Intercalibration of the GPM Microwave Radiometer Constellation

Wesley Berg Colorado State University, Fort Collins, Colorado

Search for other papers by Wesley Berg in
Current site
Google Scholar
PubMed
Close
,
Stephen Bilanow Wyle Information Systems, McLean, Virginia

Search for other papers by Stephen Bilanow in
Current site
Google Scholar
PubMed
Close
,
Ruiyao Chen University of Central Florida, Orlando, Florida

Search for other papers by Ruiyao Chen in
Current site
Google Scholar
PubMed
Close
,
Saswati Datta Data and Image Processing Consultants, LLC, Morrisville, North Carolina

Search for other papers by Saswati Datta in
Current site
Google Scholar
PubMed
Close
,
David Draper Ball Aerospace and Technologies Corporation, Boulder, Colorado

Search for other papers by David Draper in
Current site
Google Scholar
PubMed
Close
,
Hamideh Ebrahimi University of Central Florida, Orlando, Florida

Search for other papers by Hamideh Ebrahimi in
Current site
Google Scholar
PubMed
Close
,
Spencer Farrar University of Central Florida, Orlando, Florida

Search for other papers by Spencer Farrar in
Current site
Google Scholar
PubMed
Close
,
W. Linwood Jones University of Central Florida, Orlando, Florida

Search for other papers by W. Linwood Jones in
Current site
Google Scholar
PubMed
Close
,
Rachael Kroodsma Earth System Science Interdisciplinary Center, University of Maryland, College Park, College Park, and NASA Goddard Space Flight Center, Greenbelt, Maryland

Search for other papers by Rachael Kroodsma in
Current site
Google Scholar
PubMed
Close
,
Darren McKague University of Michigan, Ann Arbor, Michigan

Search for other papers by Darren McKague in
Current site
Google Scholar
PubMed
Close
,
Vivienne Payne Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California

Search for other papers by Vivienne Payne in
Current site
Google Scholar
PubMed
Close
,
James Wang Science Systems and Applications, Inc., Lanham, Maryland

Search for other papers by James Wang in
Current site
Google Scholar
PubMed
Close
,
Thomas Wilheit Texas A&M University, College Station, Texas

Search for other papers by Thomas Wilheit in
Current site
Google Scholar
PubMed
Close
, and
John Xun Yang University of Michigan, Ann Arbor, Michigan

Search for other papers by John Xun Yang in
Current site
Google Scholar
PubMed
Close
Full access

We are aware of a technical issue preventing figures and tables from showing in some newly published articles in the full-text HTML view.
While we are resolving the problem, please use the online PDF version of these articles to view figures and tables.

Abstract

The Global Precipitation Measurement (GPM) mission is a constellation-based satellite mission designed to unify and advance precipitation measurements using both research and operational microwave sensors. This requires consistency in the input brightness temperatures (Tb), which is accomplished by intercalibrating the constellation radiometers using the GPM Microwave Imager (GMI) as the calibration reference. The first step in intercalibrating the sensors involves prescreening the sensor Tb to identify and correct for calibration biases across the scan or along the orbit path. Next, multiple techniques developed by teams within the GPM Intersatellite Calibration Working Group (XCAL) are used to adjust the calibrations of the constellation radiometers to be consistent with GMI. Comparing results from multiple approaches helps identify flaws or limitations of a given technique, increase confidence in the results, and provide a measure of the residual uncertainty. The original calibration differences relative to GMI are generally within 2–3 K for channels below 92 GHz, although AMSR2 exhibits larger differences that vary with scene temperature. SSMIS calibration differences also vary with scene temperature but to a lesser degree. For SSMIS channels above 150 GHz, the differences are generally within ~2 K with the exception of SSMIS on board DMSP F19, which ranges from 7 to 11 K colder than GMI depending on frequency. The calibrations of the cross-track radiometers agree very well with GMI with values mostly within 0.5 K for the Sondeur Atmosphérique du Profil d’Humidité Intertropicale par Radiométrie (SAPHIR) and the Microwave Humidity Sounder (MHS) sensors, and within 1 K for the Advanced Technology Microwave Sounder (ATMS).

Corresponding author address: Wesley Berg, Dept. of Atmospheric Science, Colorado State University, 1371 Campus Delivery, Fort Collins, CO 80523-1371. E-mail: berg@atmos.colostate.edu

This article is included in the Global Precipitation Measurement (GPM) special collection.

Abstract

The Global Precipitation Measurement (GPM) mission is a constellation-based satellite mission designed to unify and advance precipitation measurements using both research and operational microwave sensors. This requires consistency in the input brightness temperatures (Tb), which is accomplished by intercalibrating the constellation radiometers using the GPM Microwave Imager (GMI) as the calibration reference. The first step in intercalibrating the sensors involves prescreening the sensor Tb to identify and correct for calibration biases across the scan or along the orbit path. Next, multiple techniques developed by teams within the GPM Intersatellite Calibration Working Group (XCAL) are used to adjust the calibrations of the constellation radiometers to be consistent with GMI. Comparing results from multiple approaches helps identify flaws or limitations of a given technique, increase confidence in the results, and provide a measure of the residual uncertainty. The original calibration differences relative to GMI are generally within 2–3 K for channels below 92 GHz, although AMSR2 exhibits larger differences that vary with scene temperature. SSMIS calibration differences also vary with scene temperature but to a lesser degree. For SSMIS channels above 150 GHz, the differences are generally within ~2 K with the exception of SSMIS on board DMSP F19, which ranges from 7 to 11 K colder than GMI depending on frequency. The calibrations of the cross-track radiometers agree very well with GMI with values mostly within 0.5 K for the Sondeur Atmosphérique du Profil d’Humidité Intertropicale par Radiométrie (SAPHIR) and the Microwave Humidity Sounder (MHS) sensors, and within 1 K for the Advanced Technology Microwave Sounder (ATMS).

Corresponding author address: Wesley Berg, Dept. of Atmospheric Science, Colorado State University, 1371 Campus Delivery, Fort Collins, CO 80523-1371. E-mail: berg@atmos.colostate.edu

This article is included in the Global Precipitation Measurement (GPM) special collection.

1. Introduction

The Global Precipitation Measurement (GPM) mission seeks to build upon the success of the Tropical Rainfall Measuring Mission (TRMM) through improved technology, expanded global coverage, and improved temporal sampling from a constellation of research and operational microwave sensors. The GPM Core Observatory provides a number of improvements over TRMM. These include a Dual-Frequency Precipitation Radar (DPR) that adds a Ka-band radar with better sensitivity to light precipitation, the addition of high-frequency channels to the GPM Microwave Imager (GMI) for increased sensitivity to snowfall, and an orbit inclination of 65°, which extends observations into the middle and high latitudes. Beyond the technical improvements to the GPM Core Observatory, however, the GPM mission is a constellation-based satellite mission designed to unify and advance precipitation measurements from a constellation of research and operational microwave sensors in order to improve our understanding of Earth’s water and energy cycles (Hou et al. 2014).

To provide unified estimates of precipitation from microwave radiometers built and launched by different space agencies with widely varying specifications and capabilities requires that the input brightness temperatures (Tb) be physically consistent between sensors. This means that differences in the observed Tb between sensors should agree with the expected differences based on radiative transfer model simulations that account for variations in the observing frequencies, channel bandwidths, view angles, etc. Properly accounting for sensor differences is critical to producing consistent precipitation estimates between radiometers and is the only way to ensure that observed changes in precipitation are real and not the result of sensor calibration issues.

The GPM Intersatellite Calibration Working Group (XCAL team) was established in 2007 as an ad hoc working group within the Precipitation Measurement Missions (PMM) science team. The XCAL team has responsibility for the intercalibrated level 1C Tb files that are used as input for the radiometer retrieval algorithm. Prior to the launch of GPM this involved developing the level 1C format and producing initial calibration tables for the constellation sensors using the TRMM Microwave Imager (TMI) as the reference standard. After the launch of the GPM Core Observatory, the XCAL team initially focused on the GMI calibration to ensure GMI provided the best possible calibration reference for the other microwave radiometers in the GPM constellation. Once the GMI calibration was finalized for the version 4 (V04) reprocessing, the team worked to identify issues affecting the calibration and stability of the constellation radiometers, developed corrections for these issues, and then produced intercalibration tables to adjust for residual sensor calibration differences in a physically consistent manner. XCAL has also served as a general-purpose consultant to the rest of the science team on radiometer technical issues.

The techniques and approaches used by the XCAL team build on a long history of intercalibrating microwave radiometers for consistency in retrieved geophysical parameters as well as for creating long-term data records. Early efforts focused on intercalibrating nearly identical copies of the Special Sensor Microwave Imager (SSM/I) instruments on board the Defense Meteorological Satellite Program (DMSP) spacecraft (Wentz 1997; Colton and Poe 1999). Subsequent intercalibration efforts have focused on the extension of the DMSP data record (Yan and Weng 2008; Yang et al. 2011; Sapiano et al. 2013), incorporating newer and more capable sensors for more robust longer-term data records (Wilheit 2013; Wentz et al. 2001; Wentz 2015), developing improved techniques (Ruf 2000; Brown and Ruf 2005), and addressing calibration differences with cross-track temperature and water vapor sounding radiometers (Zou and Wang 2011; John et al. 2012, 2013; Chung et al. 2013; Moradi et al. 2015). In addition, the Global Space-Based Intercalibration System (GSICS) is focused on intercalibrating a broad range of operational microwave, infrared, and visible satellite instruments (Goldberg et al. 2011). While this represents only a few of the many satellite microwave intercalibration efforts, a more comprehensive summary of published satellite intercalibration techniques is provided by Chander et al. (2013).

2. The GPM radiometer constellation

The GPM Core Observatory was launched from the Tanegashima Space Center in Japan on 28 February 2014. The passive microwave imager on board provides a number of improvements over TMI as well as other spaceborne conically scanning window channel radiometers. These include increased spatial resolution; the addition of high-frequency channels at 166, 183 ± 3, and 183 ± 7 GHz; and an emphasis on calibration accuracy and stability. The design requirements for GMI were driven by requirements both for its use in building the a priori database for the microwave precipitation retrieval algorithm and for providing the reference calibration standard for the GPM radiometer constellation (Hou et al. 2014). The GMI radiometric uncertainty requirements included the radiometric sensitivity, or noise-equivalent delta temperature (NEΔT), the calibration accuracy, and the calibration stability (Draper et al. 2015b). Some of the GMI calibration–related developments include the addition of noise diodes for a four-point calibration of the window channels (Draper et al. 2015a); design considerations to reduce/eliminate calibration issues affecting prior sensors, such as emissive reflectors (Gopalan et al. 2009; Biswas et al. 2010) and solar intrusions into the warm load (Kunkee et al. 2008b); and multiple on-orbit calibration maneuvers. Details of the GMI calibration are provided in section 4.

