51ÁÔÆæÈë¿Ú

Abstract:

A new optimal estimation algorithm for the retrieval of sulphur dioxide (SO2) has been developed for the Infrared Atmospheric Sounding Interferometer (IASI) using the channels between 1000-1200 and 1300-1410 cm -1. These regions include the two SO2 absorption bands centred at about 8.7 and 7.3 μm (the v1 and v3 bands respectively). The retrieval assumes a Gaussian SO2 profile and returns the SO2 column amount in Dobson units and the altitude of the plume in millibars (mb). Forward modelled spectra (against which the measurements are compared) are based on the Radiative Transfer for TOVS (RTTOV) code. In our implementation RTTOV uses atmospheric profiles from European Centre for Medium-Range Weather Forecasts (ECMWF) meteorological data. The retrieval includes a comprehensive error budget for every pixel derived from an error covariance matrix that is based on the SO2-free climatology of the differences between the IASI and forward modelled spectra. The IASI forward model includes the ability to simulate a cloud or ash layer in the atmosphere. This feature is used to illustrate that: (1) the SO2 retrieval is not affected by underlying cloud but is affected if the SO2 is within or below a cloud layer; (2) it is possible to discern if ash (or other atmospheric constituents not considered in the error covariance matrix) affects the retrieval using quality control based on the fit of the measured spectrum by the forward modelled spectrum. In this work, the algorithm is applied to follow the behaviour of SO2 plumes from the Eyjafjallajökull eruption during April and May 2010. From 14 April to 4 May (during Phase I and II of the eruption) the total amount of SO2 present in the atmosphere, estimated by IASI measurements, is generally below 0.02 Tg. During the last part of the eruption (Phase III) the values are an order of magnitude higher, with a maximum of 0.18 Tg measured on the afternoon of 7 May. © 2012 Author(s).

Abstract:

The overlap function of a Raman channel for a lidar system is retrieved by non-linear regression using an analytic description of the optical system and a simple model for the extinction profile, constrained by aerosol optical thickness. Considering simulated data, the scheme is successful even where the aerosol profile deviates significantly from the simple model assumed. Application to real data is found to reduce by a factor of 1.4 – 2.0 the root-mean-square difference between the attenuated backscatter coefficient as measured by the calibrated instrument and a commercial instrument.