A54A-01 INVITED 16:00h
Validation of SCIAMACHY - current and future product qualities
SCIAMACHY on board of ENVISAT measures globally the vertical distribution and total column amounts of several atmospheric constituents. The values retrieved from the observed spectra, both operationally and non-operationally, are subject to a continuous validation process, in which $38$ European and non-European institutes participate. The SCIAMACHY validation product coordinators monitor the results of this process and regularly asses the accuracy of each product. The product accuracy is described in such a way that it is of direct use for algorithm improvement, instrument characterization and atmospheric research. In this presentation we will summarize and illustrate the current quality of the SCIAMACHY products. Also we will outline our plans for the coming years to ensure a regularly updated and complete description of the accuracy of each SCIAMACHY product. Finally some ideas on synergy between SCIAMACHY and EOS-Aura validation will be discussed.
http://www.temis.nl
A54A-02 16:15h
Can one satellite data set validate another? Validation of Envisat SCIAMACHY data by comparisons with NOAA-16 SBUV/2 and ERS-2 GOME
Validation of satellite data remains a high priority for the construction of climate data sets. Traditionally ground based measurements have provided the primary comparison data for validation. For some atmospheric parameters such as ozone, a thoroughly validated satellite data record can be used to validate a new instrument's data product in addition to using ground based data. Comparing validated data with new satellite data has several advantages; availability of much more data, which will improve precision, larger geographical coverage, and the footprints are closer in size, which removes uncertainty due to different observed atmospheric volumes. To demonstrate the applicability, observations from the newly launched SCIAMACHY instrument were compared with the NOAA-16 SBUV/2 and ERS-2 GOME instruments. The SBUV/2 and GOME data had all ready undergone validation by comparing to the total ozone ground network. Overall the SCIAMACHY (near real time) data were found to be low by 3 percent with respect to satellite data and 1percent low with respect to ground station data. There appears to be seasonal and or solar zenith angle dependencies in the comparisons with SBUV/2 where differences increase with higher solar zenith angles. It is known that accuracies in both satellite and ground based total ozone algorithms decrease at high solar zenith angles. Therefore there is a strong need for more accurate ground validation measurements under these conditions.
A54A-03 16:30h
USING THE MANTRA BALLOON CAMPAIGN MEASUREMENTS FOR SATELLITE VALIDATION: FIRST RESULTS.
MANTRA (Middle Atmosphere Nitrogen TRend Assessment) is a series of high-altitude balloon campaigns conducted in late summer over Saskatchewan Canada (52°N, 107°W) to measure stratospheric trace gases from a float altitude of about 35 km. Flights have occurred in late August or early September of 1998, 2000, 2002, and 2004. By launching in late summer, dynamical variability is minimized and the changing chemical balance of the stratosphere can be studied, with an emphasis on the nitrogen partitioning. This scenario makes the measurements done during the MANTRA campaigns particularly appropriated for validation of satellite measurements. Vertical distributions of O3, NO2, N2O, BrO, HNO3, HCl, CFC-11, CFC-12, and CH4 were measured by the combination of a Fourier transform infra-red spectrometer, infrared emission filter radiometers, and UV-visible grating spectrometers. Other species such as N2O5, NO, ClONO2, may still be retrieved from the existing dataset. In addition, ozone profiles were obtained from a number of ozonesondes launched almost daily during the campaigns. A suite of ground-based instruments were operated providing O3 and NO2 total columns and vertical profiles. Here we present measurements from all the MANTRA campaigns and compare species measured with existing climatology and with the Canadian Middle Atmosphere Model (CMAM), a fully interactive chemistry-climate model. We discuss the use of the MANTRA dataset for satellite validation with a focus on NO2 and O3. Comparisons we are doing with OSIRIS instrument onboard the Odin satellite and the SCIAMACHY and MIPAS instruments onboard the ENVISAT satellite are discussed.
