Multi-source Data Fusion for Climate Variation Study—Case Study: Algeria

Further to the recent technological advances in physics and computer science researches, the implementation and deployment of solar energy plants become more and more efficient to face the exhaustion of energy sources fossils. The feasibility and studies of solar energy projects are performing by the processing of solar irradiation parameter. However, the solar irradiation data sources are in most cases unavailable in hostile or inaccessible areas making a new paradigm of solar data availability in proximity to the point of interest. Meanwhile, the researchers had developed models for the solar irradiation estimation based on different meteorological parameters like: sunshine, temperature, and cloud-cover. These terrestrial meteorological data sources represent a real mine of multi-modal and correlated data that could be fused in order to build locally a global climatic database. In this paper, we introduce a low-level data fusion approach based on multi-source and multi-modal meteorological parameters followed by a classification and mapping process of climate clusters with the objective of facilitating the identification of regions with high variation potential in Algeria.