Assessment of the spectral downward irradiance at the surface of the Mediterranean Sea using the radiative Ocean-Atmosphere Spectral Irradiance Model (OASIM)

Paolo Lazzari,Stefano Salon,David Antoine,Elena Terzić,Emanuele Organelli,Vincenzo Vellucci,Watson W Gregg,Fabrizio D'Ortenzio

doi:10.5194/os-17-675-2021

Abstract

Abstract. A multiplatform assessment of the Ocean–Atmosphere Spectral Irradiance Model (OASIM) radiative model focussed on the Mediterranean Sea for the period 2004–2017 is presented. The BOUée pour l'acquiSition d'une Série Optique à Long termE (BOUSSOLE) mooring and biogeochemical Argo (BGC-Argo) float optical sensor observations are combined with model outputs to analyse the spatial and temporal variabilities in the downward planar irradiance at the ocean–atmosphere interface. The correlations between the data and model are always higher than 0.6. With the exception of downward photosynthetic active radiation and the 670 nm channel, correlation values are always higher than 0.8 and, when removing the inter-daily variability, they are higher than 0.9. At the scale of the BOUSSOLE sampling (15 min temporal resolution), the root mean square difference oscillates at approximately 30 %–40 % of the averaged model output and is reduced to approximately 10 % when the variability between days is filtered out. Both BOUSSOLE and BGC-Argo indicate that bias is up to 20 % for the irradiance at 380 and 412 nm and for wavelengths above 670 nm, whereas it decreases to less than 5 % at the other wavelengths. Analysis of atmospheric input data indicates that the model skill is strongly affected by cloud dynamics. High skills are observed during summer when the cloud cover is low.

Highlights

IntroductionThe availability of in situ oceanic radiometric data has recently increased owing to the deployment of autonomous robotic profiling platforms called biogeochemical Argo floats (hereafter referred to as BGC-Argo floats; Johnson and Claustre, 2016; please see Appendix A for a list of abbreviations) equipped with radiometric sensors (Organelli et al, 2016)
The availability of in situ oceanic radiometric data has recently increased owing to the deployment of autonomous robotic profiling platforms called biogeochemical Argo floats equipped with radiometric sensors (Organelli et al, 2016)
The results indicated that, in general, the model reproduced the variability in the spectral downward irradiance in the Mediterranean Sea, which depends on the spatiotemporal scale

Summary

Introduction

The availability of in situ oceanic radiometric data has recently increased owing to the deployment of autonomous robotic profiling platforms called biogeochemical Argo floats (hereafter referred to as BGC-Argo floats; Johnson and Claustre, 2016; please see Appendix A for a list of abbreviations) equipped with radiometric sensors (Organelli et al, 2016). Such data may be exploited to improve the calibration and tuning of the bio-optical models embedded in threedimensional global and regional physical–biogeochemical coupled models. The evaluation of the uncertainty of the multispectral light at the ocean–atmosphere interface is important both for the solution of the radiative transfer model within the water column and for the development of assimilation schemes of radiometric parameters

Methods

Results

Discussion

Conclusion