A54A-04 16:45h
Validation of the Odin/OSIRIS Stratospheric Ozone Profiles: Statistical Comparisons With Coincident Ozonesonde, POAM III, SAGE II, and SAGE III Measurements
Since November 2001, stratospheric ozone profiles have been regularly retrieved from the Odin/OSIRIS limb measurements of the scattered sunlight in the Chappuis absorption band. Profiles are calculated on a 2 km vertical grid between 10 km and 50 km using the paired radiance method developed by Flittner et al. [GRL, 27, 2601-2604, 2000] and McPeters et al. [GRL, 27, 2597-2600, 2000] for the SOLSE/LORE experiment. OSIRIS ozone profiles (version 012) have been freely available to the international research community at http://osirus.usask.ca since January 15, 2004. In this work we present the results of statistical intercomparison of the OSIRIS stratospheric ozone profiles with coincident profiles measured by POAM III, SAGE II, SAGE III and ozonesondes. The agreement between all instruments in the altitude range of 15 km to 35 km is within 7-10 percent except for the April-July 2002, and March-June, 2003 periods, when there may be an apparent downward altitude shift of up to 1 km. This apparent altitude shift increases the difference between OSIRIS and other instruments to about 30 percent below 15 km and above 35 km. The major reason for the altitude offset is presently attributed to an incorrect Odin attitude registration.
A54A-05 17:00h
Effects of Field Inhomogeneity on High-Latitude Ozone Measurements From Spaceborne Remote Sensing
Existing and future satellite retrievals and their validation are rendered problematic by spatial gradients and variability in species composition on both large and small spatial scales (e.g., high-latitude ozone). High--solar zenith angle observations of slant column ozone made during the second SAGE~III Ozone Loss and Validation Experiment (SOLVE~II) are compared with forward model simulations based on (1) vertical column ozone profiles and assuming spatial homogeneity (1-D) and (2) simulations using 3-D ozone fields determined by an ozone proxy mapping technique, which captures ozone horizontal gradients. Retrieval--model comparisons clearly demonstrate the errors associated with the homogeneity assumption in the context of line-of-sight ozone during SOLVE~II. A discussion of inhomogeneity effects on spaceborne remote sensing measurements, such as from Aura and TOMS, and their validation is also presented.
A54A-06 17:15h
Lessons Learned From the POLARIS and TOMS$^3$-F Ozone Intercomparisons, With Applications to Aura Validation at High Latitudes
High-latitude ozone is important because this is the region where most of the global ozone loss occurs. Data from the POLARIS and TOMS$^3$-F field campaigns based in Fairbanks, AK raised the concern that Dobson spectrometers systematically underestimate total column ozone under conditions of high ozone and high solar zenith angles, exactly the conditions under which the sudden springtime loss of polar ozone occurs. Total ozone when measured at very high solar zenith angles ($>$80$\deg$) has never been rigorously validated in an absolute sense, either by satellite- or ground-based instruments or their intercomparison. UV/visible retrieval algorithms for ozone at very high solar zenith angles are complex for a variety of reasons, including low light levels and the fact that multiple scattering dominates the UV radiation field used in TOMS, OMI, Dobson and Brewer spectrometer algorithms. The validation of OMI total column ozone at high solar zenith angles is a high priority for Aura validation. While many science and technical issues were addressed and/or resolved by the POLARIS and TOMS$^3$-F campaigns, we learned a number of practical lessons on how to optimize the collection of data during intercomparisons between satellite and balloon- and ground-based instruments. Perhaps the most important issue was obtaining truly coincident observations in time and space; simple co-location of the instruments and making measurements on the same date is not sufficient for rigorous validation of satellite observations. Comparisons with POAM III observations illustrated that even near-coincident observations can have significant differences in the ozone profile between the peak of the ozone concentration and the tropopause due to dynamical activity in the lowermost stratosphere. Variability on much shorter timescales was identified by the comparison of Dobson and Brewer total ozone observations, indicating that the atmosphere can be active even on days that are relatively cloud-free and apparently quiescent to the naked eye. The use of GPS transmitters on the ozonesondes and timing the launch of the balloon sondes such that they are at the peak of the ozone layer at the time of the satellite overpasses allowed us to further narrow the definition of ``coincident'' observations. Future ozone validation campaigns should plan to include laboratory observations on the actual field instruments (measuring the instrument's wavelength-dependent slit function, stray light, and forward-scattered light through the entrance optics), which along with the atmospheric observations can yield a quantitative means of correcting Dobson and single-pass Brewer data to better match the double-pass Brewer and satellite instrument observations obtained under conditions of high slant column ozone ($>$800 DU). Post-campaign re-processing of the raw data should include taking into account the actual stratospheric temperature profile in the Dobson retrievals, making use of sonde profiles in the satellite retrievals, assessing the impact of ozone field inhomogeneity using simultaneous vertical ozone profiling instruments (sondes or lidar) co-located with ground-based solar viewing spectrometers (Dobson, Brewer, SAOZ, etc.), all of which might result in an improved agreement between satellite and ground-based observations.
