JOURNAL OF GEOPHYSICAL RESEARCH, VOL. ???, XXXX, DOI:10.1029/, Assessment of inter-calibration methods for satellite microwave humidity sounders Viju O. John 1 , Richard P. Allan 2 , William Bell 3 , Stefan A. Buehler 4 , Ajil Kottayil 4 Abstract. Three methods for inter-calibrating humidity sounding channels are com- pared to assess their merits and demerits. The methods use: 1) Natural targets or vi- carious calibration (Antarctica and tropical oceans), 2) zonal average brightness temper- atures, and 3) simultaneous nadir overpasses (SNOs). Advanced Microwave Sounding Unit- B (AMSU-B) instruments on-board the polar-orbiting NOAA-15 and NOAA-16 satel- lites are used as examples. Antarctica is shown to be useful for identifying some of the instrument problems, but less promising for inter-calibrating humidity sounders due to the large diurnal variations there. Owing to smaller diurnal cycles over tropical oceans, these are found to be a good target for estimating inter-satellite biases. Estimated bi- ases are more resistant to diurnal differences when data from ascending and descend- ing passes are combined. Biases estimated from zonal averaged brightness temperatures show large seasonal and latitude dependence which could have resulted from diurnal cy- cle aliasing and scene-radiance dependence of the biases. This method may not be the best for channels with significant surface contributions. We have also tested the impact of clouds on the estimated biases and found that it is not significant, at least for trop- ical ocean estimates. Biases estimated from SNOs are free of diurnal cycle aliasing and cloud impacts. 1. Introduction Water vapor in the troposphere (especially in the mid to upper troposphere) is an important climate variable due to its positive feedback in a warming climate [e.g., Man- abe and Wetherald , 1967; Held and Soden , 2000]. Despite this, tropospheric water vapour is poorly simulated by cur- rent climate models [e.g., John and Soden , 2007]. Tradi- tional water vapour measurements from radiosondes fail to provide an accurate and global picture of the distribution and evolution of water vapour in the mid to upper tropo- sphere due to their inabilities to measure accurately in drier and colder conditions [e.g., Soden and Lanzante , 1996; John and Buehler , 2005]. Better sources of water vapour mea- surements with global coverage are satellite infrared (IR) and microwave (MW) measurements. The IR measurements have been used to understand the role of water vapour in the climate system [e.g., Soden et al., 2005; Shi and Bates , 2011]. But there is a clear advantage in the microwave cli- mate data record (CDR), which is the availability of all-sky data, whereas infrared records sample only clear areas [John et al., 2011]. Satellite microwave humidity sounders have been mea- suring tropospheric water vapour for about 20 years with the first Special Sensor Microwave Humidity Sounder (SSM- T/2; [Wilheit and al Khalaf , 1994]) in orbit in November 1991. Later on, the first Advanced Microwave Sounding Unit-B (AMSU-B; Saunders et al. [1995]) was launched in May 1998, and the first Microwave Humidity Sounder (MHS; Bonsignori [2007]) was launched in May 2005. Microwave sounding data have been found to make significant impacts 1 Met Office Hadley Centre, Exeter, UK 2 Department of Meteorology, University of Reading, UK 3 Met Office, Exeter, UK 4 Lule˚ a University of Technology, Kiruna, Sweden Copyright 2012 by the American Geophysical Union. 0148-0227/12/$9.00 on the skills of Numerical Weather Prediction (NWP) [e.g., Andersson et al., 2007]. However, SSM-T/2 data were not assimilated at NWP centres and the error characteristics of these measurements are poorly understood. There have been efforts to inter-calibrate microwave hu- midity sounders [e.g., John et al., 2012b] in order to provide climate quality data sets for climate monitoring and use in climate quality reanalyses. Inter-calibration methods which use simultaneous nadir overpasses (SNOs; [Cao et al., 2004]) of polar orbiting satellites have become very popular [Cao et al., 2005; Shi et al., 2008; Iacovazzi and Cao , 2008; Zou et al., 2009]. During an SNO, instruments on two satellites measure the same target area with short time differences, thus the difference in their measurements can practically be taken as inter-satellite bias. SNO estimates are free from some sampling errors, for example, arising from the diur- nal cycle. However, SNOs of polar orbiting satellites nor- mally occur only in polar latitudes (above 70 ◦ ) and thus SNO measurements represent only a small portion of the dy- namic ranges of global measurements [Shi and Bates , 2011]. It is shown in John et al. [2012b] that the scene-radiance dependence of inter-satellite biases limits the usefulness of polar-only SNOs for inter-calibrating microwave humidity sounders. Another possible method for inter-calibration is the use of natural calibration targets such as Antarctica or tropical oceans [Mo , 2011]. Mo [2010] has shown that Antarctica can act as a good inter-calibration target during Antarctic win- ter months because diurnal variations there are very small. Mo and Liu [2008] have shown that tropical oceans can also act as a stable inter-calibration point. Therefore, in this study we analyse microwave humidity sounding data over these two natural targets to understand data characteristics and to estimate inter-satellite biases. The establishment of a set of natural Earth targets for instrument calibration is important for the calibration and validation of microwave radiometers [Mo , 2011]. Shi and Bates [2011] proposed and used another method for inter-calibrating upper tropospheric water vapour in- frared radiances measured by different HIRS instruments. They used zonally averaged (10 ◦ bins) monthly mean 1