Table 3. chi2/n Statistics with Respect to Surface and SCIAMACHY Observations for Selected Regions (Figure 1) and for Inversion Scenarios I1 to I11 Described in Table 1a

		I1	I2	I3	I4	I5	I6	I7	I8	I9	I10	I11
Surface Globe	n 1	400	400	400	400	400	400	400	400	400	385	385
	prior 1	7.54	7.54	5.12	9.29	7.54	7.54	7.54	7.54	6.97	9.59	9.59
	post. 1	1.99	1.82	2.21	1.95	1.91	2.16	2.05	2.09	1.76	1.60	1.17
	nout 1	19	10	21	17	15	21	18	23	13	12	7
	post. 2	1.17	1.20	1.28	1.19	1.17	1.24	1.18	1.16	1.18	1.17	0.96
SCIA Globe	n 1	98190		98190	98190	98190	98186	98190	98190	98190	130719	
	prior 1	0.96		0.95	1.05	0.90	1.52	0.97	0.97	0.97	0.99	
	post. 1	0.63		0.63	0.63	0.62	0.97	0.62	0.64	0.62	0.62	
	nout 1	237		230	237	226	866	227	238	226	322	
	post. 2	0.60		0.60	0.60	0.60	0.86	0.59	0.61	0.59	0.59	
SCIA Sahara	n 1	2308		2308	2308	2308	2308	2308	2308	2308	3448	
	prior 1	0.22		0.22	0.20	0.21	0.33	0.20	0.22	0.21	0.17	
	post. 1	0.19		0.19	0.19	0.19	0.29	0.19	0.19	0.19	0.15	
	nout 1	0		0	0	0	0	0	0	0	1	
	post. 2	0.19		0.19	0.19	0.19	0.29	0.19	0.19	0.19	0.14	
SCIA Tropical South America	n 1	4256		4256	4256	4256	4256	4256	4256	4256	6611	
	prior 1	2.13		2.02	2.89	1.84	3.44	2.23	2.15	2.25	1.94	
	post. 1	0.72		0.71	0.74	0.71	1.08	0.71	0.72	0.70	0.75	
	nout 1	7		4	8	7	38	7	7	5	16	
	post. 2	0.70		0.69	0.71	0.69	0.98	0.68	0.70	0.68	0.68	
SCIA tropical Africa	n 1	9557		9557	9557	9557	9557	9557	9557	9557	12106	
	prior 1	1.06		1.03	1.23	0.96	1.68	1.09	1.06	1.09	1.21	
	post. 1	0.61		0.60	0.62	0.59	0.91	0.57	0.62	0.62	0.64	
	nout 1	11		11	11	10	47	10	11	11	17	
	post. 2	0.60		0.58	0.60	0.58	0.85	0.56	0.60	0.60	0.62	
SCIA South Asia	n 1	2647		2647	2647	2647	2647	2647	2647	2647	2820	
	prior 1	2.68		2.76	2.60	2.57	4.14	2.59	2.68	2.65	2.77	
	post. 1	0.89		0.89	0.89	0.90	1.37	0.91	0.89	0.89	0.95	
	nout 1	9		9	10	10	50	9	9	9	24	
	post. 2	0.85		0.85	0.85	0.85	1.17	0.86	0.85	0.85	0.86	

a'n 1' and 'nout 1' are the number of assimilated observations and the number of posterior outlier observations (posterior model simulation and observation differ more than three times the observation error) in the first inversion cycle, respectively. 'prior 1' and 'post. 1' are the a priori and a posteriori chi2/n in the first inversion cycle, respectively. 'post. 2' is the a posteriori chi2/n in the second cycle. For the calculation of all chi2/n with respect to SCIAMACHY, the bias-corrected observations are used.