 |
| Titel |
Uncertainty characterization of tracer vertical mixing in the atmospheric transport model WRF-VPRM |
| VerfasserIn |
Roberto Kretschmer, Christoph Gerbig, Frank-Thomas Koch |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250053028
|
|
|
|
|
|
| Zusammenfassung |
| Large uncertainty of air quality modelling lies in the representation of vertical mixing of
tracers in the planetary boundary layer (PBL). For the atmospheric transport of greenhouse
gases (GHG) such as carbon dioxide (CO2) studies have shown that imperfect representation
of vertical mixing in the transport model can lead to substantial bias of the simulated CO2
distribution which makes the interpretation of data difficult. One of the current objectives in
carbon cycle science is to relate carbon source/sink processes to external forcing on
sub-continental scales. This so-called top-down approach tries to inversely link atmospheric
CO2/CH4 mixing ratio measurements to sources and sinks in order to infer flux
strengths by using atmospheric transport models. To make predictions by the top-down
method more reliable model errors in the simulation of PBL mixing have to be
quantified.
In the current study we applied a transport model that is widely used for air pollution
modelling studies: WRF-Chem which we apply as part of the WRF-VPRM modelling
framework. The model was set up with a horizontal resolution of 10 km and 41 vertical
levels to simulate CO2 transport over large parts of the European continent for a one
month period in summer 2006. Two simulations were conducted employing different
parameterizations of the PBL: the Mellor-Yamada-Janjic (MYJ) and the Yonsei University
(YSU) scheme. To asses the uncertainty in the vertical mixing we compared the
mixed layer height (zi) up to which CO2 gets mixed within several minutes up to an
hour. The zi is an important variable in air quality modelling because it directly
determines the integration volume for pollutants emitted from the ground. Since
photosynthetic uptake of CO2 and PBL growth are correlated a wrongly simulated
PBL height might also give rise to additional bias due to rectifier effects. In our
experiment zi differs significantly for both schemes with a bias of 200-500 m for day and
nighttime. We also present comparison results for both schemes to observed zi, such as
radio sounding measurements. To be able to relate zi errors to errors in CO2 we
developed a method to correct modeled CO2 fields by using information on the true
height of the mixed layer. We will show the results of a pseudo-data experiment in
which we defined YSU as known truth and corrected MYJ simulated CO2 offline. |
|
|
|
|
|