Tuesday, 8 January 2019
Hall 4 (Phoenix Convention Center - West and North Buildings)
Two GOES-R Series satellites (renamed GOES-16/17 after reaching the orbit) were launched on November 19, 2016 and March 1, 2018, respectively. The Advanced Baseline Imager (ABI) onboard GOES-16/17 and the Advanced Himawari Imager (AHI) onboard Himawari-8, launched on October 7, 2014, are in the same class and have almost identical thermal spectral bands. The pre-launch characterization and on-orbit calibration algorithm are similar for the thermal bands of these two instruments. The comparison between these two instruments will be very useful for assessing sensor calibration, but is challenged by the fact that they image different regions of the Earth. On the other hand, it is also challenging for the comparison of the sensors on low Earth orbit (LEO) and sun-synchronous satellites, even they have almost identical spectral response, spatial resolution, dynamic range, and calibration algorithm. The Visible Infrared Imaging Radiometer Suite (VIIRS) are on board of SNPP, launched on October 28, 2011, and on JPSS-1 (NOAA-20), launched on November 18, 2017. The two instruments have almost the same design and calibration technique. The inter-comparison between the two VIIRS is also very useful for their calibration assessment. Similarly, MODIS on Terra and Aqua will also be beneficial from their comparison. In general, the LEO-LEO sensor comparison is challenging due to the lack of simultaneous nadir overpasses (SNO). In addition to differences in observation time, other challenging factors include the solar angle and instrument view angle over selected pseudo-invariant sites.
In this work, the inter-comparisons of sensor TEB onboard of geostationary (GEO) satellites are performed over ground targets, using the observations from a LEO satellite as a bridge. Similarly, this method can be applied to LEO-LEO sensor comparison using the double difference approach. Their spectral matching bands can be identified and selected for the double difference. ABI on GOES-16 and AHI on Himawari-8 images the full disk of the Earth every 15 minutes and 10 minutes, respectively. For a selected Earth site, a large amount of samples can be used for comparison. The maximum observation time difference with a LEO sensor is approximately 7.5 minutes. The observation time difference impact on the comparison can be analyzed and, if necessary, the comparison with shorter time difference can be selected. In addition, spatial resolutions for the TEB of these GEO and LEO sensors are fairly close. The TEB spatial resolution is 2 km for ABI and AHI, 750 m for VIIRS, and 1 km for MODIS. The re-sampling, clear sky pixel filtering, and uniform scene selection can also be applied to enhance the comparison accuracy.
GOES-16/17 and Himawari-8 are all positioned over tropical ocean regions. Thus, the comparison with LEO SNO can provide assessments for brightness temperature (BT) measurements over the ocean surface type. Some other pseudo-invariant sites, such as desert site, can also be used to assess their difference with the scene dependence and BT dependence. The comparison can also be extended to simultaneous off-nadir measurement over other types of targets that cover different BT ranges. The view angle effects are analyzed, modeled empirically, and corrected. The demonstration of the comparison using their L1B data will be presented. The comparison between the sensors using same on-board calibrators and calibration algorithms will be very helpful for improvements of calibration and thus enhancements of their L1B product qualities.
In this work, the inter-comparisons of sensor TEB onboard of geostationary (GEO) satellites are performed over ground targets, using the observations from a LEO satellite as a bridge. Similarly, this method can be applied to LEO-LEO sensor comparison using the double difference approach. Their spectral matching bands can be identified and selected for the double difference. ABI on GOES-16 and AHI on Himawari-8 images the full disk of the Earth every 15 minutes and 10 minutes, respectively. For a selected Earth site, a large amount of samples can be used for comparison. The maximum observation time difference with a LEO sensor is approximately 7.5 minutes. The observation time difference impact on the comparison can be analyzed and, if necessary, the comparison with shorter time difference can be selected. In addition, spatial resolutions for the TEB of these GEO and LEO sensors are fairly close. The TEB spatial resolution is 2 km for ABI and AHI, 750 m for VIIRS, and 1 km for MODIS. The re-sampling, clear sky pixel filtering, and uniform scene selection can also be applied to enhance the comparison accuracy.
GOES-16/17 and Himawari-8 are all positioned over tropical ocean regions. Thus, the comparison with LEO SNO can provide assessments for brightness temperature (BT) measurements over the ocean surface type. Some other pseudo-invariant sites, such as desert site, can also be used to assess their difference with the scene dependence and BT dependence. The comparison can also be extended to simultaneous off-nadir measurement over other types of targets that cover different BT ranges. The view angle effects are analyzed, modeled empirically, and corrected. The demonstration of the comparison using their L1B data will be presented. The comparison between the sensors using same on-board calibrators and calibration algorithms will be very helpful for improvements of calibration and thus enhancements of their L1B product qualities.
Acknowledgment: This work is supported by the NOAA/NASA GOES-R project.
Keywords: Inter-comparison, Radiometric calibration, GOES-16/17, SNPP, NOAA-20, Himawari-8, ABI, AHI, MODIS, VIIRS
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner