Sensor Stability for SST (3S): Toward Improved Long-Term Characterization of AVHRR Thermal Bands

Abstract

Abstract:Recently, the National Oceanic and Atmospheric Administration (NOAA) performed sea surface temperature (SST) reanalysis (RAN1) from seven AVHRR/3s onboard NOAA-15 to -19 and Metop-A and -B, from 2002-present. Operational L1b data were used as input. The time series of clear-sky ocean brightness temperatures (BTs) and derived SSTs were found to be unstable. The SSTs were empirically stabilized against in situ SSTs using a 90-day moving filter, while the measured BTs were left intact. However, some users are interested in direct radiance assimilation and need stable BTs. Additionally, stabilized BTs will greatly benefit SST (by minimizing the need for their empirical stabilization), and other Level 2 products derived from AVHRR. To better understand the AVHRR calibration and stabilize its BTs, the Sensor Stability for SST (3S; www.star.nesdis.noaa.gov/sod/sst/3s/) system was established at NOAA, which monitors orbital statistics of the sensor measured blackbody temperatures (BBTs), blackbody counts (BCs), and the space counts (SCs), along with the derived calibration gains and offsets. Analyses are performed separately for the satellite night (when the satellite is in the Earth's shadow) and day (on the sunlit part of its orbit). Factors affecting the BBT, BC and SC are also monitored, including the Sun and Moon position relative to the sensor, local equator crossing time, and duration of the satellite night. All AVHRRs show long-term and band-specific smooth changes in the calibration gains and offsets, which are occasionally perturbed by spurious non-monotonic anomalies. The most prominent irregularities occur shortly after the satellite crosses from the night into day, or when it is in a (near) full Sun orbit for extended periods of time. We argue that the operational quality control (QC) and calibration procedures are suboptimal and should be improved. Analyses in 3S suggest that a more stringent QC is needed, and scan lines where the calibration coefficients cannot be derived, due to poor quality SC, BC or BBT data, should be filled in by interpolation from the best parts of orbit or more broadly satellite lifetime. Work is underway to redesign the AVHRR QC and calibration algorithms and create a more stable long-term record of AVHRR calibration and BTs, and use them in the subsequent SST RANs.

Long-Term Variations in the Pixel-to-Pixel Variability of NOAA AVHRR SST Fields from 1982 to 2015

“…From the perspective of this paper, there are three basic components contributing to this variability: electronic noise in the instrument, calibration of the instrument and the digitization of the electronic signal observed by the detector [11]. In some cases [12], NE T represents the variability of the electronic noise in the instrument plus that associated with the calibration, while, in other cases [13], NE T includes the contribution of all three terms. Variability introduced by the digitization of the signal depends on the temperature range measured by the sensor, the temperature of the target (due to the nonlinearity of the Planck Function) and the number of counts into which the temperature range is divided.…”

“…In this section, we examine our p2p σ results in the context of the Pathfinder retrieval algorithm and the NE Ts of the BTs used in the algorithm . We begin by estimating the p2p σ for the AVHRRs on NOAA-16, 17 and 18 based on NE T values available from NOAA's Sensor Stability for SST (3S) system [13]. These three instruments, all AVHRR/3s, were chosen based on the availability of the regression coefficients for the retrieval algorithm (Equation (2)) and our decision to restrict this portion of the analysis to AVHRR/3s for which we have much more data than for the AVHRR/2 instruments, hence the results are much more stable.…”

AVHRR GAC SST Reanalysis Version 1 (RAN1)

“…In particular, deriving SST regression coefficients from RTM and moving away from buoy matchups was explored in the CCI. Our experience with RAN1 and AVHRR L1b analyses in [14] suggest that the current instabilities in the AVHRR sensor calibration should be minimized first, before the advantages of physical SST retrievals can be fully realized. Future inter-comparisons of RAN1 and CCI products will help to quantify the real versus expected improvements due to the use of more physically based retrievals with the current AVHRR L1b data.…”

“…[13]). The root cause of the unstable BTs is suboptimal AVHRR calibration [14]. Work is underway to generate an improved AVHRR Level 1b dataset, and use it in ACSPO RAN.…”

“…In particular, the data in Figure 2 unambiguously support the major recommendation in the NRC (2004) report on satellite climate data records (CDR) [23] which states that high quality fundamental climate data records (FCDRs; the AVHRR BTs) are a prerequisite for the high quality thematic climate data records (TCDRs; the SST). To that end, the current priority is to better understand and resolve the root causes of the unstable AVHRR calibration and generate an improved AVHRR/3 FCDR [14], at which time the need for the variable regression coefficients will be minimized and a more stable SST TCDR (ACSPO RAN) will be produced.…”