-
In this paper we document "SOCOL", a new chemistry-climate model, which has been ported for regular PCs and shows good wall-clock performance. An extensive validation of the model results against present-day climate data obtained from observations and assimilation data sets shows that the model describes the climatological state of the atmosphere for the late 1990s with reasonable accuracy. The model has a significant temperature bias only in the upper stratosphere and near the tropopause at high latitudes. The latter is the result of the rather low vertical resolution of the model near the tropopause. The former can be attributed to a crude representation of radiation heating in the middle atmosphere. A comparison of the simulated and observed link between the tropical stratospheric structure and the strength of the polar vortex shows that in general, both observations and simulations reveal a higher temperature and ozone mixing ratio in the lower tropical stratosphere for the case with stronger Polar night jet (PNJ) and slower Brewer-Dobson circulation as predicted by theoretical studies.
-
This presentation concentrates on solar occultation measurements with the spaceborne spectrometer SCIAMACHY in the UV-Vis wavelength range. Solar occultation measurements provide unique information about the vertical distribution of atmospheric constituents. For retrieval of vertical trace gas concentration profiles, an algorithm has been developed based on the optimal estimation method. The forward model is capable of simulating the extinction signals of different species as they occur in atmospheric transmission spectra obtained from occultation measurements. Furthermore, correction algorithms have been implemented to address shortcomings of the tangent height pre-processing and inhomogeneities of measured solar spectra. First results of O<sub>3</sub> and NO<sub>2</sub> vertical profile retrievals have been validated with data from ozone sondes and satellite based occultation instruments. The validation shows very promising results for SCIAMACHY O<sub>3</sub> and NO<sub>2</sub> values between 15 to 35km with errors of the order of 10% and 15%, respectively.
-
Ozone and nitrous oxide are measured at high spatial and temporal resolution by instruments flying on the ER-2 NASA research aircraft. Comparing the airborne transects to reconstructions by ensemble of diffusive backward trajectories allows estimation of the average vertical Lagrangian turbulent diffusion experienced by the air parcels. The resulting estimates show large Lagrangian diffusion of the order of 0.1in the surf zone outside the polar vortex and smaller values of the order of 0.01 inside. Locally, large variation of Lagrangian diffusion occurs over mesoscale distances. It is found that high temporal resolution (3h or less) is required for off-line transport calculations and that the reconstructions are sensitive to spurious motion in standard analysed winds.
-
SCIAMACHY onboard the European environmental research satellite ENVISAT is an UV/visible/near-infrared spectrometer providing 3 near infrared channels covering wavelengths from 1-1.75 µm, 1.94-2.04 µm and 2.26-2.38 µm with moderate spectral resolution (0.22-1.5nm). From their structured absorption in these spectral regions, we can quantitatively determine the total column densities of the greenhouse gases CO<sub>2</sub>, CH<sub>4</sub>, N<sub>2</sub>O and H<sub>2</sub>O as well as of CO. A modified DOAS algorithm based on optimal estimation (IMAP-DOAS) has been developed at the University of Heidelberg to account for the peculiarities of these absorbers. CO is a relatively weak absorber whose spectral signature is overlapped by strong CH<sub>4</sub> and H<sub>2</sub>O absorptions. Hence, retrieval of CO from SCIAMACHY spectra (within 2.26-2.38 µm) is a challenging task. Therefore, the calibration of the raw spectra with respect to dark current issues and nonlinearity were analysed in detail and substantially improved to enable reasonable retrieval of CO. This paper focusses on first results of the CO retrieval where various sources like biomass burning events and their seasonal variability can be clearly identified.
