School of Earth and Environment

Chemical Data Assimilation

Martyn Chipperfield, Chris Wilson, Nigel Richards, Lara Gunn

Background

The chemical data assimilation work at Leeds is part of the NERC National Centre for Earth Observation (NCEO), formerly part of the Data Assimilation Research Centre (DARC).

The technique of data assimilation is used routinely in numerical weather prediction to create meteorological analyses. Over the past 10 years or so, there has been increasing interest in applying similar techniques to observations of chemical species in the atmosphere. The assimilation of such observations, and the creation of `chemical analyses' is expected to lead to better use of observations and to improvements in chemical models. The methods used for the assimilation of chemical observations can be divided into variational and sequential.

We have included a sequential chemical data assimilation in the SLIMCAT CTM. A novel feature is the ability to assimilate many species simultaneously and preserve tracer correlations. Full details are given in Chipperfield et al. (2002).

<b>Figure 1:</b> Tracer-tracer correlations as calculated by the basic SLIMCAT model. These compact correlations are well-observed features in the middle atmosphere and models should be able to reproduce them.

Sequential Data Assimilation

 

Preservation of Correlations

Compact correlations are observed in the stratosphere between long-lived species. The correlations exist for all long-lived tracers - not just those which are chemically related. A useful test of a model is its ability to reproduce these correlations.

<b>Figure 2:</b> Assimilating HALOE CH<sub>4</sub> on the CH<sub>4</sub>:N<sub>2</sub>O correlation.

By simply assimilating the tracer CH4, but no other species, although the modelled distribution of CH4 may be improved the compact correlations seen above will be destroyed. Figure 2 shows the impact of assimilating HALOE CH4 on the CH4:N2O correlation. This correlation is not at all good compared with observations, and overall the model has been degraded.

<b>Figure 3 :</b> CH<sub>4</sub> assimilation.

Therefore, in the sequential chemical data assimilation scheme in SLIMCAT we preserve the model-predicted correlations between long-lived species. We also preserve total family abundances (e.g. total Cly).

Figure 3 shows that when only CH4 is assimilated this scheme does do a good job of preserving the correlations of other long-lived tracers.

 

Example Results

Figure 4 below shown zonal mean fields of CH4, H2O and inorganic chlorine (Cly) on January 31, 1991 from a simulation with the SLIMCAT CTM. The model was forced by UKMO winds and had a horizontal resolution of 7.5o x 7.5o.

 

The top row shows fields from the basic model. The bottom row shows results from a simulation that assimilated HALOE observations of CH4 and H2O. For the assimilated species CH4, the modelled distribution now has stronger gradients in the subtropics - which is in better agreement with the direct observations. The preservation of the model correlations (see above) has transferred this information to the long-lived Cly family.

Search site