Alexander Vasilkov

First results from the OMI rotational Raman scattering cloud pressure algorithm

We have developed an algorithm to retrieve scattering cloud pressures and other cloud properties with the Aura Ozone Monitoring Instrument (OMI). The scattering cloud pressure is retrieved using the effects of rotational Raman scattering (RRS). It is defined as the pressure of a Lambertian surface that would produce the observed amount of RRS consistent with the derived reflectivity of that surface. The independent pixel approximation is used in conjunction with the Lambertian-equivalent reflectivity model to provide an effective radiative cloud fraction and scattering pressure in the presence of broken or thin cloud. The derived cloud pressures will enable accurate retrievals of trace gas mixing ratios, including ozone, in the troposphere within and above clouds. We describe details of the algorithm that will be used for the first release of these products. We compare our scattering cloud pressures with cloud-top pressures and other cloud properties from the Aqua Moderate-Resolution Imaging Spectroradiometer (MODIS) instrument. OMI and MODIS are part of the so-called A-train satellites flying in formation within 30 min of each other. Differences between OMI and MODIS are expected because the MODIS observations in the thermal infrared are more sensitive to the cloud top whereas the backscattered photons in the ultraviolet can penetrate deeper into clouds. Radiative transfer calculations are consistent with the observed differences. The OMI cloud pressures are shown to be correlated with the cirrus reflectance. This relationship indicates that OMI can probe through thin or moderately thick cirrus to lower lying water clouds.

Evaluation of the OMI cloud pressures derived from rotational Raman scattering by comparisons with other satellite data and radiative transfer simulations

[1] In this paper we examine differences between cloud pressures retrieved from the Ozone Monitoring Instrument (OMI) using the ultraviolet rotational Raman scattering (RRS) algorithm and those from the thermal infrared (IR) Aqua/MODIS. Several cloud data sets are currently being used in OMI trace gas retrieval algorithms including climatologies based on IR measurements and simultaneous cloud parameters derived from OMI. From a validation perspective, it is important to understand the OMI retrieved cloud parameters and how they differ with those derived from the IR. To this end, we perform radiative transfer calculations to simulate the effects of different geophysical conditions on the OMI RRS cloud pressure retrievals. We also quantify errors related to the use of the Mixed Lambert-Equivalent Reflectivity (MLER) concept as currently implemented of the OMI algorithms. Using properties from the Cloudsat radar and MODIS, we show that radiative transfer calculations support the following: (1) The MLER model is adequate for single-layer optically thick, geometrically thin clouds, but can produce significant errors in estimated cloud pressure for optically thin clouds. (2) In a two-layer cloud, the RRS algorithm may retrieve a cloud pressure that is either between the two cloud decks or even beneath the top of the lower cloud deck because of scattering between the cloud layers; the retrieved pressure depends upon the viewing geometry and the optical depth of the upper cloud deck. (3) Absorbing aerosol in and above a cloud can produce significant errors in the retrieved cloud pressure. (4) The retrieved RRS effective pressure for a deep convective cloud will be significantly higher than the physical cloud top pressure derived with thermal IR.

Aerosol ultraviolet absorption experiment (2002 to 2004), part 2: absorption optical thickness, refractive index, and single scattering albedo

Compared to the visible spectral region, very little is known about aerosol absorption in the UV. Without such information it is impos- sible to quantify the causes of the observed discrepancy between mod- eled and measured UV irradiances and photolysis rates. We report re- sults of a 17-month aerosol column absorption monitoring experiment conducted in Greenbelt, Maryland, where the imaginary part of effective refractive index k was inferred from the measurements of direct and diffuse atmospheric transmittances by a UV-multifilter rotating shadow- band radiometer (UV-MFRSR, U.S. Department of Agriculture (USDA) UV-B Monitoring and Research Network). Colocated ancillary measure- ments of aerosol effective particle size distribution and refractive index in the visible wavelengths (by CIMEL sun-sky radiometers, National Aero- nautics and Space Administration (NASA) Aerosol Robotic Network (AERONET)), column ozone, surface pressure, and albedo constrain the forward radiative transfer model input, so that a unique solution for k is obtained independently in each UV-MFRSR spectral channel. Inferred values of k are systematically larger in the UV than in the visible wave- lengths. The inferred k values enable calculation of the single scattering albedo v, which is compared with AERONET inversions in the visible

Improving total column ozone retrievals by using cloud pressures derived from Raman scattering in the UV

