Millores en l'estimació de la radiació solar a la Península Ibèrica. Aportacions metodològiques i comparació entre models

Abstract

Solar radiation is a first-order Essential Climate Variable due to its influence on atmospheric processes, the Earth’s energy balance, and its applications in climatology, ecology, and renewable energy. Unfortunately, in the Iberian Peninsula, knowledge of solar radiation is insufficient due to the limited density of observations and the region’s territorial complexity. This doctoral thesis presents the results of studies that improve its estimation by addressing data quality control, the development of high-resolution digital models, and the integration of satellite information. The networks of the Agencia Estatal de Meteorología (AEMET-Spain) and the Sistema Nacional de Informação de Recursos Hídricos (SNIRH-Portugal), AEMET-SNIRH, are analysed, revealing sparse spatial coverage in AEMET and low temporal completeness in SNIRH. The integration of both networks is carried out to enable the calibration and validation of the thesis’s methodological contributions. A digital model of solar radiation is also developed at 100 m spatial resolution based on the tessellation of a Digital Elevation Model, aiming to enhance the accuracy of solar illumination geometry calculations. The solar vector incidence angles are considered every 30-minutes at the centre of each tile, including intra-annual variations, the components of direct (DIR) and diffuse (DIF) solar radiation, and the cast shadows. This achieves relative differences in solar radiation calculation precision between the central point of each tile and any other point ≤ 1.00 %. The solar radiation model InsolMets has been designed for the 2004–2020 time period, integrating in-situ data and satellite-derived atmospheric variables from Meteosat’s SEVIRI sensor (Cloud Optical Thickness and Cloud Fractional Cover), along with modelling the spatiotemporal variability of the DIF component using an interpolation based on Triangulated Irregular Networks. InsolMets has been implemented in the free MiraMon GIS through the InsolDia application and validated using data from the AEMET-SNIRH network, showing significant improvements in validations compared to models assuming constant atmospheric conditions. Its performance has been compared with other solar radiation models (SARAH, PVGIS, Constant Atmospheric Conditions, Physical Solar Model, and CAMS Worldwide), standing out as the most accurate model. Validations were conducted using error statistics such as Mean Bias Error and Root-Mean-Square Error, based on 141 meteorological stations. These advances highlight the importance of integrating satellite atmospheric remote sensing and advanced spatial modelling to improve solar radiation estimates.

Date of Award	14 Jul 2025
Original language	Catalan
Supervisor	Miquel Ninyerola Casals (Director) & Xavier Pons Fernandez (Director)