Figures 1a and 1b show the coverage provided by the conically scanning window channel radiometers and the cross-track scanning microwave sounders in the GPM constellation for a single orbit from each of the sensors for 1 January 2015. Because of the inclination of the orbits and their higher-observing altitude, the radiometers on board polar-orbiting satellites provide the bulk of the global coverage needed to meet the desired 3-hourly global sampling requirements for the GPM constellation (Hou et al. 2014).

Fig. 1.
Fig. 1.

Coverage provided by a single orbit from the GPM microwave constellation for 1 Jan 2015 for (a) conical window channel radiometers and (b) cross-track sounding radiometers.

Citation: Journal of Atmospheric and Oceanic Technology 33, 12; 10.1175/JTECH-D-16-0100.1

Specifications of the available channels for each of the microwave radiometers in the GPM constellation are provided in Table 1. The conical-scanning window channel radiometers include TMI, which was turned off on 8 April 2015; the Advanced Microwave Scanning Radiometer 2 (AMSR2) on board the Global Change Observation Mission–Water (GCOM-W1); and four Special Sensor Microwave Imager/Sounder (SSMIS) instruments on board the DMSP F16F19 spacecraft. The Coriolis WindSat is also a conical-scanning radiometer, but it is not currently part of the operational GPM constellation. This is due to the fact that changes in the sensor data record (SDR) calibration (i.e., level 1B) are only applied going forward and not reprocessed for a consistent data record. In addition, cross-track sounding radiometers with channels near the 183-GHz water vapor line are used for GPM precipitation retrievals. These currently include the Advanced Technology Microwave Sounder (ATMS) on board the Suomi–National Polar-Orbiting Partnership (Suomi-NPP) spacecraft; the Microwave Humidity Sounder (MHS) on board MetOp-A, MetOp-B, NOAA-18, and NOAA-19; and the Sondeur Atmosphérique du Profil d’Humidité Intertropicale par Radiométrie (SAPHIR) on board Megha-Tropiques. Differences in channel availability and characteristics between sensors lead to significant challenges in ensuring consistency in both the input Tb and resulting precipitation estimates.

Table 1.

Frequency and polarization (v = vertical, h = horizontal, qv = quasi vertical, qh = quasi horizontal) of channels for GPM radiometer constellation. Note that the AMSR2 instrument has two independent 89-GHz feedhorns referred to as A-scan (A) and B-Scan (B).

Table 1.

3. Methodology

The XCAL team’s approach to sensor intercalibration involves several steps. The first step involves a prescreening process in which calibration biases across the scan or along the orbit path are removed (Wilheit 2013). Examples include removing cross-track biases (Yang and McKague 2016) accounting for an emissive reflector (Gopalan et al. 2009; Biswas et al. 2010) and solar intrusions and/or thermal gradients in the hot load (Berg and Sapiano 2013; Kunkee et al. 2008a,b; Bell et al. 2008; Imaoka et al. 2003). Geolocation analysis is also used to identify and correct for errors in the Earth incidence angle (EIA) resulting from mounting offsets in the feed horns and/or spacecraft attitude errors (Berg et al. 2013; Moradi et al. 2013). Although generally small, these errors can have a significant impact on the simulated Tb values and thus the resulting estimates of calibration differences. Once corrections are applied, a variety of techniques developed by teams within the XCAL working group are used to compare the calibrations of the constellation radiometers to the calibration reference sensor. The goal is to adjust the sensor calibration of the constellation radiometers to be physically consistent with the reference radiometer (i.e., GMI). To do this, the intercalibration techniques compare channels at similar frequencies, accounting for expected differences in viewing parameters, frequency, bandwidth, polarization, and view angles using radiative transfer models. Comparing results from multiple independent approaches helps to identify flaws or limitations of a given approach and/or errors in the implementation. Consistency between approaches also increases confidence that the resulting Tb differences are due to calibration issues and provides a measure of the uncertainty. For the window channels where the atmosphere is semitransparent, the XCAL team produces a two-point calibration adjustment based on oceanic observations (cold scene) and high-emissivity vegetated land scenes, such as the Amazon basin (warm scene). For the water vapor sounding channels with minimal or no sensitivity to the surface, a one-point calibration adjustment, or Tb offset, is computed. Table 2 gives a summary of the teams contributing results for the combined XCAL calibration of the imagers over cold ocean scenes and warm land scenes, as well as the sounders.

Table 2.

Contributing teams within the XCAL working group.

Table 2.

a. Double differences over oceans (window channel cold scenes)

Several of the intercalibration approaches employed by teams within the XCAL working group are variations on a technique involving double differences between sensors. The double-difference approach employs coincident observations between sensors for nonprecipitating scenes and subtracts out the expected sensor differences due to variations in channel frequency, bandwidth, view angle, etc., based on simulated Tb. As shown in Fig. 1, the orbits of the TRMM satellite (35° inclination) and the GPM satellite (65° inclination) result in regular daily coincident overpasses with all of the sun-synchronous polar-orbiting spacecraft. Coincident overpasses that occur within a specified time window (typically 30–60 min) are identified between the reference sensor (GMI) and the target sensor (e.g., F16 SSMIS). The Tb data within the crossover region are typically gridded and screened for land, the presence of precipitation, sun glint, data quality issues, etc. The black lines in Fig. 2 indicate 1° × 1° grid boxes screened to eliminate partial coverage by either sensor or land contamination. The resulting grid-averaged Tb values for each sensor are then matched with corresponding atmospheric and ocean surface parameters either from global model analyses or retrieved using the GMI Tb as input (Elsaesser and Kummerow 2008; Wilheit et al. 2015). Over oceans, these parameters are used as input into an ocean surface emissivity model (Meissner and Wentz 2012) and an atmospheric absorption model (Rosenkranz 1998; Kummerow 1993). Simulated Tb data are then computed for both the reference and target sensors using the specified channel frequencies, band widths, polarizations, and view angles. The resulting observed and simulated Tb differences are shown in Fig. 3 for both cold scenes (ocean) and warm scenes (vegetated land) for the TMI 19.35-GHz channels versus the 18.7-GHz GMI channels. As shown in Fig. 3, for ocean scenes the frequency difference between TMI and GMI for these channels results in significant differences in the simulated or expected Tb with increasing scene temperature. However, the observed minus the simulated differences, or double differences, are relatively constant with scene temperature, thus indicating a small difference between the TMI and GMI calibrations for these channels.

Fig. 2.
Fig. 2.

A coincident overpass between GMI (red dots) and DMSP F16 SSMIS (blue dots) from 20 Jan 2015. The Tb values are averaged for each sensor over 1° × 1° grid boxes as indicated by the black squares, and then screened for land, precipitation, sun glint, erroneous data, etc.

Citation: Journal of Atmospheric and Oceanic Technology 33, 12; 10.1175/JTECH-D-16-0100.1

Fig. 3.
Fig. 3.

Differences in observed, simulated, and double differences as a function of scene temperature for the TMI 19.35-GHz (a) V-pol and (b) H-pol channels. The values shown at scene temperatures below 220 K are over ocean, while the values above 280 K are over nonpolarized highly vegetated land surfaces.

Citation: Journal of Atmospheric and Oceanic Technology 33, 12; 10.1175/JTECH-D-16-0100.1

For the cold-scene calibration of the conically scanning window channel radiometers, four teams contributed results that relied on variations of the ocean double-difference method between coincident observations described above. These teams included the University of Michigan (Michigan), Colorado State University (CSU), the University of Central Florida (UCF), and Texas A&M University (TAMU). A fifth cold-scene estimate based on vicarious calibration, which does not require coincident observations between sensors, was contributed by the NASA Goddard Precipitation Processing System (PPS). Details for this approach are given in section 3b. Figure 4 shows results from these teams along with the final composite XCAL adjustments for the AMSR2 18.7-GHz channels. In the case of these AMSR2 channels, there is some variation in the resulting intercalibration differences, but all of the team’s results indicate a substantial cold-scene warm calibration bias with respect to GMI.

Fig. 4.
Fig. 4.

Intercalibration results for GCOM-W1 AMSR2 vs GMI for the 18.7-GHz (a) V-pol and (b) H-pol channels. Independent results from each of the contributing groups are shown for radiometrically cold scenes (nonprecipitating oceans) and warm scenes (vegetated nonpolarized land). The final composite XCAL calibration adjustment is indicated by the solid black line.

Citation: Journal of Atmospheric and Oceanic Technology 33, 12; 10.1175/JTECH-D-16-0100.1

Over oceans, the TAMU algorithm approach (Wilheit 2013) uses the GMI Tb to retrieve the most important geophysical variables for the spectral interval, including cloud liquid water (CLW), precipitable water, (PW), ocean surface wind speed (WS), and sea surface temperature (SST). Assumptions are used for the variables of lesser importance. For channels below 100 GHz, the TAMU algorithm assumes an average relative humidity (RH) profile and a temperature profile of fixed shape. It adjusts the WS, SST, CLW, and an offset to the temperature profile to minimize the mean square difference between the computed radiances and those observed by the reference radiometer using a nested-grid iteration. Since the relative humidity profile is fixed, the temperature profile offset becomes a proxy for PW.

