A comparison of datamining tools for geo-spatial estimation of visibility from AROME-Morocco model outputs in regression framework

Weather forecasting has evolved significantly from the late 1800s and the progress continues. Nevertheless, the numerical weather prediction (NWP) has shown some shortcoming to make cost-effective use of the increase in volume, diversity, and capabilities of observations and environmental products. Indeed, machine learning (ML) techniques are emerging as suitable tools that will have to supplement major components of operational systems. In fact, the number of available data mining tools continues to grow, particularly the open-source ones, due to the raising importance of data mining in today’s highly competitive business environment. In this paper, several, frequently used open-source data mining tools in weather forecasting, are selected and compared in a regression framework. This study aims to evaluate the sensitivity of the ML-developed model to the used platform and algorithm. To achieve this, an estimated visibility product over the north of Morocco, from the operational NWP model AROME-Morocco outputs, has been developed. The performance has been assessed using real data collected at 36 synoptic stations over 2 years. The results analysis for the various algorithms used under the various platforms, shows that the performance of the ensemble-based models is the best whatever the used platform but the model analysis and the best algorithm are very sensitive to it.