 |
| Titel |
Simultaneous Retrieval of Atmospheric CO2 and Effective Aerosol Properties From Simulated Space Borne Observations of Backscattered Near-Infrared Sunlight |
| VerfasserIn |
A. Butz, O. P. Hasekamp, C. Frankenberg, I. Aben |
| Konferenz |
EGU General Assembly 2009
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 11 (2009) |
| Datensatznummer |
250025241
|
|
|
|
|
|
| Zusammenfassung |
| Forthcoming satellite missions such as the Orbiting Carbon Observatory (OCO) and the
Greenhouse Gases Observing SATellite (GOSAT) target at inferring atmospheric CO2
abundances with high accuracy and global coverage. These missions will measure spectra of
near-infrared sunlight backscattered by the Earth‘s surface and atmosphere covering two
near-infrared CO2 absorption bands and the O2 A band. For CO2 retrievals from such
measurements, light path modification due to aerosol and cirrus cloud scattering
has been identified as a major source of error if not appropriately accounted for
in the retrieval scheme. Neglecting aerosol and cirrus effects in the retrieval can
cause errors of several percent in retrieved CO2 strongly depending on particle
amount, size, type and height distribution, ground surface properties, and viewing
geometry.
Here, we present a retrieval method that simultaneously infers aerosol properties and total
column CO2 from OCO-type nadir observations over land surfaces. The goal of our approach
is to account for the effect of aerosols and cirrus clouds on retrieved CO2. The core of the
retrieval scheme is a vector radiative transfer model that simulates the radiance measured by a
space borne observer given input aerosol and molecular optical properties. We propose a
retrieval method that infers a few effective aerosol properties of a simplified aerosols
microphysical model. A Levenberg-Marquardt least-squares method is setup to test
performance of our approximate aerosol forward model for an ensemble of simulated
spectra.
The trial ensemble covers a wide range of aerosol types, sizes and height distributions as
well as land surface scenes. Further, we consider cirrus clouds overlaying the background
aerosol. Trial spectra are generated using multi-parameter aerosol and cirrus cloud models
that allow for a detailed representation of aerosol and cirrus effects on spectra recorded by
an OCO-like observer. Here, we aim at quantifying the residual forward model
bias for retrieved CO2 when applying the approximate retrieval model to the trial
ensemble.
Preliminary results indicate that the proposed retrieval method reduces aerosol and cirrus
induced CO2 errors to mostly below 1 per cent for aerosol and cirrus optical thickness up to
0.3 and 0.1, respectively. Retrieval performance is best for aerosol-only scenes at moderate
solar zenith angle (SZA) largely independent of ground surface and aerosol type. Cases
including aerosol and cirrus particles are more challenging than aerosol-only scenes. In
general, retrieval performance is better for moderate than for high solar zenith angles. |
|
|
|
|
|
|