The techniques employed by CSU, Michigan, and UCF are similar in that they obtain corresponding atmospheric and surface parameters from global model analyses. Michigan and UCF use model analysis data from the Global Data Assimilation System (GDAS), which uses NOAA’s Global Forecast System (GFS) model (NOAA/NCEP 2000). CSU uses the interim reanalysis data from the European Centre for Medium-Range Weather Forecasts (ECMWF) (Dee et al. 2011). Further details on the implementation of the double-difference techniques for each of these groups is given in Yang and McKague (2016) for the Michigan approach, in Biswas et al. (2013) for the UCF approach, and in Sapiano et al. (2013) for the CSU approach.

b. Vicarious calibration over oceans (window channel cold scenes)

A vicarious calibration approach, originally developed by Ruf (2000), was refined by the NASA Goddard PPS for calibration of the GPM constellation (Kroodsma et al. 2012, 2016). This technique does not require observations that are coincident in space and time. Instead, vicarious calibration relies on the stability of overocean Tb histograms to calculate a cold reference Tb for intercalibration. The brightness temperatures at the cold end of overocean histograms are associated with regions of calm surface winds, no clouds, and minimal atmospheric water vapor. These conditions give rise to a relatively stable cold end of the Tb histogram, from which a cold reference Tb is calculated. This cold reference Tb is derived for both target and reference radiometer observations, along with a cold reference Tb from radiometer simulations based on the GDAS analyses. Similar to the intercalibration techniques that use coincident overpasses, vicarious calibration simulates Tb for the target and reference radiometers, but the simulated Tb for vicarious calibration are generated for each radiometer pixel to give a comparable Tb histogram. The observed and simulated cold reference Tb for the target and reference radiometers are then used in the double-difference technique to calculate the intercalibration offset. The vicarious technique not only provides estimates of calibration differences corresponding to the coldest observable temperatures but also does not rely on the limited availability and regional distribution of coincident observations.

c. Double differences over unpolarized vegetated land (window channel warm scenes)

To determine sensor calibration differences for warm scenes, double differences were computed for depolarized areas in highly vegetated regions, such as the Amazon rain forest, using variations of the approach developed by Brown and Ruf (2005). Results were contributed by the Michigan, UCF, and CSU teams, all of which implemented variations on this approach. The warm-end calibration results have inherently larger errors than the cold-end ocean comparisons due to spatial variability within the scenes, significant diurnal heating, and limitations with modeling the surface contribution to the Tb. As the example in Fig. 4 shows, however, consistency between the various approaches over both cold and warm scenes indicates significant scene temperature dependence in the calibration of the AMSR2 18-GHz channels versus GMI.

The Michigan warm-scene approach extends the methods of Brown and Ruf (2005) to include boreal and temperate forests from mid- and high-latitude regions and additional tropical rain forests to extend the range of available warm calibration sites (Yang et al. 2016). This increases the sampling of the method by a factor of 30 relative to Amazon observations alone while also extending the Tb range covered by the method at the warm end. Similarly, the UCF approach applies differential polarized Tb filters (ΔTb < 2 K) to select land regions with thick vegetation canopies for the double-difference comparison; for example, the Amazon basin, the Congo basin of equatorial Africa, and the East Indies, from Sumatra to New Guinea, are typically used, although the majority of comparisons come from the Amazon. The surface emissivity of the tropical rain forest is modeled as a linear combination of near-blackbody forest canopy and smooth open water (Biswas et al. 2013). For the forest canopy, the model assumes that the emissivity is a near blackbody (slightly less than unity emissivity and a function of frequency) that is nonpolarized and is independent of the view angle. For the open water portion, the ocean radiative transfer model is used with an assumed zero wind speed that yields a specular polarized emission. The observed differential polarization Tb is used to solve for the unknown area fraction of water. For the atmospheric portion of the model, GDAS provides the temperature and water vapor profiles to calculate the atmospheric transmissivity.

d. Sounding channel calibration

A similar approach to the cold-scene ocean calibration used for the window channels is used to determine the calibration differences between GMI and the cross-track microwave sounders. Given the atmospheric opacity of the channels near 183 GHz and the significant differences in view angle and polarization, only a single-point calibration adjustment is done for these sensors. Teams providing results for the sounding channels include TAMU, UCF, Science Systems and Applications, Inc. (SSAI), and the Jet Propulsion Laboratory (JPL) of the California Institute of Technology. For the sounding channels above 100 GHz, the TAMU algorithm uses SST, WS, and the temperature profile provided by GDAS and iteratively adjusts the RH and CLW profiles to match the radiances of the reference radiometer using an algorithm described by Blankenship et al. (2000). The key to the iteration is the Jacobian derived by Schaerer and Wilheit (1979). The algorithm adds CLW to any level that becomes supersaturated at any stage in the iteration. The surface and atmosphere so derived are used to compute the radiances for both the reference and target radiometers used in the double differences.

The SSAI approach uses the six-channel cross-track-scanning SAPHIR measurements as a transfer standard to first derive biases (with respect to SAPHIR) of any sounding channels (GMI, MHS, ATMS, and SSMIS) using the double-difference technique previously discussed. It then derives the GMI-based biases by relating the differences between SAPHIR and GMI with those of other sensors. The pixel-to-pixel data pairs between SAPHIR and another sensor are matched with time and spatial constraints of 30 min and 10 km, respectively, and for cross-track scanners like MHS and ATMS, an additional view angle or EIA constraint of ±1° is applied. The six 183.31-GHz SAPHIR channels are used along with GDAS temperature and pressure profiles matched in time and space to retrieve RH profiles for each matched pixel (Wang et al. 1997). The resulting profiles are used in radiative transfer calculations in the double-difference technique to arrive at the biases of the matched pairs. Only the matched pairs with PW ≥ 6.5 cm are used in order to minimize radiance contributions from the ocean surface. The PW threshold was determined based on an analysis of moisture profiles retrieved from SAPHIR observations. The bias is quite independent of SAPHIR EIA, implying little impact of EIA dependence of surface emission/reflection. At 89 GHz, the GMI-based MHS biases are estimated directly from the Tb differences between GMI [mix-polarized Tb calculated from vertical (V) and horizontal (H) polarization] and MHS pairs with the same time, spatial, and EIA constraints listed above.

UCF implemented the double-difference technique described by Ebrahimi et al. (2014a,b) for frequencies above 100 GHz using GMI as the reference to compute calibration biases for the sounding channels on SAPHIR, MHS, ATMS, and SSMIS. For the cross-track scanners, implementing this method is slightly tedious in that there is a strong response in the observed Tb with EIA primarily because of changes in the atmospheric slant path and the strong absorption coefficient of WV at high altitudes. However, empirical results have established that for paired cross-track and conical sounders comparisons within ±3 h, the calibration bias variation with incidence angle is quite acceptable over clear ocean scenes.

For land scenes there is an imperfect knowledge of polarized surface emissivity; however, for the sounder channels, this is a very minor component of the top-of-the-atmosphere Tb, which is a mitigating factor. The cross-track scanner antennas also introduce changes in the observed polarization with scan angle. Fortunately, the degree of polarization mixing is well known from geometry; but reasonable estimates of surface emissivity are required, which complicate the overland comparisons. Finally, the antenna instantaneous field of view (IFOV) monotonically increases over the cross-track scan. Thus, when intercomparing cross-track scanning radiometers, the sensitivity of the biases on these above-mentioned factors is carefully considered.

The JPL approach utilizes radiosonde-based clear-sky temperature and humidity profiles over U.S. Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) sites. The observed profiles are used as input to a radiative transfer model to calculate single differences (i.e., observed minus simulated Tb) for the different radiometers based on overpasses of those sites. The single differences for the constellation sensors are then compared to those from GMI in order to evaluate calibration differences. The profiles used are from the ARM “merged sounding” product (Troyan 2012) that uses a combination of observations from radiosonde soundings, ground-based microwave radiometers, surface meteorological instruments, and ECMWF model output. A sophisticated scaling/interpolation/smoothing scheme is used in order to define profiles of the atmospheric thermodynamic state at 1-min intervals (ensuring that good temporal coincidence with satellite overpasses are always available). Channel-specific column water vapor thresholds are used to exclude cases where the measured Tb is sensitive to the surface, thus enabling the use of both land and ocean sites. Cloud screening was performed using the approach described in Buehler et al. (2007) utilizing observed channel differences.

4. GMI calibration

The use of GMI as a transfer standard demands calibration stability. If the absolute calibration can also be trusted, that simplifies the problem by enabling GMI to be used as both the calibration transfer standard and the reference standard. The design of GMI was predicated on eliminating potential (and in some cases observed) calibration issues. A very high-quality reflective coating was used on the main and cold calibration subreflectors. Noise diodes were included in the calibration scheme for the channels from 10.7 to 36.64 GHz. This provides a level of redundancy that enables an explicit solution for nonlinearity in the radiometer response functions and also provides the ability to bridge across times when the sun, moon, or radio frequency interference is in the cold calibration field of view. Additionally, the cold calibration subreflector is somewhat oversized to reduce spillover issues and the warm calibration load is shrouded to minimize solar intrusion (Draper et al. 2013).

After launch, a series of spacecraft attitude maneuvers were performed to provide checks on and refinements to the calibration. One of these maneuvers involved pitching up the GPM spacecraft so that both the main beam and the cold calibration mirror viewed cold space. The voltage from the radiometers was determined to be identical to within measurement uncertainty for both the main and subreflectors, thus indicating that either there are no significant emissivity issues for either reflector or they are identical for both. While it is highly improbable both reflectors would have identical emissivity issues, even in that case the reflector emissivity would cancel out in the calibration process to the extent that the two reflectors share the same physical temperature. Thus, it appears that the choice of the antenna and reflector coating resulted in a nonemissive or very low-emissivity reflector.

The GPM spacecraft was also pitched down so that the main beam viewed directly at nadir for part of the scan. On average, at nadir there should be no difference in the antenna temperatures between the horizontal and vertical polarizations. An analysis of the nadir-viewing data indicated that the differences were less than 0.3 K when viewing the ocean and less than 0.2 K when viewing land surfaces (Farrar et al. 2016). This also provides a significant constraint on the calibration. Note that due to schedule constraints requiring that the V04 GMI calibration be finalized by mid-2015 in order to meet early 2016 deadlines for GPM product development, preliminary analyses of the attitude maneuvers were used. A more refined analysis has subsequently been performed by Wentz and Draper (2016).

For a spaceborne radiometer, the largest source of calibration uncertainty in converting from counts to Tb involves the spillover correction, or the fraction (η) of the total antenna pattern that intercepts Earth. The prelaunch values were derived from antenna patterns either measured on a near-field range or modeled. The bulk of the portion that misses Earth (1 − η) comes from the portion of the feed pattern that misses the main reflector, that is, the spillover. Unfortunately, given that the spillover estimates from the near-field antenna measurements are very uncertain, model results were used instead. The column labeled “at launch” in Table 3 gives the values of η derived from the model.

