ABSTRACT
Cross-mission sensors provide a multitude of synergistic opportunities to improve spatial and temporal coverage by merging their observations. However, discrepancies embedded in the instrumental, algorithmic, and temporal domains must be removed before merging. The main objective of this chapter is to introduce the principles of the cross-mission data merging approach, cross-mission data merging experiments, and associated performance evaluation methods. The basic data merging techniques mainly described in this chapter include:
a quantile-quantile adjustment method for multi-sensor satellite data merging
a machine learning scheme for nonlinear image reconstruction
generation of a spatially complete image by filling data gaps present in the original image
advancements of the high-quality content-based mapping for better environmental surveillance
This chapter includes a demonstration over Lake Nicaragua in Central America for the newly developed Spectral Information Adaptation and Synthesis Scheme (SIASS). In this demonstration, SIASS is designed for generating cross-mission consistent ocean color reflectance via merging 2012–2015 observations from Moderate Resolution Imaging Spectroradiometer (MODIS) and Visible Infrared Imaging Radiometer Suite (VIIRS) to fill in data gaps caused by various factors like cloud contamination. Satellite sensors of MODIS onboard Terra and Aqua, as well as VIIRS onboard Suomi-NPP are collectively used to achieve this goal.
