Skip to Main content Skip to Navigation

Analyse temps-fréquence des données de rayonnement solaire reçu au sol

Abstract : The center of focus for this PhD thesis is the intrinsic temporal variability of the surface solar irradiance (SSI). The characteristic time-scales of variability are revealed by analysing long-term time-series of daily means of SSI, such as ground measurements, satellite estimates, or radiation products from global atmospheric re-analyses, for different geographical locations around the world.To account for the wide range of the time-scales of variability, and given the non-linear and non-stationary nature of the data, the adaptive, data-driven Hilbert-Huang Transform is employed as an analysis tool. The time-varying nature of the characteristic time-scales of variability, along with variations in intensity, are thus revealed.An adaptive fractional re-sampling technique is used to discriminate between the deterministic and the stochastic variability constituents. For all datasets, the deterministic yearly cycle is found to account for the largest part of variability. Furthermore, all time-series are found to contain a high-frequency stochastic variability component, that exhibit cross-scale amplitude modulation by the yearly cycle.A refinement to existing methods for assessing the fitness for use of surrogate SSI products in lieu of ground measurements is also proposed. A case study confirms that satellite estimates outperform re-analyses across all time-scales.
Document type :
Complete list of metadata

Cited literature [197 references]  Display  Hide  Download
Contributor : ABES STAR :  Contact
Submitted on : Thursday, July 2, 2020 - 1:02:10 AM
Last modification on : Wednesday, November 17, 2021 - 12:32:43 PM
Long-term archiving on: : Thursday, September 24, 2020 - 2:21:15 AM


Version validated by the jury (STAR)


  • HAL Id : tel-02887081, version 1


Marc Bengulescu. Analyse temps-fréquence des données de rayonnement solaire reçu au sol. Météorologie. Université Paris sciences et lettres, 2017. Français. ⟨NNT : 2017PSLEM084⟩. ⟨tel-02887081⟩



Record views


Files downloads