Abstract
The Ozone Mapping & Profiler Suite (OMPS) Nadir Profiler (NP), onboard both the Suomi National Polar-orbiting Partnership (SNPP) satellite and the Joint Polar Satellite System NOAA-20 satellite, is an ultraviolet-visible imaging spectrometer that measures Earth’s albedo by registering terrestrial events to Earth image and extraterrestrial events to Solar spectral images. The NP operates from wavelengths of 250 to 310 nm to profile ozone observations in Earth’s stratosphere. Through years of intensive post-launch calibrations and validations, the Sensor Data Record (SDR) data from both SNPP and NOAA-20 NP sensors demonstrate good performance, meeting the SDR specifications (e.g., Pan et al., 2017). However, it is still challenging to characterize inter-sensor calibration radiometric biases between SNPP and NOAA-20 NP due to non-negligible differences in instrument spectral features, spatial resolution, and temporal resolution. In addition, very narrow NP swath coverage of 250 km not only increases the temporal difference up to 8 days but also extremely reduces the sample size of overlapped observations between the two satellite sensors.
To accurately quantify SNPP and NOAA-20 NP inter-sensor calibration radiometric biases, our study conducts a series of analyses by taking advantage of existing inter-sensor comparison methods, such as the 32-day averaged difference method (Yan et al. 2020), the Simultaneous Nadir Overpass (SNO) method (Cao and Heidinger, 2002), and the deep convective cloud (DCC) method (Wang et al., 2020). Due to limited sample size per geographic location during collocated observations, a 32-day consecutive data set is essential in our zonal mean analysis. Such a data set provides a statistically robust feature for our inter-sensor comparison. The significance is also recognized to ensure the consistency of geographic locations in the data sets between two sensors. The data from two satellites over a similar geographic location need to be simultaneously removed if one satellite observation fails in passing a given quality-control (QC) criterion. Otherwise, resultant mismatches of observations in location can cause large geographic distance differences of over 100 km, further producing many imprecise inter-sensor radiometric bias features. Following those analyses, the impacts of viewing condition discrepancies such as solar zenith angle (SZA) difference on the inter-sensor biases will be assessed using an existing radiative transfer modeling (RTM) to improve the QC criteria in the zonal mean analysis. This is important since the SZA difference can be as large as a couple of degrees between the two NPs’ observations over a similar geographic location. Furthermore, the zonally averaged inter-sensor calibration radiometric bias features will be validated using the RTM and third sensor (e.g., OMPS Nadir Mapper or Visible Infrared Imaging Radiometer Suite) as a transfer, respectively. The validations include some relatively homogenous regions, e.g., the DCC region and open oceans. Finally, we will assess the impacts of SNPP and NOAA-20 NP sensor spectral feature discrepancies in bandpass, radiance sensitivity, wavelength shift, and polarization sensitivity on the inter-sensor biases by using the STAR Vectorized Community Radiative Transfer Model (VCRTM) (Liu and Cao, 2019) and the Algorithm Development Library (ADL) software for OMPS SDR processing.
Characterizing Inter-Sensor Calibration Radiometric Biases between SNPP and NOAA-20 OMPS Nadir Profiler (NP) Sensor Data Records by Using Hybrid Methodologies
The Ozone Mapping & Profiler Suite (OMPS) Nadir Profiler (NP), onboard both the Suomi National Polar-orbiting Partnership (SNPP) satellite and the Joint Polar Satellite System NOAA-20 satellite, is an ultraviolet-visible imaging spectrometer that measures Earth’s albedo by registering terrestrial events to Earth image and extraterrestrial events to Solar spectral images. The NP operates from wavelengths of 250 to 310 nm to profile ozone observations in Earth’s stratosphere. Through years of intensive post-launch calibrations and validations, the Sensor Data Record (SDR) data from both SNPP and NOAA-20 NP sensors demonstrate good performance, meeting the SDR specifications (e.g., Pan et al., 2017). However, it is still challenging to characterize inter-sensor calibration radiometric biases between SNPP and NOAA-20 NP due to non-negligible differences in instrument spectral features, spatial resolution, and temporal resolution. In addition, very narrow NP swath coverage of 250 km not only increases the temporal difference up to 8 days but also extremely reduces the sample size of overlapped observations between the two satellite sensors.
To accurately quantify SNPP and NOAA-20 NP inter-sensor calibration radiometric biases, our study conducts a series of analyses by taking advantage of existing inter-sensor comparison methods, such as the 32-day averaged difference method (Yan et al. 2020), the Simultaneous Nadir Overpass (SNO) method (Cao and Heidinger, 2002), and the deep convective cloud (DCC) method (Wang et al., 2020). Due to limited sample size per geographic location during collocated observations, a 32-day consecutive data set is essential in our zonal mean analysis. Such a data set provides a statistically robust feature for our inter-sensor comparison. The significance is also recognized to ensure the consistency of geographic locations in the data sets between two sensors. The data from two satellites over a similar geographic location need to be simultaneously removed if one satellite observation fails in passing a given quality-control (QC) criterion. Otherwise, resultant mismatches of observations in location can cause large geographic distance differences of over 100 km, further producing many imprecise inter-sensor radiometric bias features. Following those analyses, the impacts of viewing condition discrepancies such as solar zenith angle (SZA) difference on the inter-sensor biases will be assessed using an existing radiative transfer modeling (RTM) to improve the QC criteria in the zonal mean analysis. This is important since the SZA difference can be as large as a couple of degrees between the two NPs’ observations over a similar geographic location. Furthermore, the zonally averaged inter-sensor calibration radiometric bias features will be validated using the RTM and third sensor (e.g., OMPS Nadir Mapper or Visible Infrared Imaging Radiometer Suite) as a transfer, respectively. The validations include some relatively homogenous regions, e.g., the DCC region and open oceans. Finally, we will assess the impacts of SNPP and NOAA-20 NP sensor spectral feature discrepancies in bandpass, radiance sensitivity, wavelength shift, and polarization sensitivity on the inter-sensor biases by using the STAR Vectorized Community Radiative Transfer Model (VCRTM) (Liu and Cao, 2019) and the Algorithm Development Library (ADL) software for OMPS SDR processing.