Spectral effects of atmospheric dust and clouds on estimation of chlorophyll-a concentration from radiation measurements

Abstract

Atmospheric dust and clouds modify the spectral distribution of the incident solar radiation. The influence of these spectral effects on the determination of chlorophyll-a concentration from the sea surface and remotely sensed radiation measurements made from ships, aircrafts or satellites was studied in the region off Northwest Africa in the Atlantic Ocean. The chlorophyll-a algorithms that are typically used in ocean colour community are based on ratios of reflectance values of different wavelengths. This fact, together with the wavelength dependent effect of atmospheric dust and clouds, causes uncertainties in the estimation of chlorophyll-a concentration. The most frequently used Morel, OC4v4 and OC3M chlorophyll-a algorithms were included in the investigation. The highest and the smallest uncertainties were found for the Morel and the OC3M algorithms, respectively. For the first time, sky conditions with dust and clouds together in the atmosphere were studied. Overestimations of the chlorophyll-a concentration of up to 8.8% were observed in the case of dusty skies without clouds. The concentration was also overestimated between 7.6% and 14.3% for skies with mixtures of dust and clouds increasing the solar irradiance. Underestimations of up to 24.1% and overestimations of up to 12.2% were found for skies with mixtures of dust and clouds decreasing the solar irradiance. Compensations of the influences of spectral effects were observed at special ratios of clouds to dust. Earlier results in relation to the impact of clouds on the estimation of chlorophyll-a concentration were verified. The spectral effects of cloudy skies cannot be neglected because it may result in an error of up to 40% of the estimated value.

Acknowledgements

The author would like to thank the Leibniz Institute for Baltic Sea Research and the German Federal Ministry of Education and Research (BMBF) for the support of the SOPRAN (Surface Ocean Processes in the Anthropocene) project (FKZ 03F0662B). The author also thank anonymous reviewers for their helpful comments.