Table 3.

The at-launch and V04 η values for GMI, and the corresponding change in Tb for a scene temperature of 200 K.

Table 3.

To refine the estimates of 1 − η, the GPM spacecraft was put into an inertial hold so that the spillover region intercepted Earth, while the main reflector viewed cold space. While some modeling is required to estimate Earth’s Tb, the impact of the uncertainty in these calculations is at least an order of magnitude less than using the Tb modeling directly given that the values of η are all greater than 0.9. The V04 η estimates for the two polarizations at each frequency were found to agree within their uncertainty, so the two values were averaged for each frequency to reduce the statistical error and to ensure that the constraints from the nadir-viewing observations were satisfied. The V04 η values are also provided in Table 3 along with the corresponding change in Tb between using the at-launch and V04 η values for a scene temperature of 200 K. These, along with straightforward numbers such as the calibration of the thermistors, constitute a purely physical calibration that is not tuned to any model and is completely independent of the calibration of any other spaceborne microwave radiometer.

Having arrived at a calibration based on the calibration maneuvers, it is worthwhile to compare it with other high-quality radiometers. Here we have chosen to compare GMI with WindSat on board the Coriolis satellite and MHS on board MetOp-B. For these comparisons, the TAMU (below 100 GHz) algorithm was used to compute the differences between the corresponding GMI and WindSat radiances (up to 37 GHz), and the TAMU water vapor profile retrieval (above 100 GHz) algorithm was used to compute differences between the corresponding GMI and MHS radiances (89 GHz and above). Table 4 shows a comparison of the GMI versus WindSat double differences using both the at-launch and V04 calibrations. The at-launch comparison with WindSat show large differences, as much as 2.84 K for the 10V channel and an RMS difference of 1.80 K across all seven channels. The V04 calibration reduces this error to a maximum (in an absolute sense) of −1.92 K for the 18H channel with a seven-channel RMS of 1.25, a reduction of the variance by more than a factor of 2.

Table 4.

Double differences between GMI and WindSat (all values in K).

Table 4.

The GMI calibration discussed here and the WindSat calibration are entirely independent so that we may examine the differences statistically. Both the WindSat and GMI instruments were carefully designed for maximum calibration accuracy and both used on-orbit maneuvers to determine the emissivity of the main reflector (Gaiser et al. 2004; Jones et al. 2006; Twarog et al. 2006; Draper et al. 2013, 2015a,b). The maneuvers to check polarization consistency at nadir and to look at the backlobes, however, were done only for GMI. Based on the RMS difference of 1.25 K and the calibration corrections and checks done on orbit, it is conservative to state that for the 10–37-GHz channels the RMS calibration error for GMI is less than 1 K.

For the GMI–MHS comparison, MHS radiances were used in conjunction with GDAS temperature profiles to retrieve water vapor and cloud liquid water profiles. The atmospheric profiles were then used to compute both GMI and MHS radiances. Although it varies by sensor and channel, there is significant improvement in using the retrieved water vapor and cloud liquid water profiles, as the retrieval accurately captures spatial and temporal variations that frequently occur in the analysis datasets. The results are given in Table 5 for both the at-launch and V04 GMI calibrations. The at-launch differences for the 166- and 183-GHz channels are quite large with an RMS difference of more than 2 K. This is not surprising, since these channels used a prelaunch value of η = 1, a very unphysical value. Using the V04 η values results in a significant decrease in the GMI–MHS channel differences and reduces the RMS to 0.48 K. Results for the 89-GHz channels were virtually unchanged. As with the WindSat comparison, the independence of the calibrations suggests that the GMI calibration for these channels is better than 0.5 K. Note that radiosonde-based comparisons for the GMI sounding channels (discussed in section 3d above) also show a significant reduction in the residual differences between simulated and observed Tb using the V04 GMI calibration.

Table 5.

Double differences between MetOp-B MHS and GMI (all values in K).

Table 5.

For the V04 GMI calibration, the 89H channel exhibits the largest difference with MHS. For this channel we use MHS with a single rotating polarization to predict the two GMI polarizations. This is very demanding on the surface emissivity model used. If we turn the problem around and use GMI to retrieve the water vapor profiles rather than MHS and predict the MHS channels from GMI, we now have two polarizations at 89 and 166 GHz to derive the radiance at a single polarization. This reduces the RMS difference to 0.35 K. It is also of interest to note that the difference for the MHS 183.31 ± 1 GHz channel, which has no true GMI equivalent, is only 0.16 K.

The polarization issue illustrates another contributor to the apparent calibration differences between sensors, shortcomings in the algorithms, and methods used to translate between sensors. As is discussed further in section 6, uncertainties in the methods used to compare sensor calibrations contribute to the apparent differences. Thus, it is conservative to state that the absolute calibration accuracy of GMI is within 1 K at 37 GHz and below, and 0.5 K above. Furthermore, as mentioned previously, the GMI orbit creates many coincident observations with all of the other constellation members from which intercomparisons can be made. For version 4, therefore, GMI is used as the calibration reference standard to which all the constellation sensors are adjusted.

5. Radiometer prescreening and calibration monitoring

The first step in dealing with a new sensor involves a prescreening process to identify and remove calibration biases that occur across the scan or along the orbit path. This typically involves investigating cross-scan biases, ascending versus descending calibration differences, and seasonal or other time-dependent calibration changes (Yang and McKague 2016; Berg and Sapiano 2013). An assessment of the previously identified TMI emissive reflector issue (Wentz et al. 2001) led to an improved time-dependent correction (Gopalan et al. 2009; Biswas et al. 2010). Other corrections applied to the TMI data included corrections for cross-track biases (Wentz et al. 2001) and cone angle adjustments for the 10v, 10h, and 37v channels based on prelaunch measurements (J. Shiue 1997, personal communication). The SSMIS sensors on board DMSP F16F19 were also found to exhibit significant cross-scan and ascending/descending scan biases. Corrections for cross-scan biases and biases based on the position of the sun relative to the spacecraft were developed for the imager channels (i.e., 19, 22, 37, and 91 GHz) on board F16F18 by Berg and Sapiano (2013). The sun-angle correction approach was developed to account for complicated calibration bias patterns associated with emissive reflector issues on F16 and F17, direct and indirect solar intrusions into the warm load, and thermal heating issues (Kunkee et al. 2008a,b; Bell et al. 2008). Cross-track and sun-angle calibration corrections were subsequently developed for SSMIS on board DMSP F19 by teams at Michigan, CSU, and the PPS based on an initial analysis of 4 months of data after it became operational in November 2014. The sun-angle corrections were also extended to include the sounding channels for F17 and F18. The F16 SSMIS 183-GHz channels were set to missing in December 2013 due to increasing noise and drifts in the antenna temperatures (Ta). Similarly, the 150-GHz channels failed for F16 in May 2015 and for F18 in February 2012. These recent SSMIS channel failures were identified from ongoing monitoring of the raw sensor data, channel noise, and time-dependent calibration for each of the radiometers in the GPM constellation by the XCAL team.

While the XCAL team has full access to all of the raw sensor data and the ability to implement corrections for the TMI and GMI instruments, the same is not true for most of the other radiometers in the constellation. The magnitude of calibration issues with the DMSP SSMIS instruments impacting the operational SDR Tb dataset compelled the XCAL team to develop and implement corrections for these sensors. For the remaining constellation radiometers, corrections and calibration of the level 1B Tb data obtained by PPS are done by the respective space agencies. As a result, the XCAL team endeavors to work with the agencies responsible for the level 1B calibration in order to address issues found as a result of the data prescreening. Changes to the level 1B calibration by these agencies, however, require the team to quickly evaluate the impact and adjust the resulting intercalibration adjustments applied to produce the level 1C Tb data.

6. Intercalibration versus GMI

The individual estimates (contributing teams given in Table 2) of the constellation radiometer calibration differences versus GMI are composited for each sensor to create the final XCAL calibration adjustment tables. This was done by removing any outliers and averaging the remaining calibration difference estimates. For the window channels, the cold-scene and warm-scene results were averaged separately to provide a two-point calibration adjustment. The resulting intercalibration adjustments derived by the XCAL team and subsequently applied to the GPM constellation sensors are given in Table 6 and shown for comparison in Fig. 5.

Table 6.

Window channel radiometer calibration vs GMI. The two columns for each sensor correspond to the cold-scene (left) and warm-scene (right) calibration differences vs GMI with the corresponding mean GMI Tb value in parentheses.

Table 6.
Fig. 5.
Fig. 5.

Composite XCAL team intercalibration differences between GMI, which is the reference sensor, and the microwave window channel radiometers for both cold ocean scenes (blue bars) and warm vegetated land scenes (red bars).

Citation: Journal of Atmospheric and Oceanic Technology 33, 12; 10.1175/JTECH-D-16-0100.1

TMI has differences relative to GMI of up to ~2.5 K that are relatively consistent between warm and cold scenes with temperature-dependent variations in the calibration of less than ~1 K. Comparisons with WindSat over multiple years also indicate that the TMI calibration is quite stable after correcting for the emissive reflector issue (Biswas et al. 2010). AMSR2 has a substantial warm bias over cold ocean scenes relative to GMI, with values more than 4 K higher for the 18v and 36h channels. In addition, calibration differences over warm land scenes are quite small, even negative for some channels, thus leading to large temperature-dependent variations in the calibration versus GMI. For most sensors, including the 18-GHz AMSR2 channels shown in Fig. 4, the dominant sources of calibration error are antenna pattern corrections (for the fraction of the signal missing Earth), emissive reflectors, and warm calibration target temperature errors. All of these can be corrected for using a two-point linear correction. There are questions, however, related to large nonlinearity corrections applied to the AMSR2 antenna temperatures that are based on prelaunch measurements. Unfortunately, uncertainties in the intercalibration estimates are as large or larger than errors due to a nonlinear detector response. This is apparent in Fig. 4, where the spread of points at the cold end, which provides a measure of the residual uncertainty, prevents fitting any sort of nonlinear function with any degree of confidence. As such there may be residual calibration errors with some of the AMSR2 channels related to this nonlinear detector response.