-
This paper describes a new ozone column retrieval algorithm and its application to SCIAMACHY measurements. The TOSOMI algorithm is based on the Differential Optical Absorption Spectroscopy (DOAS) technique and implements several improvements over older algorithms. These improvements include aspects like (i) the explicit treatment of rotational Raman scattering, (ii) an improved air-mass factor formulation which is based on a simulation of the reflectivity spectrum and a subsequent DOAS fit of this simulated spectrum, (iii) the use of an improved ozone climatology and a column dependent air-mass factor, (iv) the use of daily varying ECMWF temperature profile analyses. The results of three validation exercises are reported. The TOSOMI columns are compared with an extensive set of ground-based observations (Brewer, Dobson) for the years 2003 and 2004. Secondly, a direct comparison for January–June 2003 with two new GOME retrievals, GDP Version 4 and TOGOMI, is presented. Third, data assimilation is used to study the dependence of the TOSOMI columns with retrieval parameters such as the viewing angle, cloud fraction and geographical location. These comparisons show a good consistency on the percent level between the GOME and SCIAMACHY algorithms. The present TOSOMI implementation (v0.32) shows an offset of about −1.5% with respect to ground-based observations and the GOME retrievals.
-
Understanding the impact of atmospheric dynamical variability on observed changes in stratospheric O<sub>3</sub> is a key to understanding how O<sub>3</sub> will change with future climate dynamics and trace gas abundances. In this paper we examine the linkage between interannual variability in total column O<sub>3</sub> at northern high latitudes in March and lower-to-mid stratospheric vortex O<sub>3</sub> in the prior November. We find that these two quantities are significantly correlated in the years available from TOMS, SBUV, and POAM data (1978-2004). Additionally, we find that the increase in March O<sub>3</sub> variability from the 1980s to years post-1990 is also seen in the November vortex O<sub>3</sub>, i.e., interannual variability in both quantities is much larger in the later years. The cause of this correlation is not clear, however. Interannual variations in March total O<sub>3</sub> are known to correspond closely with variations in winter stratospheric wave driving consistent with the effects of varying residual circulation, temperature, and chemical loss. Variation in November vortex O<sub>3</sub> may also depend on dynamical wave activity, but the dynamics in fall are less variable than in winter and spring. We do not find significant correlations of dynamic indicators for November such as temperature, heat flux, or polar average total O<sub>3</sub> with the November vortex O<sub>3</sub>, nor with dynamical indicators later in winter and spring that might lead to a connection to March. We discuss several potential hypotheses for the observed correlation but do not find strong evidence for any considered mechanism. We present the observations as a phenomenon whose understanding may improve our ability to predict the dependence of O<sub>3</sub> on changing dynamics and chemistry.
-
The Optimal Estimation Method is used to retrieve temperature and water vapor profiles from simulated radio occultation measurements in order to assess how different retrieval schemes may affect the assimilation of this data. High resolution ECMWF global fields are used by a state-of-the-art radio occultation simulator to provide quasi-realistic bending angle and refractivity profiles. Both types of profiles are used in the retrieval process to assess their advantages and disadvantages. The impact of the GPS measurement is expressed as an improvement over the a priori knowledge (taken from a 24h old analysis). Large improvements are found for temperature in the upper troposphere and lower stratosphere. Only very small improvements are found in the lower troposphere, where water vapor is present. Water vapor improvements are only significant between about 1 km to 7 km. No pronounced difference is found between retrievals based upon bending angles or refractivity. Results are compared to idealized retrievals, where the atmosphere is spherically symmetric and instrument noise is not included. Comparing idealized to quasi-realistic calculations shows that the main impact of a ray tracing algorithm can be expected for low latitude water vapor, where the horizontal variability is high. We also address the effect of altitude correlations in the temperature and water vapor. Overall, we find that water vapor and temperature retrievals using bending angle profiles are more CPU intensive than refractivity profiles, but that they do not provide significantly better results.
-
Monthly mean aerosol optical depth (AOD) over ocean is compared from a total of 9 aerosol retrievals during a 40 months period. Comparisons of AOD have been made both for the entire period and sub periods. We identify regions where there is large disagreement and good agreement between the aerosol satellite retrievals. Significant differences in AOD have been identified in most of the oceanic regions. Several analyses are performed including spatial correlation between the retrievals as well as comparison with AERONET data. During the 40 months period studied there have been several major aerosol field campaigns as well as events of high aerosol content. It is studied how the aerosol retrievals compare during such circumstances. The differences found in this study are larger than found in a previous study where 5 aerosol retrievals over an 8 months period were compared. Part of the differences can be explained by limitations and deficiencies in some of the aerosol retrievals. In particular, results in coastal regions are promising especially for aerosol retrievals from satellite instruments particularly suited for aerosol research. In depth analyses explaining the differences between AOD obtained in different retrievals are clearly needed. We limit this study to identifying differences and similarities and indicating possible sources that affect the quality of the retrievals. This is a necessary first step towards understanding the differences and improving the retrievals.
