 |
| Titel |
Four dimensional variational assimilation of in-situ and remote-sensing aerosol data |
| VerfasserIn |
L. P. Nieradzik, H. Elbern |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250059245
|
|
|
|
|
|
| Zusammenfassung |
| Aerosols play an increasingly important role in atmospheric modelling. They have a strong
influence on the radiative transfer balance and a significant impact on human health. Their
origin is various and so are its effects. Most of the measurement sites in Europe account for
an integrated aerosol load PMx (Particulate Matter of less than x μm in diameter) which does
not give any qualitative information on the composition of the aerosol. Since very different
constituents contribute to PMx, like e.g. mineral dust derived from desert storms or
sea salt, it is necessary to make aerosol forecasts not only of load, but also type
resolved.
The method of four dimensional variational data assimilation (4Dvar) is a widely known
technique to enhance forecast skills of CTMs (Chemistry-Transport-Models) by ingesting
in-situ and, especially, remote-sensing measurements. The EURAD-IM (EURopean Air
pollution Dispersion - Inverse Model), containing a full adjoint gas-phase model, has
been expanded with an adjoint of the MADE (Modal Aerosol Dynamics model for
Europe) to optimise initial and boundary values for aerosols using 4Dvar. A forward
and an adjoint radiative transfer model is driven by the EURAD-IM as mapping
between BLAOT (Boundary Layer Aerosol Optical Thickness) and internal aerosol
species. Furthermore, its condensation scheme has been bypassed by an HDMR
(High-Dimensional-Model-Representation) to ensure differentiability.
In this study both in-situ measured PMx as well as satellite retrieved aerosol
optical thicknesses have been assimilated and the effect on forecast performance has
been investigated. The source of BLAOT is the aerosol retrieval system SYNAER
(SYNergetic AErosol Retrieval) from DLR-DFD that retrieves AOT by making use of both
AATSR/SCIAMACHY and AVHRR/GOME-2 data respectively. Its strengths are a large
spatial coverage, near real-time availability, and the classification of five intrinsic aerosol
species, namely water-solubles, water-insolubles, soot, sea salt, and mineral dust which are
furthermore size resolved in terms of modes.
The skill of the aerosol 4Dvar system was tested in two episodes: 1) July through August
2003, a dry period with strong wildfire activity in Europe, and 2) October through November
2008, the period of the ZEPTER-2 (Second ZEPpelin based Tropospheric photochemical
chemistry expERiment) measurement campaign in the area of Lake Constance. In the latter
case one-way nesting has been applied from a horizontal grid resolution of 45 km down to 5
km. Overall, the results showed a significant increase in forecast quality of tropospheric
aerosol loads. |
|
|
|
|
|
|