The worldwide growing development of PV capacity requires an accurate forecast for a safer and cheaper PV grid penetration. Solar energy variability mainly depends on cloud cover evolution. Thus, relationships between weather variables and forecast uncertainties may be quantified to optimize forecast use. An intraday solar energy forecast algorithm using satellite images is fully described and validated over three years in the Paris (France) area. For all tested horizons (up to 6 h), the method shows a positive forecast skill score compared to persistence (up to 15%) and numerical weather predictions (between 20% and 40%). Different variables, such as the clear-sky index (Kc), solar zenith angle (SZA), surrounding cloud pattern observed by satellites and northern Atlantic weather regimes have been tested as predictors for this forecast method. Results highlighted an increasing absolute error with a decreasing SZA and Kc. Root mean square error (RMSE) is significantly affected by the mean and the standard deviation of the observed Kc in a 10 km surrounding area. The highest (respectively, lowest) errors occur at the Atlantic Ridge (respectively, Scandinavian Blocking) regime. The differences of relative RMSE between these two regimes are from 8% to 10% in summer and from 18% to 30% depending on the time horizon. These results can help solar energy users to anticipate—at the forecast start time and up to several days in advance—the uncertainties of the intraday forecast. The results can be used as inputs for other solar energy forecast methods.
This study focuses on the water cycle around the Northern seasonal polar cap from the end of autumn to the following spring season, and more precisely on the progression and retreat of CO2 and H2O frosts observed by the Martian Global Climate Model (GCM) of the LMD and by the OMEGA imaging spectrometer onboard Mars Express.Based on a series of OMEGA observations from the end of autumn of MY 27 (Ls ~260°) to the end of spring of MY 28, Appéré et al. (2011) described the temporal evolution of H2O and CO2 ice deposits, constantly evolving northwards through sublimation and deposition of the corresponding ice/frost. This ends just before the summer solstice (around Ls ~70°) after the complete disappearance of CO2 ice. At high latitudes, the sublimation of frost then contributes to an abundant emission of water vapor.The LMD Martian GCM is able to reproduce the global and seasonal water and CO2 cycles during the winter-spring seasons. However, it releases excessive humidity in the polar region. In order to improve the model, we examine and compare the southernmost position of frosts and their poleward progression on Martian GCM data and on spectral images from OMEGA.In OMEGA data, water and CO2 frosts can be detected by absorption bands at 1.5 μm, respectively at 1.43 μm (Langevin et al., 2007). Similarly, when the depth of the absorption band falls below a chosen value, the frost is considered as having disappeared. On one orbit-segment image, the southernmost pixels form a more or less continuous line corresponding to the frost boundary (“crocus-line” type).In the model simulation, we use the surface ice contents provided by the LMD GCM (Forget et al., 1999) in order to detect the frost dissipation. Water (resp. CO2) ice content values (in kg/m2) have been calculated on a regular grid (5.625° longitude x 3.75° latitude) 4 times per sol (at 0, 6, 12 and 18 h LT) over one Martian year. Starting at the end of the northern autumn (Ls ~ 260°), the evolution of the water (CO2) ice content can be examined at every grid point.In most cases, all the OMEGA pixels of an image are observed at the same local time. We calculate an average GCM frost dissipation time Lsfd_GCM from the 4 closest GCM neighbor grid points, weighted by the distance between each GCM grid point and the OMEGA frost line. Then the time interval between the dissipation of frost in OMEGA water (CO2) ice absorption depth profile and in the collocated (interpolated) water ice disappearance on the GCM can be determined.With a perfect GCM and well-chosen frost-detection thresholds on both datasets, the dissipation of frost should be simultaneous for collocated data in both datasets. Otherwise, when the frost time dissipation interval DLsfd = Lsfd_OMEGA - Lsfd_GCM is positive (respectively negative), the model is late (in advance) w.r.t. observations. We will present results of the evolution of the frost time dissipation during the winter-spring season.