-
Long-term and large-scale correlations between Advanced Very High-Resolution Radiometer (AVHRR) aerosol optical depth and International Satellite Cloud Climatology Project (ISCCP) monthly cloud amount data show significant regional scale relationships between cloud amount and aerosols, consistent with aerosol-cloud interactions. Positive correlations between aerosols and cloud amount are associated with North American and Asian aerosols in the North Atlantic and Pacific storm tracks, and mineral aerosols in the tropical North Atlantic. Negative correlations are seen near biomass burning regions of North Africa and Indonesia, as well as south of the main mineral aerosol source of North Africa. These results suggest that there are relationships between aerosols and clouds in the observations that can be used by general circulation models to verify the correct forcing mechanisms for both direct and indirect radiative forcing by clouds.
-
Atmospheric aerosol particles typically consist of inorganic salts and organic material. The inorganic compounds as well as their hygroscopic properties are well defined, but the effect of organic compounds on cloud droplet activation is still poorly characterized. The focus of the present study is the organic compounds that are surface active i.e. tend to concentrate on droplet surface and decrease the surface tension. Gibbsian surface thermodynamics was used to find out how partitioning between droplet surface and the bulk of the droplet affects the surface tension and the surfactant bulk concentration in droplets large enough to act as cloud condensation nuclei. Sodium dodecyl sulfate (SDS) was used together with sodium chloride to investigate the effect of surfactant partitioning on the Raoult effect (solute effect). While accounting for the surface to bulk partitioning is known to lead to lowered bulk surfactant concentration and thereby to increased surface tension compared to a case in which the partitioning is neglected, the present results show that the partitioning also alters the Raoult effect, and that the change is large enough to further increase the critical supersaturation and hence decrease cloud droplet activation. The fraction of surfactant partitioned to droplet surface increases with decreasing droplet size, which suggests that surfactants might enhance the activation of larger particles relatively more thus leading to less dense clouds. Cis-pinonic acid-ammonium sulfate aqueous solutions were studied in order to study the partitioning with compounds found in the atmosphere and to find out the combined effects of dissolution and partitioning behavior. The results show that the partitioning consideration presented in this paper alters the shape of the Köhler curve when compared to calculations in which the partitioning is neglected either completely or in the Raoult effect. In addition, critical supersaturation was measured for SDS particles with dry radii of 25-60nm using a static parallel plate Cloud Condensation Nucleus Counter. The experimentally determined critical supersaturations agree very well with theoretical calculations taking the surface to bulk partitioning fully into account and are much higher than those calculated neglecting the partitioning.
-
Smoke-haze episodes caused by vegetation and peat fires affect parts of Indonesia every year with significant impacts on human health and climate. Particularly fires in degenerated peat areas release huge amounts of trace gases, e.g. CO<sub>2</sub>, CO and CH<sub>4</sub>, and particles into the atmosphere, exceeding by far the emissions per unit area from fires in surface vegetation. However, only limited information is available about the current distribution of pristine and degenerated peat areas in Indonesia, their depth, drainage condition and modification by fire. Particularly during the strong El Niño event in 1997/1998 a huge uncertainty exists about the contribution of Indonesian peat fire emissions to the measured increase of atmospheric CO<sub>2</sub>, as the published estimates of the peat area burned differ considerably. In this paper we study the contribution of peat fire emissions in Indonesia during the El Niño event 1997/1998. A regional three-dimensional atmosphere-chemistry model is applied over Indonesia using two emission estimates. These vegetation and peat fire emission inventories for Indonesia are set up in 0.5° resolution in weekly intervals and differ only in the size of the fire affected peat areas. We evaluate simulated rainfall and particle concentrations by comparison with observations to draw conclusions on the total carbon emissions released from the vegetation and peat fires in Indonesia in 1997/1998.