[1] The higher spectral resolution, coverage, and sampling of the Aura satellite ozone monitoring instrument (OMI), as compared with the total ozone mapping spectrometer (TOMS) should allow for improved ozone retrievals. By default, the TOMS-like OMI total column ozone algorithm uses climatological cloud-top pressures based on infrared (IR) measurements to estimate the column ozone below the clouds. Alternatively, cloud pressure can be retrieved using atmospheric rotational Raman scattering. The retrieved cloud pressures should be more consistent with assumptions made in the total ozone algorithm. Here, we use data from the global ozone monitoring experiment (GOME) to estimate total ozone using both the IR-climatological and retrieved cloud pressures. The resulting ozone differences can be significant but do not exceed � 15 DU. Use of the cloud pressure retrievals leads to a smoother distribution of ozone along a satellite track by reducing small spatial irregularities presumably caused by the difference between the retrieved and climatological cloud pressures. INDEX TERMS: 0320 Atmospheric Composition and Structure: Cloud physics and chemistry; 0360 Atmospheric Composition and Structure: Transmission and scattering of radiation; 3360 Meteorology and Atmospheric Dynamics: Remote sensing. Citation: Vasilkov, A. P., J. Joiner, K. Yang, and P. K. Bhartia (2004), Improving total column ozone retrievals by using cloud pressures derived from Raman scattering in the UV, Geophys. Res. Lett., 31, L20109, doi:10.1029/ 2004GL020603.

Seasonal variation of UV radiation in the ocean under clear and cloudy conditions

Seasonal variability of solar UV radiation in ocean waters is estimated on a global scale by combining satellite measurements of scene reflectivity (TOMS), column ozone (TOMS) and chlorophyll concentration (SeaWiFS) with radiative transfer calculations for an ocean-atmosphere system. The new features are an extension of underwater radiative transfer (scattering and absorption) into the UV, inclusion of polarization in the above water diffuse radiances, the proper treatment of Fresnel reflection, and first order atmospheric backscatter of water-leaving radiance to the oceans. Maps of downwelling diffuse irradiances (Ed) at ocean surface and at different depths in the ocean, diffuse attenuation coefficient (Kd), and ten percent penetration depth (Z10) of solar irradiation are computed for open ocean waters. Results on spectral irradiances at 310 nm in UV-B and at 380 nm in UV-A part of the spectrum are presented with particular emphasis on the role of aerosols, clouds, and ozone in the atmosphere and chlorophyll concentrations in the ocean.

Measuring aerosol UV absorption optical thickness by combining use of shadowband and almucantar techniques

We report final results of an aerosol UV absorption closure experiment where a UV-shadow-band radiometer (UV-MFRSR, USDA UVB Monitoring and Research Network) and 4 rotating sun-sky radiometers (CIMEL, NASA AERONET network) were run side-by-side continuously for 17 months at NASA/GSFC site in Greenbelt, MD. The aerosol extinction optical thickness τext, was measured by the CIMEL direct-sun technique in the visible and at two UV wavelengths 340 and 380 nm. These results were used for UV-MFRSR daily on-site calibration and 3-min measurements of τext at 325nm, 332nm and 368nm. The τext measurements were used as input to the radiative transfer model along with AERONET retrievals of the column-integrated particle size distribution (PSD)to infer an effective imaginary part of the UV aerosol refractive index, k, by fitting MFRSR measured voltage ratios. Using all cases for cloud-free days, we derive diurnal and seasonal dependence of the aerosol absorption optical thickness, τabs with an uncertainty 0.01-0.02. At our site τabs follows pronounced seasonal dependence with maximum values ~0.07 at 368nm (~0.15 at 325nm) occurring in summer hazy conditions and <0.02 in winter-fall seasons, when aerosol loadings are small. Inferred values of k allow calculation of the single scattering albedo, ω, in UVA and comparisons with AERONET almucantar ω440 retrievals at 440nm. Overall, ω was slightly lower in UV than in the visible: case average =0.93 compared to =0.95. However, the differences ( ~0.02, rms difference ~0.016) are smaller than uncertainties of both retrievals (δω~0.03). Low values are consistent with higher values for imaginary refractive index, k: ~0.01 compare to ~0.006. However, mean differences in k (~0.004) were only slightly larger than AERONET retrieval uncertainty δk ~0.00327. We also found that ω decreases with decrease in τext, suggesting different aerosol composition in summer and winter months. So far, our results do not allow explaining the causes of apparent larger aerosol absorption in UV. Continuing co-located measurements at GFSC is important to improve the comparison statistics, but conducting aerosol absorption measurements at different sites with varying conditions is also desirable.

