One feature of climate change evident in IPCC data is that an increase in CO2 leads to a poleward shift in the tropospheric jet. Earlier studies have suggested that this poleward shift in the tropospheric jet may be forced by the rise in tropopause height that occurs as the tropopause warms and the stratosphere cools. In addition, this response seems to be dominated by shifts in tropopause height near the poles (Lorenz and DeWeaver 2007). The first question we attempt to address is whether or not we can use data assimilation to gain insight into this theory. The basis for using data assimilation is that a common bias that occurs in climate models is unobserved cold temperatures near the tropopause in polar regions. By assimilating data in this region, the region warms and effectively lowers the polar tropopause. As expected, the response is opposite to that of GHG forcing. Through this process, we find that data assimilation is a useful tool for identifying, quantifying and analyzing model bias and dynamical relationships.

In addition, the Arctic has been identified as being highly sensitive to shifts in global climate. This sensitivity is generally associated with feedbacks involving sea ice. However, there are substantial uncertainties in modeling the Arctic because it is poorly observed and poorly simulated (IPCC chapter 8 2007). Errors in the sea ice distribution are generally attributed to errors in surface winds, which make an accurate simulation of atmospheric circulation in the Arctic imperative to climate studies. In this study we use an ensemble adjustment Kalman filter (EAKF) provided by the Data Assimilation Research Testbed (DART) to assimilate COSMIC data into the Community Atmosphere Model (CAM) in order to investigate the impact of radio occultation (RO) data assimilation in polar regions. By comparing CAM runs that assimilate RO data with reanalysis products relevant to the Arctic such as tropospheric temperature and surface winds, we will begin to quantify the impact of assimilating COSMIC data into the CAM model. Further, COSMIC data is a relatively new source of observations, and this investigation allows us to better understand the impact that the integration of COSMIC data can have on EAKF data assimilation.