-
Phenols are a major class of volatile organic compounds (VOC) whose reaction within, and partitioning between, the gas and liquid phases affects their lifetime within the atmosphere, the local oxidising capacity, and the extent of production of nitrophenols, which are toxic chemicals. In this work, a zero-dimension box model was constructed to quantify the relative importance of different nitration pathways, and partitioning into the liquid phase, of mono-aromatic compounds in order to help elucidate the formation pathways of 2- and 4-nitrophenol in the troposphere. The liquid phase contributed significantly to the production of nitrophenols for liquid water content (<i>L<sub>c</sub></i>) values exceeding 3x10<sup>-9</sup>, and for a range of assumed liquid droplet diameter, even though the resultant equilibrium partitioning to the liquid phase was much lower. For example, in a "typical" model scenario, with <i>L<sub>c</sub></i>=3x10<sup>-7</sup>, 58% of nitrophenol production occurred in the liquid phase but only 2% of nitrophenol remained there, i.e. a significant proportion of nitrophenol observed in the gas phase may actually be produced via the liquid phase. The importance of the liquid phase was enhanced at lower temperatures, by a factor ~1.5-2 at 278K c.f. 298K. The model showed that nitrophenol production was particularly sensitive to the values of the rate coefficients for the liquid phase reactions between phenol and OH or NO<sub>3</sub> reactions, but insensitive to the rate coefficient for the reaction between benzene and OH, thus identifying where further experimental data are required.
-
A decadal-scale trend in the tropical radiative energy budget has been observed recently by satellites, which however is not reproduced by climate models. In the present study, we have computed the outgoing shortwave radiation (OSR) at the top of atmosphere (TOA) at 2.5° longitude-latitude resolution and on a mean monthly basis for the 17-year period 1984-2000, by using a deterministic solar radiative transfer model and cloud climatological data from the International Satellite Cloud Climatology Project (ISCCP) D2 database. Anomaly time series for the mean monthly pixel-level OSR fluxes, as well as for the key physical parameters, were constructed. A significant decreasing trend in OSR anomalies, starting mainly from the late 1980s, was found in tropical and subtropical regions (30° S-30° N), indicating a decadal increase in solar planetary heating equal to 1.9±0.3Wm<sup>-2</sup>/decade, reproducing well the features recorded by satellite observations, in contrast to climate model results. This increase in solar planetary heating, however, is accompanied by a similar increase in planetary cooling, due to increased outgoing longwave radiation, so that there is no change in net radiation. The model computed OSR trend is in good agreement with the corresponding linear decadal decrease of 2.5±0.4Wm<sup>-2</sup>/decade in tropical mean OSR anomalies derived from ERBE S-10N non-scanner data (edition 2). An attempt was made to identify the physical processes responsible for the decreasing trend in tropical mean OSR. A detailed correlation analysis using pixel-level anomalies of model computed OSR flux and ISCCP cloud cover over the entire tropical and subtropical region (30° S-30° N), gave a correlation coefficient of 0.79, indicating that decreasing cloud cover is the main reason for the tropical OSR trend. According to the ISCCP-D2 data derived from the combined visible/infrared (VIS/IR) analysis, the tropical cloud cover has decreased by 6.6±0.2% per decade, in relative terms. A detailed analysis of the inter-annual and long-term variability of the various parameters determining the OSR at TOA, has shown that the most important contribution to the observed OSR trend comes from a decrease in low-level cloud cover over the period 1984-2000, followed by decreases in middle and high-level cloud cover. Note, however, that there still remain some uncertainties associated with the existence and magnitude of trends in ISCCP-D2 cloud amounts. Opposite but small trends are introduced by increases in cloud scattering optical depth of low and middle clouds.