The SSMIS window channels on board DMSP F16F18 are relatively consistent with GMI, but as shown in Fig. 5 they vary significantly between cold and warm scenes. The calibration of SSMIS on board F19 appears a bit colder than the three older ones, but it also exhibits a temperature-dependent bias for several channels. Finally, the WindSat calibration varies with respect to GMI by as much as 2 K, but there is very little temperature dependence in the differences.

Calibration differences versus GMI for the cross-track sounders and the SSMIS sounding channels are given in Table 7 and shown in Fig. 6. As discussed previously, for these channels a single-point calibration adjustment or offset was computed. The calibration of SAPHIR and the MHS instruments on board MetOp-A, MetOp-B, NOAA-18, and NOAA-19 are remarkably consistent with GMI, with differences consistently below 0.5 K. There are slightly larger differences for the 183.31 ± 1 GHz channel, although still within 1 K. This is not unexpected given the sensitivity to upper-tropospheric water vapor in this channel and with the lack of a comparable GMI channel, thus resulting in larger uncertainties in the double-difference estimates. The consistency of the calibrations between GMI and these cross-track sensors, however, emphasizes the importance of the on-orbit-derived η values given in Table 3. The ATMS instrument on board Suomi-NPP has slightly larger differences versus GMI, but the values are still all within 1 K. Note that a planned reprocessing of the Suomi-NPP ATMS SDR (i.e., Tb) dataset is expected to change the Tb for some channels by up to 0.5 K. Once the new ATMS SDR data are available, comparisons with GMI will be redone to determine whether it improves the overall agreement.

Table 7.

Sounding channel radiometer calibration vs GMI. As for the imagers, but only a single calibration difference reported for overocean scenes. The mean GMI Tb values corresponding to the calibration differences are given in parenthesis.

Table 7.
Fig. 6.
Fig. 6.

Composite XCAL team intercalibration differences between GMI and the microwave sounding channels for both the DMSP SSMIS sensors and the cross-track humidity sounding sensors. Note that the y-axis range is ±2 K for all instruments except F19, which shows substantially larger biases.

Citation: Journal of Atmospheric and Oceanic Technology 33, 12; 10.1175/JTECH-D-16-0100.1

Differences between the SSMIS sounding channels and GMI are within ~2 K with the exception of F19, which is substantially colder. Note that the differences for the F19 SSMIS window channels do not exhibit such large differences. The four affected channels, however, all share the same feed horn. One theory that has been suggested is that there may be a piece of debris lodged in the feed horn for those channels. Regardless, the calibration differences for these channels appear stable and thus useable once corrected for their cold bias. While there may be residual time-varying errors or temperature dependence in the calibration bias, the impact on the resulting precipitation estimates is likely to be small given that the retrieval algorithm applies more weight to the low-frequency window channels for this sensor.

7. Summary and conclusions

Providing consistent input Tb from the GPM constellation radiometers involves dealing with a wide variation in the available channels, channel characteristics, and viewing geometries from both conical and cross-track scanners, as well as identifying and correcting for numerous issues affecting the individual instrument calibrations. The approach taken by the XCAL team involves prescreening the data for cross-scan or along-orbit calibration issues, developing and applying corrections, and finally comparing the observed Tb to a calibration reference (i.e., GMI) accounting for channel differences. Fortunately, the GMI instrument on board the GPM Core Observatory appears to be extremely well calibrated and stable. On-orbit calibration maneuvers for GMI were used to check for potential calibration issues, such as polarization, reflector emissivity, and solar intrusions into the warm load. In addition, corrections were developed for magnetic-induced anomalies and spillover corrections were updated. Subsequent analysis based on radiative transfer simulations and comparisons with other well-calibrated sensors, including WindSat and MHS, suggest residual calibration errors less than 1 K for all channels. In addition, the 65° inclination of the GPM satellite provides frequent coincident overpasses with polar-orbiting sensors. GMI also has a full complement of window and water vapor sounding channels, thus making it an excellent calibration reference for the entire constellation of conical and cross-track microwave radiometers.

Prescreening of the GPM constellation radiometers led to corrections for TMI emissive reflector issues and sun-angle-dependent corrections for the SSMIS sensors to account for emissive reflector and solar intrusion issues. Cross-scan bias corrections were also applied for the TMI and SSMIS instruments. Subsequent intercalibration with GMI finds typical calibration differences for the window channels below 92 GHz within 2–3 K. AMSR2 exhibits larger differences, however, with the 18v and 36h channels more than 4 K warmer than GMI. The AMSR2 calibration differences also vary significantly with scene temperature as determined from separate cold-scene ocean and warm-scene land analyses. SSMIS calibration differences also vary with scene temperature, although to a much lesser degree. For the SSMIS channels above 150 GHz, the differences are generally within ~2 K with the exception of F19, which ranges from 7 to 11 K colder than GMI depending on frequency. Finally, the calibrations of the cross-track radiometers agree very well with GMI with differences mostly within 0.5 K for the SAPHIR and all four MHS sensors. Differences between GMI and ATMS on board the Suomi-NPP are slightly larger, but they still remain within 1 K for all channels.

While the calibration of the cross-track sounders is simpler given that the main reflector directly views the cold target (i.e., deep space), the consistency with GMI validates the GMI calibration and emphasizes the importance of the postlaunch calibration maneuvers. Updates to the spillover correction in particular point to significant uncertainties in the prelaunch measured values at 10 and 18 GHz, as well as for the 166- and 183-GHz channels for which prelaunch measured values were not available. It seems quite likely that uncertainties in the spillover corrections for the other conically scanning imagers, particularly at the lower frequencies, where the corrections are largest, may account for a significant portion of the resulting calibration differences between sensors. The XCAL approach involves multiple teams analyzing calibration differences using different techniques and/or implementations. While this increases confidence in the results, uncertainties due to the radiative transfer models and errors in the geophysical parameter remain, particularly when comparing channels with significant frequency, polarization, and/or view angle differences. Differences in the intercalibration estimates from the various techniques indicate that residual uncertainties after the composite intercalibration offsets have been applied to the constellation radiometers are less than 1 K and below 0.5 K for most sensors/channels. Better understanding and quantifying the residual uncertainties, however, are some of the major tasks for the XCAL team going forward. Other challenges involve improving calibration methods, adapting to changes in the radiometer constellation, and revisiting previous radiometers to develop a long-term intercalibrated TRMM–GPM constellation data record. Given that the TMI data record extends an impressive 17-plus years, the combined TRMM–GPM datasets provide the opportunity to create a consistent long-term data record.

Acknowledgments

The authors thank the NASA Precipitation Processing System for providing the datasets used for the intercalibration analysis. Funding for this work was provided by NASA’s Global Precipitation Measurement (GPM) mission; NASA contracts at Ball Aerospace (Goddard SESDA III No. S-Ball-01), the Jet Propulsion Laboratory, California Institute of Technology, and Science Systems and Applications, Inc. (SSAI); and the NASA Precipitation Measurement Missions (PMM) science team under Grants NNX13AG30G, NNX16AE35G, NNX13AG70G, and NNX13AG46G. Reference herein to any specific commercial product, process, or service by trade name, manufacturer, or otherwise does not constitute or imply its endorsement by the U.S. government or the Jet Propulsion Laboratory, California Institute of Technology.

REFERENCES

  • Bell, W., and Coauthors, 2008: The assimilation of SSMIS radiances in numerical weather prediction models. IEEE Trans. Geosci. Remote Sens., 46, 884900, doi:10.1109/TGRS.2008.917335.

    • Search Google Scholar
    • Export Citation
  • Berg, W., and Sapiano M. , 2013: Corrections and APC for SSMIS. Colorado State University Tech. Rep., 30 pp. [Available online at http://rain.atmos.colostate.edu/FCDR/doc/CSU_FCDR_ssmis_corrections_tech_report.pdf.]

  • Berg, W., Sapiano M. , Horsman J. , and Kummerow C. , 2013: Improved geolocation and Earth incidence angle information for a fundamental climate data record of the SSM/I sensors. IEEE Trans. Geosci. Remote Sens., 51, 15041513, doi:10.1109/TGRS.2012.2199761.

    • Search Google Scholar
    • Export Citation
  • Biswas, S. K., Gopalan K. , Jones W. L. , and Bilanow S. , 2010: Correction of time-varying radiometric errors in TRMM Microwave Imager calibrated brightness temperature products. IEEE Geosci. Remote Sens. Lett., 7, 851855, doi:10.1109/LGRS.2010.2050135.

    • Search Google Scholar
    • Export Citation
  • Biswas, S. K., Gopalan K. , Santos-Garcia A. , and Jones W. L. , 2013: Intercalibration of the microwave radiometer brightness temperature for the Global Precipitation Measurement Mission. IEEE Trans. Geosci. Remote Sens., 51, 14651477, doi:10.1109/TGRS.2012.2217148.

    • Search Google Scholar
    • Export Citation
  • Blankenship, C., Al-Khalaf A. , and Wilheit T. , 2000: Retrieval of water vapor profiles using SSM/T-2 and SSM/I data. J. Atmos. Sci., 57, 939955, doi:10.1175/1520-0469(2000)057<0939:ROWVPU>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Brown, S. T., and Ruf C. S. , 2005: Determination of an Amazon hot reference target for the on-orbit calibration of microwave radiometers. J. Atmos. Oceanic Technol., 22, 13401352, doi:10.1175/JTECH1769.1.

    • Search Google Scholar
    • Export Citation
  • Buehler, S. A., Kuvatov M. , Sreerekha T. R. , John V. O. , Rydberg B. , Eriksson P. , and Notholt J. , 2007: A cloud filtering method for microwave upper tropospheric humidity measurements. Atmos. Chem. Phys., 7, 55315542, doi:10.5194/acp-7-5531-2007.

    • Search Google Scholar
    • Export Citation
  • Chander, G., Hewison T. J. , Fox N. , Wu X. , Xiong X. , and Blackwell W. J. , 2013: Overview of intercalibration of satellite instruments. IEEE Trans. Geosci. Remote Sens., 51, 10561080, doi:10.1109/TGRS.2012.2228654.

    • Search Google Scholar
    • Export Citation
  • Chung, E.-S., Soden B. , and John V. O. , 2013: Intercalibrating microwave satellite observations for monitoring long-term variations in upper- and midtropospheric water vapor. J. Atmos. Oceanic Technol., 30, 23032319, doi:10.1175/JTECH-D-13-00001.1.

