Table 1. Summary of the Linear Error Analysis of the SCIAMACHY Retrievala

	Day and Year	ILS	Abs Gain	Rel Gain	Offset	T	Aerosol	Pol
		Ch4	Ch6		Ch4	Ch6	Ch4	Ch6		tau	Type	hscale	
Deltab		5%	5%	5%	1%	1%	1%	1%	2K	100%	-	+1km (+3km)	-
XCO2, ppm	194/04	0.4	0.4	0.3	0.07	0.01	4.3	2.4	0.08	0.1	0.35 (0.2)	0.2 (0.3)	0.08
	284/04	1.0	0.6	0.8	0.2	0.03	6.2	3.3	0.09	0.06	0.08 (0.2)	0.4 (1.7)	1.1
	104/05	1.1	0.4	1.0	0.2	0.02	6.4	3.3	0.07	0.5	0.7 (0.03)	1.8 (5.8)	1.8
psurf, hPa	194/04	1.4	0.	1.4	0.3	0.	16.5	0.	0.9	0.9	1.5 (0.4)	1.1 (0.6)	0.1
	284/04	3.0	0.	3.0	0.6	0.	19.1	0.	0.8	0.8	0.2 (0.1)	4.2 (14.6)	4.0
	104/05	3.2	0.	3.2	0.6	0.	18.9	0.	0.9	1.8	2.0 (0.5)	7.3 (23.6)	6.3
aErrors in XCO2 and surface pressure due to uncertainties in instrument calibration (ILS, absolute and relative radiometric gain and intensity offset), assumed atmospheric temperature, assumed aerosol characteristics, and due to neglecting polarization (Pol) are given. Also given are the assumed uncertainties Deltab of the studied parameters. For ILS, relative gain and intensity offset, errors are given independently for channel 4 (Ch4) and channel 6 (Ch6). To investigate errors due to uncertainties in the atmospheric aerosol characteristics, we have changed the aerosol optical depth (tau), the aerosol type and the scale height (hscale) of the vertical optical depth profile. The error due to aerosol type has been obtained by replacing the aerosol optical properties of the aerosol type used in the retrieval (see section 4.2) with those of water soluble aerosols with a monomodal distribution with a mean radius of 0.004 mum or dust-like aerosols with a bimodal distribution with mean radii of 0.13 and 0.3 mum (numbers in parentheses) [Shettle and Fenn, 1979].