Calibration Algorithm Resolution and Assessment of the SNPP CrIS Scan Baffle Anomaly using a Model Orbital Scan Baffle Temperature

In May 2023, The Scene Select Module (SSM) scan baffle temperature on SNPP CrIS entered a non-nominal state during which the scan baffle temperature readout dropped to an unphysical low value and remained fixed, despite the actual temperature of the baffle likely being nominal. It is suspected that this is due to either the scan baffle platinum resistance thermometer (PRT) becoming faulty, or due to a component in the surrounding electronics. For SNPP CrIS, the SSM scan baffle is a main contributor of the external environmental model of the Internal Calibration target. Thus, the calculated ICT radiance has a strong dependency on the Scan baffle temperature, and the anomalous fixed temperature value results in a cold radiometric bias with an estimated magnitude of about 0.1 K over the Longwave infrared (LWIR) band and a 0.2 K bias over the SWIR band, and this bias fluctuates with time due to the actual unmeasured scan baffle temperature changing over its orbit as well as slower temperature drifts. To mitigate this bias, a solution is being deployed where a model scan baffle temperature shall be used as a surrogate of the true scan baffle temperature. This model temperature is based on using a mean temperature (averaged from historical temperature data over a year) with an orbital averaged temperature data (averaged over a year for each orbital position), and adding these to the preexisting temperature offset values used in the algorithm; then removing the telemetry input in the algorithm. A straightforward solution using new static temperature offset values using an engineering packet update allows for fast implementation with minimal changes to the existing calibration algorithm. The expected radiometric performance of this algorithm is assessed by comparing the operational SDR data from selected days throughout the year (before the anomaly occurred) to the operational data with a new engineering packet update with the new proxy scan baffle temperature data, and comparing this with the performance of the anomaly emulated for these same days. It was found that the absolute mean radiometric difference between the operational radiances versus the same data with the proxy SSM scan baffle temperature solution is less than 5 millikelvin for the longwave band (compared to as much as 125 millikelvin for the emulated anomaly versus operational data), and less than 10 millikelvin difference for the shortwave band (compared to up to 200 millikelvin for the emulated anomaly versus operational data). In addition, the expected percent difference between the scan baffle temperature with the model scan baffle temperature, as well as changes in the relative phases of these temperature oscillations are assessed using historical SNPP CrIS data. One major caveat of this solution that is noted, and revealed in the radiometric assessment is that, using static model temperature values, this does not take into account seasonal changes in the daily mean SSM temperature, changes in the relative phase in the real vs model SSM temperature over long periods of time, or long timescale drifts in the SSM temperature. Nonetheless, radiometric assessment on the representative days, time series assessment, and scene temperature dependency analysis support the robustness of the solution. Overall, this solution is expected to have a positive impact on the radiometric performance of the SDR data. In addition, this solution is expected to have a positive impact on the CrIS Reprocessed SDR Products, which are important for long-term climate monitoring and assessment.

Disclaimer : The scientific results and conclusions, as well as any views or opinions expressed herein, are those of the author(s) and do not necessarily reflect those of NOAA or the Department of Commerce.