    • Search Google Scholar
    • Export Citation
  • Colton, M. C., and Poe G. A. , 1999: Intersensor calibration of DMSP SSM/I’s: F-8 to F-14, 1987-1997. IEEE Trans. Geosci. Remote Sens., 37, 418439, doi:10.1109/36.739079.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Draper, D. W., Newell D. A. , Teusch D. A. , and Yoho P. K. , 2013: Global Precipitation Measurement Microwave Imager prelaunch hot load calibration. IEEE Trans. Geosci. Remote Sens., 51, 47314742, doi:10.1109/TGRS.2013.2239300.

    • Search Google Scholar
    • Export Citation
  • Draper, D. W., Newell D. A. , McKague D. S. , and Piepmeier J. R. , 2015a: Assessing calibration stability using the Global Precipitation Measurement (GPM) Microwave Imager (GMI) noise diodes. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 8, 42394247, doi:10.1109/JSTARS.2015.2406661.

    • Search Google Scholar
    • Export Citation
  • Draper, D. W., Newell D. A. , Wentz F. J. , Krimchansky S. , and Skofronick-Jackson G. , 2015b: The Global Precipitation Measurement (GPM) Microwave Imager (GMI): Instrument overview and early on-orbit performance. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 8, 34523462, doi:10.1109/JSTARS.2015.2403303.

    • Search Google Scholar
    • Export Citation
  • Ebrahimi, H., Datta S. , Santos-Garcia A. , and Jones W. L. , 2014a: Radiometric inter-calibration of SAPHIR using the Microwave Humidity Sounders. 13th Specialist Meeting on Microwave Radiometry and Remote Sensing of the Environment: MicroRad 2014; Proceedings, IEEE, 211–214, 10.1109/MicroRad.2014.6878942.

  • Ebrahimi, H., Datta S. , and Jones W. L. , 2014b: Radiometric inter-calibration of satellite microwave humidity sounders using the GPM Microwave Imager. 2014 IEEE International Geoscience and Remote Sensing Symposium: Proceedings, IEEE, 2566–2569, doi:10.1109/IGARSS.2014.6946997.

  • Elsaesser, G. S., and Kummerow C. D. , 2008: Toward a fully parametric retrieval of the nonraining parameters over the global oceans. J. Appl. Meteor. Climatol., 47, 15991618, doi:10.1175/2007JAMC1712.1.

    • Search Google Scholar
    • Export Citation
  • Farrar, S., Draper D. , Jones L. , and Alquaied F. , 2016: Advantages of calibration attitude maneuvers for spaceborne microwave radiometer missions. 14th Specialist Meeting on Microwave Radiometry and Remote Sensing of the Environment: Proceedings, IEEE, 186–190, doi:10.1109/MICRORAD.2016.7530531.

  • Gaiser, P. W., and Coauthors, 2004: The WindSat spaceborne polarimetric microwave radiometer: Sensor description and early orbit performance. IEEE Trans. Geosci. Remote Sens., 42, 23472361, doi:10.1109/TGRS.2004.836867.

    • Search Google Scholar
    • Export Citation
  • Goldberg, M., and Coauthors, 2011: The Global Space-based Inter-Calibration System. Bull. Amer. Meteor. Soc., 92, 467475, doi:10.1175/2010BAMS2967.1.

    • Search Google Scholar
    • Export Citation
  • Gopalan, K., Jones L. , Biswas S. , Bilanow S. , Wilheit T. , and Kasparis T. , 2009: A time-varying radiometric bias correction for the TRMM Microwave Imager. IEEE Trans. Geosci. Remote Sens., 47, 37223730, doi:10.1109/TGRS.2009.2028882.

    • Search Google Scholar
    • Export Citation
  • Hou, A. Y., and Coauthors, 2014: The Global Precipitation Measurement Mission. Bull. Amer. Meteor. Soc., 95, 701722, doi:10.1175/BAMS-D-13-00164.1.

    • Search Google Scholar
    • Export Citation
  • Imaoka, K., and Coauthors, 2003: Post-launch calibration and data evaluation of AMSR-E. 2003 IEEE International Geoscience and Remote Sensing Symposium, Vol. 1, IEEE, 666–668, doi:10.1109/IGARSS.2003.1293875.

  • John, V. O., Holl G. , Buehler S. A. , Candy B. , Saunders R. W. , and Parker D. E. , 2012: Understanding intersatellite biases of microwave humidity sounders using global simultaneous nadir overpasses. J. Geophys. Res., 117, D02305, doi:10.1029/2011JD016349.

    • Search Google Scholar
    • Export Citation
  • John, V. O., Allan R. P. , Bell W. , Buehler S. A. , and Kottayil A. , 2013: Assessment of intercalibration methods for satellite microwave humidity sounders. J. Geophys. Res. Atmos., 118, 49064918, doi:10.1002/jgrd.50358.

    • Search Google Scholar
    • Export Citation
  • Jones, W. L., Park J. D. , Soisuvarn S. , Hong L. , Gaiser P. W. , and St. Germain K. M. , 2006: Deep-space calibration of the WindSat radiometer. IEEE Trans. Geosci. Remote Sens., 44, 476495, doi:10.1109/TGRS.2005.862499.

    • Search Google Scholar
    • Export Citation
  • Kroodsma, R. A., McKague D. S. , and Ruf C. S. , 2012: Inter-calibration of microwave radiometers using the vicarious cold calibration double difference method. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 5, 10061013, doi:10.1109/JSTARS.2012.2195773.

    • Search Google Scholar
    • Export Citation
  • Kroodsma, R. A., McKague D. S. , and Ruf C. S. , 2016: Vicarious cold calibration for conical scanning microwave imagers. IEEE Trans. Geosci. Remote Sens., doi:10.1109/TGRS.2016.2615552, in press.

    • Search Google Scholar
    • Export Citation
  • Kummerow, C., 1993: On the accuracy of the Eddington approximation for radiative transfer in the microwave frequencies. J. Geophys. Res., 98, 27572765, doi:10.1029/92JD02472.

    • Search Google Scholar
    • Export Citation
  • Kunkee, D. B., Poe G. A. , Boucher D. J. , Swadley S. D. , Hong Y. , Wessel J. E. , and Uliana E. A. , 2008a: Design and evaluation of the first Special Sensor Microwave Imager/Sounder. IEEE Trans. Geosci. Remote Sens., 46, 863883, doi:10.1109/TGRS.2008.917980.

    • Search Google Scholar
    • Export Citation
  • Kunkee, D. B., Swadley S. D. , Poe G. A. , Hong Y. , and Werner M. F. , 2008b: Special Sensor Microwave Imager Sounder (SSMIS) radiometric calibration anomalies—Part I: Identification and characterization. IEEE Trans. Geosci. Remote Sens., 46, 10171033, doi:10.1109/TGRS.2008.917213.

    • Search Google Scholar
    • Export Citation
  • Meissner, T., and Wentz F. J. , 2012: The emissivity of the ocean surface between 6 and 90 GHz over a large range of wind speeds and Earth incidence angles. IEEE Trans. Geosci. Remote Sens., 50, 30043026, doi:10.1109/TGRS.2011.2179662.

    • Search Google Scholar
    • Export Citation
  • Moradi, I., Meng H. , Ferraro R. R. , and Bilanow S. , 2013: Correcting geolocation errors for microwave instruments aboard NOAA satellites. IEEE Trans. Geosci. Remote Sens., 51, 36253637, doi:10.1109/TGRS.2012.2225840.

    • Search Google Scholar
    • Export Citation
  • Moradi, I., Ferraro R. R. , Eriksson P. , and Weng F. , 2015: Intercalibration and validation of observations from ATMS and SAPHIR microwave sounders. IEEE Trans. Geosci. Remote Sens., 53, 59155925, doi:10.1109/TGRS.2015.2427165.

    • Search Google Scholar
    • Export Citation
  • NOAA/NCEP, 2000: NCEP FNL Operational Model Global Tropospheric Analyses, continuing from July 1999 (updated daily). National Center for Atmospheric Research Computational and Information Systems Laboratory Research Data Archive, accessed 12 May 2016, doi:10.5065/D6M043C6.

  • Rosenkranz, P. W., 1998: Water vapor microwave continuum absorption: A comparison of measurements and models. Radio Sci., 33, 919928, doi:10.1029/98RS01182.

    • Search Google Scholar
    • Export Citation
  • Ruf, C. S., 2000: Detection of calibration drifts in spaceborne microwave radiometers using a vicarious cold reference. IEEE Trans. Geosci. Remote Sens., 38, 4452, doi:10.1109/36.823900.

    • Search Google Scholar
    • Export Citation
  • Sapiano, M., Berg W. , McKague D. , and Kummerow C. , 2013: Towards an intercalibrated fundamental climate data record of the SSM/I sensors. IEEE Trans. Geosci. Remote Sens., 51, 14921503, doi:10.1109/TGRS.2012.2206601.

    • Search Google Scholar
    • Export Citation
  • Schaerer, G., and Wilheit T. T. , 1979: A passive microwave technique for profiling of atmospheric water vapor. Radio Sci., 14, 371375, doi:10.1029/RS014i003p00371.

    • Search Google Scholar
    • Export Citation
  • Troyan, D., 2012: Merged sounding value added product. U.S. Dept. of Energy Rep. DOE/SC-ARM-TR-087, 13 pp. [Available online at https://www.arm.gov/publications/tech_reports/doe-sc-arm-tr-087.pdf.]

  • Twarog, E. M., Purdy W. E. , Gaiser P. W. , Cheung K. H. , and Kelm B. E. , 2006: WindSat on-orbit warm load calibration. IEEE Trans. Geosci. Remote Sens., 44, 516529, doi:10.1109/TGRS.2005.863300.

