B. G. Arends

The Kleiner Feldberg Cloud Experiment 1990. An overview

An overview is given of the Kleiner Feldberg cloud experiment performed from 27 October until 13 November 1990. The experiment was carried out by numerous European research groups as a joint effort within the EUROTRAC-GCE project in order to study the interaction of cloud droplets with atmospheric trace constituents. After a description of the observational site and the measurements which were performed, the general cloud formation mechanisms encountered during the experiment are discussed. Special attention is given here to the process of moist adiabatic lifting. Furthermore, an overview is given regarding the pollutant levels in the gas phase, the particulate and the liquid phase, and some major findings are presented with respect to the experimental objectives. Finally, a first comparison attempts to put the results obtained during this campaign into perspective with the previous GCE field campaign in the Po Valley.

Night-time formation and occurrence of new particles associated with orographic clouds

Abstract The formation and occurrence of new ultrafine aerosol particles were studied in association with an orographic cloud during a field experiment at Great Dun Fell (GDF), Northern England. Three size spectrometers to measure submicrometer aerosol particles were located upwind, on top, and downwind of GDF Summit to investigate changes in the aerosol size distribution. During two night-time cloud periods, ultrafine particles were observed downwind of the hill while no particles were detected upwind of the hill. During one cloud event, there was some evidence of entrainment. In this case, the occurrence of ultrafine particles may have been due to entrainment from aloft or by homogenous nucleation downwind of the hill. During the other cloud event, the formation of an ultrafine particle mode (nucleation mode) occurred probably after the cloud passage. There was no evidence of entrainment during this time period. Multicomponent homogeneous nucleation models were used to simulate the formation of new particle downwind of an orographic cloud. Possible homogeneous nucleation processes for this could be the formation of sulphuric acid or ammonium chloride due to outgassing of hydrochloric acid. It was not possible, however, to simulate formation rates of new particles as observed downwind the hill using a model for the binary or ternary homogeneous nucleation process of ammonia and hydrochloric acid. During the first event with high sulphur dioxide concentrations, the formation of new particle via binary homogeneous nucleation of sulphuric acid and water could be only predicted using a high nighttime hydroxyl radical concentration. No formation of sulphuric acid particle could be simulated during the second event with low sulphur dioxide concentrations.

Cloud processing of soluble gases

Abstract Experimental data from the Great Dun Fell Cloud Experiment 1993 were used to investigate interactions between soluble gases and cloud droplets. Concentrations of H 2 O 2 , SO 2 , CH 3 COOOH, HCOOH, and HCHO were monitored at different sites within and downwind of a hill cap cloud and their temporal and spatial evolution during several cloud events was investigated. Significant differences were found between in-cloud and out-of-cloud concentrations, most of which could not be explained by simple dissolution into cloud droplets. Concentration patterns were analysed in relation to the chemistry of cloud droplets and the gas/liquid equilibrium. Soluble gases do not undergo similar behaviour: CH 3 COOH simply dissolves in the aqueous phase and is outgassed upon cloud dissipation; instead, SO 2 is consumed by its reaction with H 2 O 2 . The behaviour of HCOOH is more complex because there is evidence for in-cloud chemical production. The formation of HCOOH interferes with the odd hydrogen cycle by enhancing the liquid-phase production of H 2 O 2 . The H 2 O 2 concentration in cloud therefore results from the balance of consumption by oxidation of SO 2 in-cloud production, and the rate by which it is supplied to the system by entrainment of new air into the clouds.

Phase partitioning of aerosol particles in clouds at Kleiner Feldberg

The partitioning of aerosol particles between cloud droplets and interstitial air by number and volume was determined both in terms of an integral value and as a function of size for clouds on Mt. Kleiner Feldberg (825 m asl), in the Taunus Mountains north-west of Frankfurt am Main, Germany. Differences in the integral values and the size dependent partitioning between two periods during the campaign were observed. Higher number and volume concentrations of aerosol particles in the accumulation mode were observed during Period II compared to Period I. In Period I on average 87±11% (±one standard deviation) and 73±7% of the accumulation mode volume and number were incorporated into cloud droplets. For Period II the corresponding fractions were 42±6% and 12±2% in one cloud event and 64±4% and 18±2% in another cloud event. The size dependent partitioning as a function of time was studied in Period II and found to have little variation. The major processes influencing the partitioning were found to be nucleation scavenging and entrainment.

Observations and modelling of the processing of aerosol by a hill cap cloud

Abstract Observations are presented of the aerosol size distribution both upwind and downwind of the Great Dun Fell cap cloud. Simultaneous measurements of the cloud microphysics and cloud chemistry, and of the chemical composition of the aerosol both upwind and downwind of the hill were made along with measurements of sulphur dioxide, hydrogen peroxide and ozone. These observations are used for initialisation of, and for comparison with the predictions of a model of the air flow, cloud microphysics and cloud chemistry of the system. A broad droplet size distribution is often observed near to the hill summit, seemingly produced as a result of a complex supersaturation profile and by mixing between parcels with different ascent trajectories. The model generates several supersaturation peaks as the airstream ascends over the complex terrain, activating increasing numbers of droplets. In conditions where sulphate production in-cloud (due to the oxidation of S(IV) by hydrogen peroxide and ozone) is observed, there is a marked effect on the chemical evolution of the aerosol particles on which the droplets form. When sulphate production occurs, a significant modification of the aerosol size distribution and hygroscopic properties is both predicted and observed. The addition of sulphate mass to those aerosol particles nucleation scavenged by the cloud generally increases the ease with which they are subsequently able to act as cloud condensation nuclei (CCN). Often, this will lead to an increase in the number of CCN available for subsequent cloud formation, although this latter effect is shown to be strongly dependent upon the activation history of the droplets and the concentration of pollutant gases present in the interstitial air. Situations are also identified where cloud processing could lead to a reduction in the capacity of smaller aerosol to act as CCN.

