Diffuse attenuation coefficient of the photosynthetically available radiation Kd(PAR) for global open ocean and coastal waters

Abstract

Satellite-based observations of the diffuse attenuation coefficient for the downwelling spectral irradiance at the wavelength of 490nm, Kd(490) and the diffuse attenuation coefficient for the downwelling photosynthetically available radiation (PAR), Kd(PAR) in the ocean can play important roles for ocean–atmospheric circulation, biogeochemical, and ecosystem models. Since existing Kd(PAR) models for the satellite ocean color data have wide regional variations, we need to improve the Kd(PAR) algorithm for global ocean applications. In this study, we propose a new blended Kd(PAR) model for both open oceans and turbid coastal waters. The new method has been assessed using in situ optical measurements from the NASA Sea-viewing Wide Field-of-view Sensor (SeaWiFS) Bio-Optical Archive and Storage System (SeaBASS) database. Next, the new method is applied to the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Visible Infrared Imaging Radiometer Suite (VIIRS) to derive Kd(PAR) products, and is compared with in situ measurements. Results show that there are significant improvements in model-derived Kd(PAR) values using the new approach compared to those from some existing Kd(PAR) algorithms. In addition, matchup comparisons between MODIS-derived and in situ-measured Kd(PAR) data for the global ocean show a good agreement with mean and median ratios of 1.109 and 1.035, respectively. Synoptic maps of MODIS- and VIIRS-derived Kd(PAR) data generated using the new method provide very similar and consistent spatial patterns in the U.S. East Coast region, although there are some slight differences between two satellite-derived Kd(PAR) images (~1–5% higher in VIIRS Kd(PAR) compared with those from MODIS-Aqua in the shallow water region), which are possibly due to differences in spectral bands and sensor performance (e.g., calibrations). Monthly maps of VIIRS-derived Kd(PAR) data for the global ocean are also generated using the new Kd(PAR) model, and provide spatial and temporal Kd(PAR) distributions that show consistent results with those from previous studies. Thus, results show that satellite-derived Kd(PAR) data using the new Kd(PAR) model, e.g., from MODIS and VIIRS, can provide more accurate Kd(PAR) data to science communities, in particular, as an important input for ocean–atmospheric circulation, biogeochemical, and ecosystem models.