    • Search Google Scholar
    • Export Citation
  • Wang, J. R., Racette P. , and Chang L. A. , 1997: MIR measurements of atmospheric water vapor profiles. IEEE Trans. Geosci. Remote Sens., 35, 212221, doi:10.1109/36.563259.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., 1997: A well-calibrated ocean algorithm for special sensor microwave/imager. J. Geophys. Res., 102, 87038718, doi:10.1029/96JC01751.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., 2015: A 17-year climate record of environmental parameters derived from the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager. J. Climate, 28, 68826902, doi:10.1175/JCLI-D-15-0155.1.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., and Draper D. , 2016: On-orbit absolute calibration of the Global Precipitation Mission Microwave Imager. J. Atmos. Oceanic Technol., 33, 13931412, doi:10.1175/JTECH-D-15-0212.1.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., Ashcroft P. , and Gentemann C. , 2001: Post-launch calibration of the TRMM microwave imager. IEEE Trans. Geosci. Remote Sens., 39, 415422, doi:10.1109/36.905249.

    • Search Google Scholar
    • Export Citation
  • Wilheit, T., 2013: Comparing calibrations of similar conically scanning window-channel microwave radiometers. IEEE Trans. Geosci. Remote Sens., 51, 14531464, doi:10.1109/TGRS.2012.2207122.

    • Search Google Scholar
    • Export Citation
  • Wilheit, T., Berg W. , Ebrahami H. , Kroodsma R. , McKague D. , Payne V. , and Wang J. , 2015: Intercalibrating the GPM constellation using the GPM Microwave Imager (GMI). 2015 IEEE International Geoscience and Remote Sensing Symposium: Proceedings, IEEE, 5162–5165, doi:10.1109/IGARSS.2015.7326996.

  • Yan, B., and Weng F. , 2008: Intercalibration between Special Sensor Microwave Imager/Sounder and Special Sensor Microwave Imager. IEEE Trans. Geosci. Remote Sens., 46, 984995, doi:10.1109/TGRS.2008.915752.

    • Search Google Scholar
    • Export Citation
  • Yang, J. X., and McKague D. S. , 2016: Improving collocation-based scan-dependent intercalibration over the ocean for spaceborne radiometry. IEEE Geosci. Remote Sens. Lett., 13, 589593, doi:10.1109/LGRS.2016.2528261.

    • Search Google Scholar
    • Export Citation
  • Yang, J. X., McKague D. S. , and Ruf C. S. , 2016: Boreal, temperate, and tropical forests as vicarious calibration sites for spaceborne microwave radiometry. IEEE Trans. Geosci. Remote Sens., 54, 10351051, doi:10.1109/TGRS.2015.2472532.

    • Search Google Scholar
    • Export Citation
  • Yang, S., Weng F. , Yan B. , Sun N. , and Goldberg M. , 2011: Special Sensor Microwave Imager (SSM/I) intersensor calibration using a simultaneous conical overpass technique. J. Appl. Meteor. Climatol., 50, 7795, doi:10.1175/2010JAMC2271.1.

    • Search Google Scholar
    • Export Citation
  • Zou, C.-Z., and Wang W. , 2011: Intersatellite calibration of AMSU-A observations for weather and climate applications. J. Geophys. Res., 116, D23113, doi:10.1029/2011JD016205.

    • Search Google Scholar
    • Export Citation
Save
  • Bell, W., and Coauthors, 2008: The assimilation of SSMIS radiances in numerical weather prediction models. IEEE Trans. Geosci. Remote Sens., 46, 884900, doi:10.1109/TGRS.2008.917335.

    • Search Google Scholar
    • Export Citation
  • Berg, W., and Sapiano M. , 2013: Corrections and APC for SSMIS. Colorado State University Tech. Rep., 30 pp. [Available online at http://rain.atmos.colostate.edu/FCDR/doc/CSU_FCDR_ssmis_corrections_tech_report.pdf.]

  • Berg, W., Sapiano M. , Horsman J. , and Kummerow C. , 2013: Improved geolocation and Earth incidence angle information for a fundamental climate data record of the SSM/I sensors. IEEE Trans. Geosci. Remote Sens., 51, 15041513, doi:10.1109/TGRS.2012.2199761.

    • Search Google Scholar
    • Export Citation
  • Biswas, S. K., Gopalan K. , Jones W. L. , and Bilanow S. , 2010: Correction of time-varying radiometric errors in TRMM Microwave Imager calibrated brightness temperature products. IEEE Geosci. Remote Sens. Lett., 7, 851855, doi:10.1109/LGRS.2010.2050135.

    • Search Google Scholar
    • Export Citation
  • Biswas, S. K., Gopalan K. , Santos-Garcia A. , and Jones W. L. , 2013: Intercalibration of the microwave radiometer brightness temperature for the Global Precipitation Measurement Mission. IEEE Trans. Geosci. Remote Sens., 51, 14651477, doi:10.1109/TGRS.2012.2217148.

    • Search Google Scholar
    • Export Citation
  • Blankenship, C., Al-Khalaf A. , and Wilheit T. , 2000: Retrieval of water vapor profiles using SSM/T-2 and SSM/I data. J. Atmos. Sci., 57, 939955, doi:10.1175/1520-0469(2000)057<0939:ROWVPU>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Brown, S. T., and Ruf C. S. , 2005: Determination of an Amazon hot reference target for the on-orbit calibration of microwave radiometers. J. Atmos. Oceanic Technol., 22, 13401352, doi:10.1175/JTECH1769.1.

    • Search Google Scholar
    • Export Citation
  • Buehler, S. A., Kuvatov M. , Sreerekha T. R. , John V. O. , Rydberg B. , Eriksson P. , and Notholt J. , 2007: A cloud filtering method for microwave upper tropospheric humidity measurements. Atmos. Chem. Phys., 7, 55315542, doi:10.5194/acp-7-5531-2007.

    • Search Google Scholar
    • Export Citation
  • Chander, G., Hewison T. J. , Fox N. , Wu X. , Xiong X. , and Blackwell W. J. , 2013: Overview of intercalibration of satellite instruments. IEEE Trans. Geosci. Remote Sens., 51, 10561080, doi:10.1109/TGRS.2012.2228654.

    • Search Google Scholar
    • Export Citation
  • Chung, E.-S., Soden B. , and John V. O. , 2013: Intercalibrating microwave satellite observations for monitoring long-term variations in upper- and midtropospheric water vapor. J. Atmos. Oceanic Technol., 30, 23032319, doi:10.1175/JTECH-D-13-00001.1.

    • Search Google Scholar
    • Export Citation
  • Colton, M. C., and Poe G. A. , 1999: Intersensor calibration of DMSP SSM/I’s: F-8 to F-14, 1987-1997. IEEE Trans. Geosci. Remote Sens., 37, 418439, doi:10.1109/36.739079.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Draper, D. W., Newell D. A. , Teusch D. A. , and Yoho P. K. , 2013: Global Precipitation Measurement Microwave Imager prelaunch hot load calibration. IEEE Trans. Geosci. Remote Sens., 51, 47314742, doi:10.1109/TGRS.2013.2239300.

    • Search Google Scholar
    • Export Citation
  • Draper, D. W., Newell D. A. , McKague D. S. , and Piepmeier J. R. , 2015a: Assessing calibration stability using the Global Precipitation Measurement (GPM) Microwave Imager (GMI) noise diodes. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 8, 42394247, doi:10.1109/JSTARS.2015.2406661.

    • Search Google Scholar
    • Export Citation
  • Draper, D. W., Newell D. A. , Wentz F. J. , Krimchansky S. , and Skofronick-Jackson G. , 2015b: The Global Precipitation Measurement (GPM) Microwave Imager (GMI): Instrument overview and early on-orbit performance. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 8, 34523462, doi:10.1109/JSTARS.2015.2403303.

    • Search Google Scholar
    • Export Citation
  • Ebrahimi, H., Datta S. , Santos-Garcia A. , and Jones W. L. , 2014a: Radiometric inter-calibration of SAPHIR using the Microwave Humidity Sounders. 13th Specialist Meeting on Microwave Radiometry and Remote Sensing of the Environment: MicroRad 2014; Proceedings, IEEE, 211–214, 10.1109/MicroRad.2014.6878942.

  • Ebrahimi, H., Datta S. , and Jones W. L. , 2014b: Radiometric inter-calibration of satellite microwave humidity sounders using the GPM Microwave Imager. 2014 IEEE International Geoscience and Remote Sensing Symposium: Proceedings, IEEE, 2566–2569, doi:10.1109/IGARSS.2014.6946997.

  • Elsaesser, G. S., and Kummerow C. D. , 2008: Toward a fully parametric retrieval of the nonraining parameters over the global oceans. J. Appl. Meteor. Climatol., 47, 15991618, doi:10.1175/2007JAMC1712.1.

    • Search Google Scholar
    • Export Citation
  • Farrar, S., Draper D. , Jones L. , and Alquaied F. , 2016: Advantages of calibration attitude maneuvers for spaceborne microwave radiometer missions. 14th Specialist Meeting on Microwave Radiometry and Remote Sensing of the Environment: Proceedings, IEEE, 186–190, doi:10.1109/MICRORAD.2016.7530531.

  • Gaiser, P. W., and Coauthors, 2004: The WindSat spaceborne polarimetric microwave radiometer: Sensor description and early orbit performance. IEEE Trans. Geosci. Remote Sens., 42, 23472361, doi:10.1109/TGRS.2004.836867.

    • Search Google Scholar
    • Export Citation
  • Goldberg, M., and Coauthors, 2011: The Global Space-based Inter-Calibration System. Bull. Amer. Meteor. Soc., 92, 467475, doi:10.1175/2010BAMS2967.1.

    • Search Google Scholar
    • Export Citation
  • Gopalan, K., Jones L. , Biswas S. , Bilanow S. , Wilheit T. , and Kasparis T. , 2009: A time-varying radiometric bias correction for the TRMM Microwave Imager. IEEE Trans. Geosci. Remote Sens., 47, 37223730, doi:10.1109/TGRS.2009.2028882.

    • Search Google Scholar
    • Export Citation
  • Hou, A. Y., and Coauthors, 2014: The Global Precipitation Measurement Mission. Bull. Amer. Meteor. Soc., 95, 701722, doi:10.1175/BAMS-D-13-00164.1.

    • Search Google Scholar
    • Export Citation
  • Imaoka, K., and Coauthors, 2003: Post-launch calibration and data evaluation of AMSR-E. 2003 IEEE International Geoscience and Remote Sensing Symposium, Vol. 1, IEEE, 666–668, doi:10.1109/IGARSS.2003.1293875.

  • John, V. O., Holl G. , Buehler S. A. , Candy B. , Saunders R. W. , and Parker D. E. , 2012: Understanding intersatellite biases of microwave humidity sounders using global simultaneous nadir overpasses. J. Geophys. Res., 117, D02305, doi:10.1029/2011JD016349.

