Surface solar radiation trends over Europe assessed from ground-based measurements and satellite imagery and their comparison with climate models

<p><span dir="ltr" role="presentation">Downward surface solar radiation (SSR) is the main component in the surface energy balance and </span><span dir="ltr" role="presentation">the climate system, as well as being the fundamental source of energy in various forms of solar and</span> <span dir="ltr" role="presentation">photovoltaic technologies.</span> <span dir="ltr" role="presentation">It is therefore of great importance to know in detail the spatio-temporal </span><span dir="ltr" role="presentation">variation of SSR, as well as its long-term trends. Scientific evidence has shown </span><span dir="ltr" role="presentation">that the amount of solar radiation incident on the Earth&#8217;s surface is not stable over the years, but</span> <span dir="ltr" role="presentation">undergoes significant variations every decade</span><span dir="ltr" role="presentation">. Until recently, ground-based observations have been the most reliable data source for</span> <span dir="ltr" role="presentation">SSR monitoring. Nevertheless, satellite-derived SSR measurements have a better spatial and temporal </span><span dir="ltr" role="presentation">coverage, though the scientific literature on the use of satellite imagery for the study of SSR</span><span dir="ltr" role="presentation"> is still limited.</span></p> <p><span dir="ltr" role="presentation">This study covers several purposes. First, a direct comparison between ground-based observations </span><span dir="ltr" role="presentation">and satellite-derived estimates has been carried out, to determine the capability of the latter to repro</span><span dir="ltr" role="presentation">duce measurements from surface observations. Monthly averaged time series of 108 land stations from</span><span dir="ltr" role="presentation"> GEBA (</span><span dir="ltr" role="presentation">Global Energy Balance Archive</span><span dir="ltr" role="presentation">) dataset (ground observations) have been compared</span> <span dir="ltr" role="presentation">to those estimated from satellite imagery by the Solargis model over the same locations. Solargis is </span><span dir="ltr" role="presentation">a company based in Bratislava, dedicated to the assessment of the solar resource worldwide, </span><span dir="ltr" role="presentation">using GIS (Geographic Information Systems)</span><span dir="ltr" role="presentation">. </span><span dir="ltr" role="presentation">SSR anomalies measured at the surface and estimated from satellite images have been compared over </span><span dir="ltr" role="presentation">Europe for the period 1994-2019</span><span dir="ltr" role="presentation">. Second, multiannual SSR trends have also been calculated in detail </span><span dir="ltr" role="presentation">(station-averaged and station-separated) for both ground-based and satellite-derived datasets, in the </span><span dir="ltr" role="presentation">period of study.</span> <span dir="ltr" role="presentation">Finally, SSR time series have been compared to several CMIP6 (</span><span dir="ltr" role="presentation">Coupled Model </span><span dir="ltr" role="presentation">Intercomparison Project Phase 6</span> <span dir="ltr" role="presentation">) climate models runs.</span></p> <p><span dir="ltr" role="presentation">The results show that the method of estimating SSR from satellite images is able to reproduce </span><span dir="ltr" role="presentation">around 94% of the variability of the SSR measured by ground-based methods in Europe. In addition,</span> <span dir="ltr" role="presentation">trend analysis shows a general increase of SSR over the continent in the period of this study, with </span><span dir="ltr" role="presentation">an average trend of 3.5 Wm^-2</span><span dir="ltr" role="presentation">decade</span>^{<span dir="ltr" role="presentation">-1</span>} <span dir="ltr" role="presentation">for the observational data and 1.7 Wm</span>^{<span dir="ltr" role="presentation">-2</span>}<span dir="ltr" role="presentation">decade</span>^{<span dir="ltr" role="presentation">-1</span>} <span dir="ltr" role="presentation">for the </span><span dir="ltr" role="presentation">satellite estimations. This increase in SSR may be associated with changes in the transmission of the</span> <span dir="ltr" role="presentation">atmosphere due to variations in cloud properties and aerosols.</span> <span dir="ltr" role="presentation">Finally, CMIP6 time series average </span><span dir="ltr" role="presentation">over all models for RCP8.5 scenario shows exactly the same trend as the satellite-derived dataset</span><span dir="ltr" role="presentation">, which suggests there are still some variables not considered by satellite </span><span dir="ltr" role="presentation">imagery methods and climate models.</span></p>