Improved scheme for Cross‐track Infrared Sounder geolocation assessment and optimization

Abstract

An improved scheme for Cross‐track Infrared Sounder (CrIS) geolocation assessment for all scan angles (from −48.5° to 48.5°) is developed in this study. The method uses spatially collocated radiance measurements from the Visible Infrared Imaging Radiometer Suite (VIIRS) image band I5 to evaluate the geolocation performance of the CrIS Sensor Data Records (SDR) by taking advantage of its high spatial resolution (375 m at nadir) and accurate geolocation. The basic idea is to perturb CrIS line‐of‐sight vectors along the in‐track and cross‐track directions to find a position where CrIS and VIIRS data matches more closely. The perturbation angles at this best matched position are then used to evaluate the CrIS geolocation accuracy. More importantly, the new method is capable of performing postlaunch on‐orbit geometric calibration by optimizing mapping angle parameters based on the assessment results and thus can be further extended to the following CrIS sensors on new satellites. Finally, the proposed method is employed to evaluate the CrIS geolocation accuracy on current Suomi National Polar‐orbiting Partnership satellite. The error characteristics are revealed along the scan positions in the in‐track and cross‐track directions. It is found that there are relatively large errors (~4 km) in the cross‐track direction close to the end of scan positions. With newly updated mapping angles, the geolocation accuracy is greatly improved for all scan positions (less than 0.3 km). This makes CrIS and VIIRS spatially align together and thus benefits the application that needs combination of CrIS and VIIRS measurements and products.

Intercomparison Between VIIRS and CrIS by Taking Into Account the CrIS Subpixel Cloudiness and Viewing Geometry

“…Compared with results without the geolocation correction, the STDs for all scenes are reduced by 18%, 38%, 39%, and 38% for M13, M15, M16, and I5, respectively, but there is little change in the radiance biases. From the last column in Figure , the STD for all scenes without the uniformity test is 0.27 K for I5, which is comparable to the results from Figure 15 in Wang, Zhang, et al (). This result indicates that the simplified correction method used in this study is able to mitigate the geolocation errors identified by Wang, Zhang, et al ().…”

“…The geolocation correction is important for VIIRS/CrIS intercomparisons (Tobin et al, ; Wang, Zhang, et al, ). As mentioned, the results presented in Figure are after the simplified geolocation error correction.…”

“…Because the time period selected for this study has not been corrected for such errors, we have implemented a simplified correction scheme. While Wang, Zhang, et al () corrects the CrIS geolocation so that it has no geolocation error compared with VIIRS, our simplified scheme focuses on finding the correct collocated VIIRS pixels for each CrIS FOV. Take CrIS field‐of‐regard (FOR, each FOR contains 3 by 3 FOVs) 30 as an example.…”

“…Other FORs are processed in a similar way, in both cross‐track and along‐track directions. The estimated geolocation errors shown in Figure 8 in Wang, Zhang, et al () are used to determine how many VIIRS pixels to shift and which direction to shift, for each CrIS FOR. Note that, as both fly on SNPP, the VIIRS and CrIS have the same directions in the along‐track axis, but opposite scan directions in the cross‐track axis (Wang et al, ).…”

“…Along with the Atmospheric InfraRed Sounder (AIRS; Chahine et al, ) on the Aqua satellite and the Infrared Atmospheric Sounding Interferometer (IASI; Chalon et al, ; Collard & McNally, ) on the European MetOp‐A/B satellites, CrIS provides needed information about global atmospheric profiles with high vertical resolution for numerical weather prediction models (Cucurull et al, ; Joo et al, ; Wang, Li, et al, ). Calibration and validation of the CrIS Sensor Data Records (SDRs) are performed before (Zavyalov et al, ) and after the launch (Han et al, ; Zhou et al, ) in order to determine the instrument noise characteristics (Zavyalov et al, ), the spectral/radiometric performance (Strow et al, ; Tobin, Revercomb, Knuteson, Best, et al, ; Tobin, Revercomb, Knuteson, Taylor, et al, ; Wang et al, ), and the geolocation accuracy (Wang et al, ; Wang, Zhang, et al, ).…”