    • Search Google Scholar
    • Export Citation
  • John, V. O., Allan R. P. , Bell W. , Buehler S. A. , and Kottayil A. , 2013: Assessment of intercalibration methods for satellite microwave humidity sounders. J. Geophys. Res. Atmos., 118, 49064918, doi:10.1002/jgrd.50358.

    • Search Google Scholar
    • Export Citation
  • Jones, W. L., Park J. D. , Soisuvarn S. , Hong L. , Gaiser P. W. , and St. Germain K. M. , 2006: Deep-space calibration of the WindSat radiometer. IEEE Trans. Geosci. Remote Sens., 44, 476495, doi:10.1109/TGRS.2005.862499.

    • Search Google Scholar
    • Export Citation
  • Kroodsma, R. A., McKague D. S. , and Ruf C. S. , 2012: Inter-calibration of microwave radiometers using the vicarious cold calibration double difference method. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 5, 10061013, doi:10.1109/JSTARS.2012.2195773.

    • Search Google Scholar
    • Export Citation
  • Kroodsma, R. A., McKague D. S. , and Ruf C. S. , 2016: Vicarious cold calibration for conical scanning microwave imagers. IEEE Trans. Geosci. Remote Sens., doi:10.1109/TGRS.2016.2615552, in press.

    • Search Google Scholar
    • Export Citation
  • Kummerow, C., 1993: On the accuracy of the Eddington approximation for radiative transfer in the microwave frequencies. J. Geophys. Res., 98, 27572765, doi:10.1029/92JD02472.

    • Search Google Scholar
    • Export Citation
  • Kunkee, D. B., Poe G. A. , Boucher D. J. , Swadley S. D. , Hong Y. , Wessel J. E. , and Uliana E. A. , 2008a: Design and evaluation of the first Special Sensor Microwave Imager/Sounder. IEEE Trans. Geosci. Remote Sens., 46, 863883, doi:10.1109/TGRS.2008.917980.

    • Search Google Scholar
    • Export Citation
  • Kunkee, D. B., Swadley S. D. , Poe G. A. , Hong Y. , and Werner M. F. , 2008b: Special Sensor Microwave Imager Sounder (SSMIS) radiometric calibration anomalies—Part I: Identification and characterization. IEEE Trans. Geosci. Remote Sens., 46, 10171033, doi:10.1109/TGRS.2008.917213.

    • Search Google Scholar
    • Export Citation
  • Meissner, T., and Wentz F. J. , 2012: The emissivity of the ocean surface between 6 and 90 GHz over a large range of wind speeds and Earth incidence angles. IEEE Trans. Geosci. Remote Sens., 50, 30043026, doi:10.1109/TGRS.2011.2179662.

    • Search Google Scholar
    • Export Citation
  • Moradi, I., Meng H. , Ferraro R. R. , and Bilanow S. , 2013: Correcting geolocation errors for microwave instruments aboard NOAA satellites. IEEE Trans. Geosci. Remote Sens., 51, 36253637, doi:10.1109/TGRS.2012.2225840.

    • Search Google Scholar
    • Export Citation
  • Moradi, I., Ferraro R. R. , Eriksson P. , and Weng F. , 2015: Intercalibration and validation of observations from ATMS and SAPHIR microwave sounders. IEEE Trans. Geosci. Remote Sens., 53, 59155925, doi:10.1109/TGRS.2015.2427165.

    • Search Google Scholar
    • Export Citation
  • NOAA/NCEP, 2000: NCEP FNL Operational Model Global Tropospheric Analyses, continuing from July 1999 (updated daily). National Center for Atmospheric Research Computational and Information Systems Laboratory Research Data Archive, accessed 12 May 2016, doi:10.5065/D6M043C6.

  • Rosenkranz, P. W., 1998: Water vapor microwave continuum absorption: A comparison of measurements and models. Radio Sci., 33, 919928, doi:10.1029/98RS01182.

    • Search Google Scholar
    • Export Citation
  • Ruf, C. S., 2000: Detection of calibration drifts in spaceborne microwave radiometers using a vicarious cold reference. IEEE Trans. Geosci. Remote Sens., 38, 4452, doi:10.1109/36.823900.

    • Search Google Scholar
    • Export Citation
  • Sapiano, M., Berg W. , McKague D. , and Kummerow C. , 2013: Towards an intercalibrated fundamental climate data record of the SSM/I sensors. IEEE Trans. Geosci. Remote Sens., 51, 14921503, doi:10.1109/TGRS.2012.2206601.

    • Search Google Scholar
    • Export Citation
  • Schaerer, G., and Wilheit T. T. , 1979: A passive microwave technique for profiling of atmospheric water vapor. Radio Sci., 14, 371375, doi:10.1029/RS014i003p00371.

    • Search Google Scholar
    • Export Citation
  • Troyan, D., 2012: Merged sounding value added product. U.S. Dept. of Energy Rep. DOE/SC-ARM-TR-087, 13 pp. [Available online at https://www.arm.gov/publications/tech_reports/doe-sc-arm-tr-087.pdf.]

  • Twarog, E. M., Purdy W. E. , Gaiser P. W. , Cheung K. H. , and Kelm B. E. , 2006: WindSat on-orbit warm load calibration. IEEE Trans. Geosci. Remote Sens., 44, 516529, doi:10.1109/TGRS.2005.863300.

    • Search Google Scholar
    • Export Citation
  • Wang, J. R., Racette P. , and Chang L. A. , 1997: MIR measurements of atmospheric water vapor profiles. IEEE Trans. Geosci. Remote Sens., 35, 212221, doi:10.1109/36.563259.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., 1997: A well-calibrated ocean algorithm for special sensor microwave/imager. J. Geophys. Res., 102, 87038718, doi:10.1029/96JC01751.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., 2015: A 17-year climate record of environmental parameters derived from the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager. J. Climate, 28, 68826902, doi:10.1175/JCLI-D-15-0155.1.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., and Draper D. , 2016: On-orbit absolute calibration of the Global Precipitation Mission Microwave Imager. J. Atmos. Oceanic Technol., 33, 13931412, doi:10.1175/JTECH-D-15-0212.1.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., Ashcroft P. , and Gentemann C. , 2001: Post-launch calibration of the TRMM microwave imager. IEEE Trans. Geosci. Remote Sens., 39, 415422, doi:10.1109/36.905249.

    • Search Google Scholar
    • Export Citation
  • Wilheit, T., 2013: Comparing calibrations of similar conically scanning window-channel microwave radiometers. IEEE Trans. Geosci. Remote Sens., 51, 14531464, doi:10.1109/TGRS.2012.2207122.

    • Search Google Scholar
    • Export Citation
  • Wilheit, T., Berg W. , Ebrahami H. , Kroodsma R. , McKague D. , Payne V. , and Wang J. , 2015: Intercalibrating the GPM constellation using the GPM Microwave Imager (GMI). 2015 IEEE International Geoscience and Remote Sensing Symposium: Proceedings, IEEE, 5162–5165, doi:10.1109/IGARSS.2015.7326996.

  • Yan, B., and Weng F. , 2008: Intercalibration between Special Sensor Microwave Imager/Sounder and Special Sensor Microwave Imager. IEEE Trans. Geosci. Remote Sens., 46, 984995, doi:10.1109/TGRS.2008.915752.

    • Search Google Scholar
    • Export Citation
  • Yang, J. X., and McKague D. S. , 2016: Improving collocation-based scan-dependent intercalibration over the ocean for spaceborne radiometry. IEEE Geosci. Remote Sens. Lett., 13, 589593, doi:10.1109/LGRS.2016.2528261.

    • Search Google Scholar
    • Export Citation
  • Yang, J. X., McKague D. S. , and Ruf C. S. , 2016: Boreal, temperate, and tropical forests as vicarious calibration sites for spaceborne microwave radiometry. IEEE Trans. Geosci. Remote Sens., 54, 10351051, doi:10.1109/TGRS.2015.2472532.

    • Search Google Scholar
    • Export Citation
  • Yang, S., Weng F. , Yan B. , Sun N. , and Goldberg M. , 2011: Special Sensor Microwave Imager (SSM/I) intersensor calibration using a simultaneous conical overpass technique. J. Appl. Meteor. Climatol., 50, 7795, doi:10.1175/2010JAMC2271.1.

    • Search Google Scholar
    • Export Citation
  • Zou, C.-Z., and Wang W. , 2011: Intersatellite calibration of AMSU-A observations for weather and climate applications. J. Geophys. Res., 116, D23113, doi:10.1029/2011JD016205.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    Coverage provided by a single orbit from the GPM microwave constellation for 1 Jan 2015 for (a) conical window channel radiometers and (b) cross-track sounding radiometers.

  • Fig. 2.

    A coincident overpass between GMI (red dots) and DMSP F16 SSMIS (blue dots) from 20 Jan 2015. The Tb values are averaged for each sensor over 1° × 1° grid boxes as indicated by the black squares, and then screened for land, precipitation, sun glint, erroneous data, etc.

  • Fig. 3.

    Differences in observed, simulated, and double differences as a function of scene temperature for the TMI 19.35-GHz (a) V-pol and (b) H-pol channels. The values shown at scene temperatures below 220 K are over ocean, while the values above 280 K are over nonpolarized highly vegetated land surfaces.

  • Fig. 4.

    Intercalibration results for GCOM-W1 AMSR2 vs GMI for the 18.7-GHz (a) V-pol and (b) H-pol channels. Independent results from each of the contributing groups are shown for radiometrically cold scenes (nonprecipitating oceans) and warm scenes (vegetated nonpolarized land). The final composite XCAL calibration adjustment is indicated by the solid black line.

  • Fig. 5.

    Composite XCAL team intercalibration differences between GMI, which is the reference sensor, and the microwave window channel radiometers for both cold ocean scenes (blue bars) and warm vegetated land scenes (red bars).

  • Fig. 6.

    Composite XCAL team intercalibration differences between GMI and the microwave sounding channels for both the DMSP SSMIS sensors and the cross-track humidity sounding sensors. Note that the y-axis range is ±2 K for all instruments except F19, which shows substantially larger biases.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 3827 2079 554
PDF Downloads 6106 622 42