Goddard UV aerosol absorption closure experiment (2002-03)

Compared to the visible spectral region very little is known about aerosol absorption in UV. Without such information it is impossible to quantify a cause to the observed discrepancy between modeled and measured UV irradiances and photolysis rates. We report preliminary results of an aerosol closure experiment where a UV-shadow-band radiometer (UVMFRSR, USDA UVB Monitoring and Research Network) and well-calibrated sun-sky radiometer (CIMEL, NASA AERONET network) were run side-by-side for several months at NASA/GSFC site in Greenbelt, MD. The aerosol optical thickness, τ, was measured at 340nm and 380nm by the CIMEL direct-sun technique. These results compared well with independent MFRSR τ measurements at 368nm (using total minus diffuse irradiance technique). Such comparisons provide an independent check of both instrument’s radiometric and MFRSR’s angular calibration and allow precise tracking of the UV filter degradation by repeating the comparisons made at somewhat regular time intervals. The τ measurements were used as input to a radiative transfer model along with AERONET retrievals of the column-integrated particle size distribution (PSD) to infer an effective imaginary part of the UV aerosol refractive index (k). This was done by fitting the MFRSR diffuse fraction measurements to the calculated values for each UV spectral channel. Inferred values of refractive index and PSD allow calculation of the single scattering albedo, ω, in the UV and comparisons with AERONET ω retrievals. The advantage of utilizing diffuse fraction measurements is that radiometric calibration is not needed for the MFRSR since the same detector measures both the total and diffuse flux. The additional advantage is that surface albedo is much smaller in the UV than in the visible spectral range and has much less effect on aerosol measurements.

A cloud algorithm based on the O 2 -O 2 477 nm absorption band featuring an advanced spectral fitting method and the use of surface geometry-dependent Lambertian-equivalent reflectivity

We discuss a new cloud algorithm that retrieves an effective cloud pressure, also known as cloud optical centroid pressure (OCP), from oxygen dimer (O2-O2) absorption at 477 nm after determining an effective cloud fraction (ECF) at 466 nm, a wavelength not significantly affected by trace-gas absorption and rotational Raman scattering. The retrieved cloud products are intended for use as inputs to the operational nitrogen dioxide (NO2) retrieval algorithm for the Ozone Monitoring Instrument (OMI) flying on the Aura satellite. The cloud algorithm uses temperature-dependent O2-O2 cross sections and incorporates flexible spectral fitting techniques that account for specifics of the surface reflectivity. The fitting procedure derives O2-O2 slant column densities (SCDs) from radiances after O3, NO2, and H2O absorption features have been removed based on estimates of the amounts of these species from independent OMI algorithms. The cloud algorithm is based on the frequently used mixed Lambertian-equivalent reflectivity (MLER) concept. A geometry-dependent Lambertian-equivalent reflectivity (GLER), which is a proxy of surface bidirectional reflectance, is used for the ground reflectivity in our implementation of the MLER approach. The OCP is derived from a match of the measured O2-O2 SCD to that calculated with the MLER method. Temperature profiles needed for computation of vertical column densities are taken from the Global Modeling Initiative (GMI) model. We investigate the effect of using GLER instead of climatological LER on the retrieved ECF and OCP. For evaluation purposes, the retrieved ECFs and OCPs are compared with those from the operational OMI cloud product, which is also based on the same O2-O2 absorption band. Impacts of the application of the newly developed cloud algorithm to the OMI NO2 retrieval are discussed.

Using Rotational Raman Scattering in the Atmosphere for Satellite Retrieval of Aerosol Properties

Raman scattering is used for retrieval of aerosol properties from satellite hyperspectral measurements in UV. Comparisons of retrieved aerosol heights and single scattering albedo with CALIOP and OMI data show reasonable agreement.

A model for the assessment of UV penetration into ocean waters from space-based measurements and full radiative transfer calculations

Increased levels of biologically harmful Uv radiatonhave beenshown to affec aquatic ecosystems, marine photocynmetiry, and their imapct on carbon cycling. A quantiative assessment of UV effectw requires an estimate of the in-water raiationfield. An esitmate of underwater UV radiatonis porosed based on satellit meausrments fromthe TOMS and SeaWiFS and modesl fo radiatve transfer (RT). The Hydrolight code, modified toe xtnd it to the 290 - 400 nm wavleength range, is used for REt calucaitons in theocean. Solar direc tandidffuse radiances at the ocean surfce are calculated using a fulll RT code for clear-sky coditions, whicha re then modified for clouds and aerosols.Teh TOMS total column ozone and reflectivity productsa reinputs for RT calcuaitons in the atmosphere. An essential component of the in-water RT model is a model of seawater inherent optical properties (IOP). The IOP model is an extension of the Case-1 water model to the UV spectral region. Pure water absorption is interpolated between experimental datasets available in the literature. A new element of the IOP model is parameterization of particulate matter absorption in the UV based on recent in situ data. The SeaWiFS chlorophyll product is input for the IOP model. The in-water computational scheme is verified by comparing the calculated diffuse attenuation coefficient Kd, with one measured for a variety of seawater IOP. The calculated Kd is in a good agreement with the measured Kd. The relative RMS error for all of the cruise stations is about 20%. The error may be partially attributed to variability of solar illumination conditions not accounted for in calculations. The conclusion is that we are now able to model ocean UV irradiances and IOP properties with accuracies approaching those visible region, and in agreement with experimental in situ data.