Nc relationships varies from −29.1% to 25.2%, compared to the case without considering spectral dispersion (ɛ = 0). Our results suggest that an accurate description of ɛ − Nc relationships helps to reduce the uncertainty of the first indirect aerosol effect and advances our scientific understanding of aerosol-cloud-radiation interactions.

Aerosol–climate interactions in the Norwegian Earth System Model – NorESM1-M

Abstract. The objective of this study is to document and evaluate recent changes and updates to the module for aerosols and aerosol–cloud–radiation interactions in the atmospheric module CAM4-Oslo of the core version of the Norwegian Earth System Model (NorESM), NorESM1-M. Particular attention is paid to the role of natural organics, sea salt, and mineral dust in determining the gross aerosol properties as well as the anthropogenic contribution to these properties and the associated direct and indirect radiative forcing. The aerosol module is extended from earlier versions that have been published, and includes life-cycling of sea salt, mineral dust, particulate sulphate, black carbon, and primary and secondary organics. The impacts of most of the numerous changes since previous versions are thoroughly explored by sensitivity experiments. The most important changes are: modified prognostic sea salt emissions; updated treatment of precipitation scavenging and gravitational settling; inclusion of biogenic primary organics and methane sulphonic acid (MSA) from oceans; almost doubled production of land-based biogenic secondary organic aerosols (SOA); and increased ratio of organic matter to organic carbon (OM/OC) for biomass burning aerosols from 1.4 to 2.6. Compared with in situ measurements and remotely sensed data, the new treatments of sea salt and dust aerosols give smaller biases in near-surface mass concentrations and aerosol optical depth than in the earlier model version. The model biases for mass concentrations are approximately unchanged for sulphate and BC. The enhanced levels of modeled OM yield improved overall statistics, even though OM is still underestimated in Europe and overestimated in North America. The global anthropogenic aerosol direct radiative forcing (DRF) at the top of the atmosphere has changed from a small positive value to −0.08 W m−2 in CAM4-Oslo. The sensitivity tests suggest that this change can be attributed to the new treatment of biomass burning aerosols and gravitational settling. Although it has not been a goal in this study, the new DRF estimate is closer both to the median model estimate from the AeroCom intercomparison and the best estimate in IPCC AR4. Estimated DRF at the ground surface has increased by ca. 60%, to −1.89 W m−2. We show that this can be explained by new emission data and omitted mixing of constituents between updrafts and downdrafts in convective clouds. The increased abundance of natural OM and the introduction of a cloud droplet spectral dispersion formulation are the most important contributions to a considerably decreased estimate of the indirect radiative forcing (IndRF). The IndRF is also found to be sensitive to assumptions about the coating of insoluble aerosols by sulphate and OM. The IndRF of −1.2 W m−2, which is closer to the IPCC AR4 estimates than the previous estimate of −1.9 W m−2, has thus been obtained without imposing unrealistic artificial lower bounds on cloud droplet number concentrations.

Numerical simulation of clouds and precipitation depending on different relationships between aerosol and cloud droplet spectral dispersion

ABSTRACTThe aerosol effects on clouds and precipitation in deep convective cloud systems are investigated using the Weather Research and Forecast (WRF) model with the Morrison two-moment bulk microphysics scheme. Considering positive or negative relationships between the cloud droplet number concentration (Nc) and spectral dispersion (ɛ), a suite of sensitivity experiments are performed using an initial sounding data of the deep convective cloud system on 31 March 2005 in Beijing under either a maritime (‘clean’) or continental (‘polluted’) background. Numerical experiments in this study indicate that the sign of the surface precipitation response induced by aerosols is dependent on the ɛ−Nc relationships, which can influence the autoconversion processes from cloud droplets to rain drops. When the spectral dispersion ɛ is an increasing function of Nc, the domain-average cumulative precipitation increases with aerosol concentrations from maritime to continental background. That may be because the existence of large-sized rain drops can increase precipitation at high aerosol concentration. However, the surface precipitation is reduced with increasing concentrations of aerosol particles when ɛ is a decreasing function of Nc. For the ɛ−Nc negative relationships, smaller spectral dispersion suppresses the autoconversion processes, reduces the rain water content and eventually decreases the surface precipitation under polluted conditions. Although differences in the surface precipitation between polluted and clean backgrounds are small for all the ɛ−Nc relationships, additional simulations show that our findings are robust to small perturbations in the initial thermal conditions.

Parameterization of cloud droplet size distributions: Comparison with parcel models and observations

This work examines the efficacy of various physically based approaches derived from one‐dimensional adiabatic parcel model frameworks (a numerical model and a simplified parameterization) to parameterize the cloud droplet distribution characteristics for computing cloud effective radius and autoconversion rate in regional/global atmospheric models. Evaluations are carried out for integrations with single (average) and distributions of updraft velocity, assuming that (1) conditions at smax are reflective of the cloud column or (2) cloud properties vary vertically, in agreement with one‐dimensional parcel theory. The predicted droplet distributions are then compared against in situ cloud droplet observations obtained during the CRYSTAL‐FACE and CSTRIPE missions. Good agreement of droplet relative dispersion between parcel model frameworks indicates that the parameterized parcel model essentially captures one‐dimensional dynamics; the predicted distributions are overly narrow, with relative dispersion being a factor of 2 lower than observations. However, if conditions at cloud maximum supersaturation are used to predict relative dispersion and applied throughout the cloud column, better agreement is seen with observations, especially if integrations are carried out over the distribution of updraft velocity. When considering the efficiency of the method, calculating cloud droplet spectral dispersion at smax is preferred for linking aerosol with droplet distributions in large‐scale models.