-
A Lagrangian trajectory model used to simulate photochemistry has been extended to include a simple parameterisation of primary and secondary aerosol particles. The model uses emission inventories of primary particles for the UK from the NAEI (National Atmospheric Emissions Inventory for the UK), and for Europe from the TNO (Institute of Environmental Sciences, Energy Research and Process Innovation, the Netherlands) respectively, to transport tracers representing PM<sub>10</sub>. One biogenic and two anthropogenic organic compounds were chosen as surrogates to model the formation of condensable material suitable for the production of secondary organic aerosol (SOA). The SOA is added to the primary PM<sub>10</sub> and compared to measured PM<sub>10</sub> at one urban and two rural UK receptor sites. The results show an average under-prediction by factors of 4.5 and 8.9 in the urban and rural cases respectively. The model is also used to simulate production of two secondary inorganic species, H<sub>2</sub>SO<sub>4</sub> and HNO<sub>3</sub>, which are assumed, as a limiting case, to be present in the particle phase. The relationships between modelled and measured total PM<sub>10</sub> improved with the addition of secondary inorganic compounds, and the overall model under-prediction factors are reduced to 3.5 and 3.9 in the urban and rural cases respectively. Nevertheless, our conclusion is that current emissions and chemistry do not appear to provide sufficient information to model PM<sub>10</sub> well (i.e. to within a factor of two). There is a need for further process studies to inform global climate modelling that includes climate forcing by aerosol.
-
Water vapour has been measured from the International Scientific Station Jungfraujoch (ISSJ, 47° N, 7° E, 3580m above sea level) during the winters of 1999/2000 and 2000/2001 by microwave radiometry and Raman lidar. The abundance of atmospheric water vapour between the planetary boundary layer and the upper stratosphere varies over more than three orders of magnitude. The currently used measurement techniques are only suited to determine the abundance of water vapour in different atmospheric regimes. None can resolve the vertical distribution profile from ground level to the top of the stratosphere by itself. We present such a water vapour profile where simultaneous measurements from a Raman lidar and a microwave radiometer were combined to cover both the troposphere and the stratosphere, respectively. We also present a study of the stratospheric and tropospheric water vapour variability for the two consecutive winters.
-
Chemical ozone loss in the Arctic stratosphere was investigated for the twelve years between 1991 and 2003 employing the ozone-tracer correlation method. For this method, the change in the relation between ozone and a long-lived tracer is considered for all twelve years over the lifetime of the polar vortex to calculate chemical ozone loss. Both the accumulated local ozone loss in the lower stratosphere and the column ozone loss were derived consistently, mainly on the basis of HALOE satellite observations. HALOE measurements do not cover the polar region homogeneously over the course of the winter. Thus, to derive an early winter reference function for each of the twelve years, all available measurements were additionally used; for two winters climatological considerations were necessary. Moreover, a detailed quantification of uncertainties was performed. This study further demonstrates the interaction between meteorology and ozone loss. The connection between temperature conditions and chlorine activation, and in turn, the connection between chlorine activation and ozone loss, becomes obvious in the HALOE HCl measurements. Additionally, the degree of homogeneity of ozone loss within the vortex was shown to depend on the meteorological conditions. <P style="line-height: 20px;"> Results derived here are in general agreement with the results obtained by other methods for deducing polar ozone loss. Differences occur mainly owing to different time periods considered in deriving accumulated ozone loss. However, very strong ozone losses as deduced from SAOZ for January in winters 1993-1994 and 1995-1996 cannot be identified using available HALOE observations in the early winter. In general, strong accumulated ozone loss was found to occur in conjunction with a strong cold vortex containing a large volume of possible PSC existence (V<sub>PSC</sub>), whereas moderate ozone loss was found if the vortex was less strong and moderately warm. Hardly any ozone loss was calculated for very warm winters with small amounts of V<sub>PSC</sub> during the entire winter. This study supports the linear relationship between V<sub>PSC</sub> and the accumulated ozone loss reported by Rex et al. (2004) if V<sub>PSC</sub> was averaged over the entire winter period. Here, further meteorological factors controlling ozone loss were additionally identified if V<sub>PSC</sub> was averaged over the same time interval as that for which the accumulated ozone loss was deduced. A significant difference in ozone loss (of ≈36DU) was found due to the different duration of solar illumination of the polar vortex of at maximum 4 hours per day in the observed years. Further, the increased burden of aerosols in the atmosphere after the Pinatubo volcanic eruption in 1991 significantly increased the extent of chemical ozone loss.