A54A-07 17:30h
Direct Tropospheric Ozone Retrieval from Global Ozone Monitoring Experiment (GOME)
We present the first directly retrieved global distribution of tropospheric ozone from satellite-based measurements. Tropospheric ozone and ozone profiles are derived from backscattered radiances in the ultraviolet spectra (i.e., 290-340 nm) measured by the nadir-viewing GOME. Tropospheric ozone is directly retrieved with daily NCEP/NCAR tropopause pressure as one of the retrieval levels. We perform detailed treatments of radiometric and wavelength calibrations and improve forward model inputs to achieve the fitting precision needed for tropospheric ozone. The calibrations include (a) variable slit width and shifts between radiance and irradiances, (b) real-time first-order Ring effect correction, (c) undersampling correction, (d) shift between measurements and trace gas cross sections, and (e) polarization correction. The TOMS Verison-8 ozone climatology is used to initialize a priori ozone profiles. The retrieved total ozone agrees well with TOMS V8 and Dobson total ozone, and tropospheric ozone and ozone profiles agree well with ozonesonde observations at 24 WOUDC stations from north pole (e.g., NyAlesund, 78.9°N) to south pole (e.g., Neumayer, 70.7°S).
A54A-08 17:45h
Insights into Tropospheric Ozone from the INTEX Ozonesonde Network Study (IONS)
Ozone profile data from soundings integrate models, aircraft and other ground-based measurements for better interpretation of atmospheric chemistry and dynamics. A well- designed network of ozonesonde stations, with consistent sampling, can answer questions not possible with short campaigns or current satellite technology. The SHADOZ (Southern Hemisphere Additional Ozonesondes) project, $<$http://croc.gsfc.nasa.gov/shadoz$>$, for example, has led to these findings about tropical ozone: definition of the zonal tropospheric wave-one pattern in equatorial ozone, characterization of the "Atlantic ozone paradox" and establishment of a link between tropical Atlantic and Indian Ocean pollution. Building on the SHADOZ concept, a short-term ozone network was formed in July-August 2004 to coordinate ozonesonde launches during the ICARTT/INTEX/NEAQS (International Consortium on Atmospheric Research on Transport and Transformation)/Intercontinental Transport Experiment/New England Air Quality Study. In IONS (INTEX Ozonesonde Network Study $<$http://croc.gsfc.nasa.gov/intex/ions.html$>$), more than 250 soundings, with daily frequency at half the sites, were launched from eleven North American stations and an oceanographic ship in the Gulf of Maine. Although the goal was to examine pollution influences under stable high-pressure systems and transport associated with "warm conveyor belt" flows, the INTEX study region was dominated by a series of weak frontal systems that mixed aged pollution with stratospheric ozone in the middle troposphere. Deconvoluting ozone sources provides new insights into ozone in the transition between mid-latitude and polar air.
http://croc.gsfc.nasa.gov/intex/ions.html