EFFECT OF MICROWAVE RADIOMETER INTER-CALIBRATION ON RAINFALL ACCUMULATION FOR THE GLOBAL PRECIPITATION MEASUREMENT MISSION Rachael Kroodsma, Darren McKague, and Christopher Ruf University of Michigan, USA ABSTRACT The effect of inter-calibration on a Level 3 rainfall product for the Global Precipitation Measurement (GPM) mission is examined using two spaceborne microwave radiometers that are currently used to derive rain measurements, the Tropical Rainfall Measuring Mission Microwave Imager (TMI) and the Special Sensor Microwave/Imager (SSM/I). It is found that inter-calibrating the microwave radiometer brightness temperatures from the two instruments improves the agreement of the derived rain accumulations between the two radiometers. The average difference between TMI and F13 derived rain accumulations is 0.60 mm/day before inter- calibration is applied. This difference decreases to 0.08 mm/day when F13 is inter-calibrated to TMI. Index Terms—Microwave radiometry, Inter-calibration, Rainfall retrieval 1. INTRODUCTION Spaceborne microwave radiometers are an important tool used to measure global precipitation. Since the instruments are onboard satellites, they are able to achieve measurements of precipitation over otherwise hard to reach areas compared to in situ instruments or ground radar (e.g. oceans). It is therefore imperative that the microwave radiometers be properly calibrated so that the precipitation derived from the radiometer measurements is accurate. Furthermore, if several radiometers are to be used to derive precipitation, such as in the Global Precipitation Measurement (GPM) mission, the radiometers need to be inter-calibrated to ensure that the precipitation derived from the various instruments is consistent [1]. The objective of the GPM X-Cal team is to develop algorithms to inter-calibrate the microwave radiometers used in the constellation for GPM [2]. These algorithms have been successfully applied to several current spaceborne microwave radiometers to calculate calibration differences between the radiometers, which are then applied to the brightness temperature (TB) measurements of the radiometers to make the TB measurements between different radiometers consistent with each other. Inter-calibration adjustments are based on the TBs and not the derived precipitation products in order to minimize their dependence on any model assumptions that may be incorporated into the precipitation retrieval algorithms. It is not well understood what, if any, effect the inter- calibration has on the derived precipitation products for GPM. Since the inter-calibration of the TBs is done to make the measurements, and thereby the derived products, more consistent with each other, this is an important analysis to be done. This paper will examine the effect of microwave radiometer inter-calibration on one of the precipitation products that will be produced for GPM. 2. RAIN RETRIEVAL ALGORITHM The predecessor for the GPM mission, the Tropical Rainfall Measuring Mission (TRMM), is currently on orbit and has several rain products that are derived using the Precipitation Radar (PR) and the TRMM Microwave Imager (TMI) onboard the spacecraft. Since we are concerned here with the effect of the inter-calibration of the microwave radiometers on the rain retrievals, we will examine the products that only make use of the radiometer data. One of these is the Level 3 3A11 product that gives monthly rainfall accumulations. This will be used as an example here. The 3A11 product operates directly on the Level 1 microwave radiometer TBs to produce monthly rainfall accumulations over the ocean for 5° latitude/longitude gridded regions [3]. This retrieval algorithm will hereafter be referred to as the WCC algorithm. The WCC algorithm was developed for the Special Sensor Microwave/Imager (SSM/I), but has since been adapted for TMI. It makes use of the 19 GHz and 22 GHz vertically polarized (V-pol) channels to derive the rain accumulations. Monthly histograms of the radiometer TBs are generated for these two channels for each 5° grid box, and the freezing level is calculated using a radiative transfer model. This freezing level is then used with the histogram of a combined channel, 2*TB 19V – TB 22V , to derive the monthly rain accumulations in mm/day. A full description of the WCC algorithm is given by Wilheit et al. [3]. The goal of this study is to quantify the effect that inter- calibrating the microwave radiometers has on the Level 3 rainfall accumulations derived using the WCC retrieval algorithm. A comparison is made between the derived rain 1147 978-1-4799-1114-1/13/$31.00 ©2013 IEEE IGARSS 2013