Henry’s Law and the Behavior of Weak Acids and Bases in Fog and Cloud

Experimental data from two field experiments on ground based clouds were used to study the distribution of formic acid, acetic acid, ammonia and S(IV) species between liquid and gas phase. The ratio of the concentrations of these compounds between the phases during concurrent measurements was compared to ratios expected according to Henrys law (considering the pH influence). Large discrepancies of several orders of magnitude were seen. Three hypotheses have been investigated to explain the observed discrepancies: The existence of a microscale equilibrium which does not persist in a bulk sample, a thermodynamic shift of the equilibrium due to competing reactions, and nonequilibrium conditions due to mass transfer limitations. Approximate quantitative calculations show that none of these hypotheses is sufficient to explain all of the discrepancies, so a combination of different effects seems to be responsible for this observation. The same theoretical considerations also suggest that mass transfer limitation may be an important factor for highly soluble compounds. The data presented here indicates that it is not possible to simply extrapolate interstitial gas phase composition from measured bulk liquid phase concentrations of a fog or cloud.

Multiphase chemistry and acidity of clouds at Kleiner Feldberg

The chemistry of cloud multiphase systems was studied within the Kleiner Feldberg Cloud Experiment 1990. The clouds encountered during this experimental campaign could be divided into two categories according to the origin of air masses in which the clouds formed. From the chemical point of view, clouds passing the sampling site during the first period of the campaign (26 October-4 November) were characterized by lower pollutant loading and higher pH, as compared to clouds during the final period of the experimental campaign (10–13 November). The study of multiphase partitioning of the main chemical constituents of the cloud systems and of atmospheric acidity within the multiphase systems themselves (gas + interstitial aerosol + liquid droplets) are presented in this paper. A general lack of gaseous NH3 was found in these cloud systems, which caused a lack of buffer capacity toward acid addition. Evidence supports the hypothesis that the higher acidity of the cloud systems during this final period of the campaign was due to input of HNO3. Our measurements, however, could not determine whether the observed input was due to scavenging of gaseous HNO3 from the air feeding into the cloud, or to heterogeneous HNO3 formation via NO2 oxidation by O3 to NO3 and N2O5. Sulfate in cloud droplets mainly originated from aerosol SO42− scavenging, since S(IV) to S(VI) liquid phase conversion was inhibited due to both lack of H2O2 and low pH of cloud droplets, which made O3 and metal catalyzed S(IV) oxidation inefficient.

METEOROLOGY OF THE GREAT DUN FELL CLOUD EXPERIMENT 1993

Synoptic and local meteorological conditions during the Spring 1993 Ground-based Cloud Experiment on Great Dun Fell are described, including cloud microphysics, general pollution levels and sources of air, especially for five case studies selected for detailed analysis. Periods when air was flowing across the hill are identified and the extent to which air mixed into the cloud from above reached the ground is estimated. To aid the interpretation of cloud chemistry and microphysics measurements, the horizontal and vertical extent of the cloud are used to estimate droplet lifetimes and to comment on the influence of complex terrain on peak supersaturation.

Microphysics of clouds at Kleiner Feldberg

During a field measuring campaign at Kleiner Feldberg (Taunus) in 1990, microphysical characteristics of clouds have been measured by Forward Scattering Spectrometer Probes (FSSP). The aim was to study the influence of aerosol and meteorological factors on droplet size and number. The results are: More mass in the accumulation size range of the aerosol leads to more droplets in stratocumulus clouds and to higher soluble masses in droplets of stratus clouds. However, the aerosol distribution was coarser in the stratus clouds compared to the stratocumulus clouds. Within the first 200 m from cloud base, the droplets grow while their number decreases. The growth results in a stable size of about 14 µm diameter over a large distance from cloud base in many stratocumulus clouds. Two types of mixing processes were observed: processes with reductions in the number of droplets (inhomogeneous mixing) and with reductions in the size of the droplets (homogeneous mixing).

Computer modelling of clouds at Kleiner Feldberg

The airflow, cloud microphysics and gas- and aqueous-phase chemistry on Kleiner Feldberg have been modelled for the case study of the evening of 1 November 1990, in order to calculate parameters that are not easily measured in the cloud and thus to aid the interpretation of the GCE experimental data-set. An airflow model has been used to produce the updraught over complex terrain for the cloud model, with some care required to ensure realistic modelling of the strong stable stratification of the atmosphere. An extensive set of measurements has been made self-consistent and used to calculate gas and aerosol input parameters for the model. A typical run of the cloud model has calculated a peak supersaturation of 0.55% which occurs about 20 s after entering cloud where the updraught is 0.6 m s−1. This figure has been used to calculate the efficiency with which aerosol particles were scavenged; it is higher than that calculated by other methods, and produces a cloud with slightly too many droplets. A broad cloud droplet size spectrum has been produced by varying the model inputs to simulate turbulent mixing and fluctuations in cloud parameters in space and time, and the ability of mixing processes near cloud-base to produce a lower peak supersaturation is discussed. The scavenging of soluble gases by cloud droplets has been observed and departures from Henrys Law in bulk cloud-water samples seen to be caused by variation of pH across the droplet spectrum and the inability of diffusion to adjust initial distributions of highly soluble substances across the spectrum in the time available. Aqueous-phase chemistry has been found to play a minor role in the cloud as modelled, but circumstances in which these processes would be more important are identified.