Cloud droplet spectral dispersion and the indirect aerosol effect: Comparison of two treatments in a GCM

Two parameterizations of cloud droplet spectral dispersion and their impact on the indirect aerosol effect are compared in a global climate model. The earlier scheme specifies β, the ratio of droplet effective radius to volume‐mean radius, in terms of N, the cloud droplet number concentration. The new scheme specifies β in terms of mean droplet mass (L/N), where L is liquid water content, to account for the effect of variations in L. For low to moderate N, the new scheme gives a stronger increase of β with increasing N than the old scheme. In a present‐climate simulation, the new scheme shows a stronger gradient between remote regions (small β) and polluted/continental regions (large β). The new scheme also offsets the first indirect aerosol forcing (ΔF) more strongly: ΔF = −0.65 W m−2 with constant β, −0.43 W m−2 with the old β, and −0.38 W m−2 with the new β.

Aerosol‐cloud relationships in continental shallow cumulus

Aerosol‐cloud relationships are derived from 14 warm continental cumuli cases sampled during the 2006 Gulf of Mexico Atmospheric Composition and Climate Study (GoMACCS) by the Center for Interdisciplinary Remotely‐Piloted Aircraft Studies (CIRPAS) Twin Otter aircraft. Cloud droplet number concentration is clearly proportional to the subcloud accumulation mode aerosol number concentration. An inverse correlation between cloud top effective radius and subcloud aerosol number concentration is observed when cloud depth variations are accounted for. There are no discernable aerosol effects on cloud droplet spectral dispersion; the averaged spectral relative dispersion is 0.30 ± 0.04. Aerosol‐cloud relationships are also identified from comparison of two isolated cloud cases that occurred under different degrees of anthropogenic influence. Cloud liquid water content, cloud droplet number concentration, and cloud top effective radius exhibit subadiabaticity resulting from entrainment mixing processes. The degree of LWC subadiabaticity is found to increase with cloud depth. Impacts of subadiabaticity on cloud optical properties are assessed. It is estimated that owing to entrainment mixing, cloud LWP, effective radius, and cloud albedo are decreased by 50–85%, 5–35%, and 2–26%, respectively, relative to adiabatic values of a plane‐parallel cloud. The impact of subadiabaticity on cloud albedo is largest for shallow clouds. Results suggest that the effect of entrainment mixing must be accounted for when evaluating the aerosol indirect effect.

Aircraft measurements of cloud droplet spectral dispersion and implications for indirect aerosol radiative forcing

Using a large amount of aircraft measurements of cloud droplet size distributions, the relationship between cloud spectral relative dispersion (ɛ) and cloud droplet number concentration (Nc) is studied. The results indicate that the value of ɛ varies between 0.2 to 0.8 when the cloud droplet number concentration is low (about 50 cm−3), and converges toward a narrow range of 0.4 to 0.5 when the cloud number concentration is higher. Because the distribution of the cloud droplet size is an important parameter in estimating the first indirect radiative effect of aerosols on the climate system, the uncertainty in the corresponding radiative forcing can be reduced by 10–40% (depending on cloud droplet number density) under high aerosol loading. This finding is important for improving climate change projections, especially for the regions where aerosol loading is high and continues to increase.

Uncertainty analysis for estimates of the first indirect aerosol effect

Abstract. The IPCC has stressed the importance of producing unbiased estimates of the uncertainty in indirect aerosol forcing, in order to give policy makers as well as research managers an understanding of the most important aspects of climate change that require refinement. In this study, we use 3-D meteorological fields together with a radiative transfer model to examine the spatially-resolved uncertainty in estimates of the first indirect aerosol forcing. The global mean forcing calculated in the reference case is -1.30 Wm-2. Uncertainties in the indirect forcing associated with aerosol and aerosol precursor emissions, aerosol mass concentrations from different chemical transport models, aerosol size distributions, the cloud droplet parameterization, the representation of the in-cloud updraft velocity, the relationship between effective radius and volume mean radius, cloud liquid water content, cloud fraction, and the change in the cloud drop single scattering albedo due to the presence of black carbon are calculated. The aerosol burden calculated by chemical transport models and the cloud fraction are found to be the most important sources of uncertainty. Variations in these parameters cause an underestimation or overestimation of the indirect forcing compared to the base case by more than 0.6 Wm-2. Uncertainties associated with aerosol and aerosol precursor emissions, uncertainties in the representation of the aerosol size distribution (including the representation of the pre-industrial size distribution), and uncertainties in the representation of cloud droplet spectral dispersion effect cause uncertainties in the global mean forcing of 0.2~0.6 Wm-2. There are significant regional differences in the uncertainty associated with the first indirect forcing with the largest uncertainties in industrial regions (North America, Europe, East Asia) followed by those in the major biomass burning regions.

Sensitivity of the First Indirect Aerosol Effect to an Increase of Cloud Droplet Spectral Dispersion with Droplet Number Concentration

Observations show that an increase in anthropogenic aerosols leads to concurrent increases in the cloud droplet concentration and the relative dispersion of the cloud droplet spectrum, other factors being equal. It has been suggested that the increase in effective radius resulting from increased relative dispersion may substantially negate the indirect aerosol effect, but this is usually not parameterized in global climate models (GCMs). Empirical parameterizations, designed to represent the average of this effect, as well as its lower and upper bounds, are tested in the CSIRO GCM. Compared to a control simulation, in which the relative dispersion of the cloud droplet spectrum is prescribed separately over land and ocean, inclusion of this effect reduces the magnitude of the first indirect aerosol effect by between 12% and 35%.