Cross-calibration of MODIS and VIIRS long near infrared bands

Assessment of long-term change using satellite-borne instruments necessitates consistent merged-sensor datasets, and thereby requires regular efforts to identify and resolve potential sources of sensor-specific biases. Among such sources is a fundamental assumption within atmospheric correction of ocean color radiometry - that satellite-derived remote sensing reflectances (Rrs, the ratio of water leaving radiance to downwelling irradiance) in the visible bands are resilient to moderate errors in the vicarious calibration of the long-wave near infrared band (NIR_L). Recent research suggests that varying the vicarious calibration gain of the NIR_L band may introduce sensor-specific biases in space and / or time. In this work, we assess the calibration of this band across sensors – performed here for MODIS/Aqua and VIIRS/NPP using ‘simultaneous same view’ matchups (SSV). SSV occur during near-simultaneous nadir overpasses (nSNO), with matchups including measurements throughout the sensor swaths’ overlap. Through this research, we assess commonly used geometry, temporal, quality control, and spatial homogeneity metrics used to identify SSV matchups and develop a band-shifting approach applicable within standard satellite data processing routines. Using data from ocean gyres, we calculate that VIIRS L_t(862) are approximately 3% higher than equivalent (band-shifted) measurements from MODIS. As this approach is applicable to other satellite-borne instruments, we anticipate multi-sensor NIR_L cross-calibration may reduce existing discrepancies in the overall satellite ocean color record.