-
We investigate the impact of convection on the thermal structure of the Tropical Tropopause Layer (TTL). We use temperature profiles measured by the Atmospheric Infrared Sounder (AIRS) onboard the Aqua satellite, and the time evolution of local convection determined by the National Centers for Environmental Protection/Aviation Weather Center (NCEP/AWS) half-hourly infrared global geostationary composite. The observations demonstrate that the TTL is cooled by convection, in agreement with previous observations and model simulations. By using a global data set, we are able to investigate the variations in this convective cooling by season and region. The estimated cooling rate during active convection is 7.5~9 K/day. While we cannot unambiguously identify the cause of this cooling, our analysis suggests that radiative cooling is likely not an explanation.
-
Photolysis of water-soluble components inside cloud droplets by ultraviolet/visible radiation may play an important role in atmospheric chemistry. Two earlier studies have suggested that the actinic flux and hence the photolysis frequency within spherical droplets is enhanced relative to that in the surrounding air, but have given different values for this enhancement. Here, we reconcile these discrepancies by noting slight errors in both studies that, when corrected, lead to consistent results. Madronich (1987) examined the geometric (large droplet) limit and concluded that refraction leads to an enhancement factor, averaged over all incident directions, of 1.56. However, the physically relevant quantity is the enhancement of the average actinic flux (rather than the average enhancement factor) which we show here to be 1.26 in the geometric limit. Ruggaber et al. (1997) used Mie theory to derive energy density enhancements slightly larger than 2 for typical droplet sizes, and applied these directly to the calculation of photolysis rates. However, the physically relevant quantity is the actinic flux (rather than the energy density) which is obtained by dividing the energy density by the refractive index of water, 1.33. Thus, the Mie-predicted enhancement for typical cloud droplet sizes is in the range 1.5, only coincidentally in agreement with the value originally given by Madronich. We also investigated the influence of resonances in the actinic flux enhancement. These narrow spikes which are resolved only by very high resolution calculations are orders of magnitude higher than the intermediate values but contribute only little to the actinic flux enhancement when averaged over droplet size distributions. Finally, a table is provided which may be used to obtain the actinic flux enhancement for the photolysis of any dissolved species.
-
A new 3-D mercury model has been developed within the Danish Arctic Monitoring and Assessment Programme (AMAP). The model is based on the Danish Eulerian Hemispheric Model, which in the original version has been used to study the transport of SO<sub>2</sub>, SO<sub>4</sub><sup>2-</sup> and Pb into the Arctic. It was developed for sulphur in 1990 and in 1999 also lead was included. For the current study a chemical scheme for mercury has been included and the model is now applied to the mercury transport problem. Some experiments with the formulation of the mercury chemistry during the Polar Sunrise are carried out in order to investigate the observed depletion. Some of the main conclusions of the work described in this paper are that atmospheric transport of mercury is a very important pathway into the Arctic and that mercury depletion in the Arctic troposphere during the Polar Sunrise contributes considerably to the deposition of mercury in the Arctic.
-
Trends in derived from a 45 year integration of a chemistry-climate model (CCM) run have been compared with ground-based measurements at Lauder (45° S) and Arrival Heights (78° S). Observed trends in at both sites exceed the modelled trends in N<sub>2</sub>O, the primary source gas for stratospheric NO<sub>2</sub>. This suggests that the processes driving the trend are not solely dictated by changes in but are coupled to global atmospheric change, either chemically or dynamically or both. If CCMs are to accurately estimate future changes in ozone, it is important that they comprehensively include all processes affecting NO<sub>x</sub> (NO+NO<sub>2</sub>) because NO<sub>x</sub> concentrations are an important factor affecting ozone concentrations. Comparison of measured and modelled NO<sub>2</sub> trends is a sensitive test of the degree to which these processes are incorporated in the CCM used here. At Lauder the 1980-2000 CCM NO<sub>2</sub> trends (4.2% per decade at sunrise, 3.8% per decade at sunset) are lower than the observed trends (6.5% per decade at sunrise, 6.0% per decade at sunset) but not significantly different at the 2σ level. Large variability in both the model and measurement data from Arrival Heights makes trend analysis of the data difficult. CCM predictions (2001-2019) of NO<sub>2</sub> at Lauder and Arrival Heights show significant reductions in the rate of increase of NO<sub>2</sub> compared with the previous 20 years (1980-2000). The model results indicate that the partitioning of oxides of nitrogen changes with time and is influenced by both chemical forcing and circulation changes.
