RETRACTED: Critical analysis and performance comparison of thirty-eight (38) clear-sky direct irradiance models under the climate of Chilean Atacama Desert. (June 2020)

• Record Type: Journal Article
• Authors: Behar, O.; Sbarbaro, D.; Marzo, A.; Gonzalez, M. Trigo; Vidal, E. Fuentealba; Moran, L.
• Is Part Of: Renewable energy. Volume 153(2020)
• Page Start: 49
• Page End: 60
• Publication Date: 2020-06
• Subjects: Renewable energy sources -- Periodicals; Power resources -- Periodicals; Énergies renouvelables -- Périodiques; Ressources énergétiques -- Périodiques; 333.794
• DOI: 10.1016/j.renene.2019.08.006
• Languages: English
• ISSNs: 0960-1481

Abstract

Abstract : This article has been retracted: please see Elsevier Policy on Article Withdrawal (https://www.elsevier.com/about/our-business/policies/article-withdrawal ). This article has been retracted at the request of the Editor-in-Chief, following concerns raised by a reader. The paper is retracted because of: 1. A preliminary release of Meteorological Radiation Model (MRM) has been used rather than the current version (6.1), which affects the results presented in Table 1. Some inputs required by the MRM code are not used. Therefore, the categorization of the models shown in Table 1 is not correct. 2. The use of 1-min radiation values is uncertain in the sense that short time intervals may contain peculiar circumstances, such as, reflection of radiation by clouds, and short passage of Cumulus clouds. If such measurements are not correctly filtered, they may not be well compared to the simulations obtained with a clear-sky radiation model, which is insensitive to such a dynamic atmospheric occurrence. In this regard, the method selected to detect cloud artifacts is not satisfactory. 3. There is a mistake with the maximum value of P registered at the altitude of z = 965 m amsl (i.e., P = 1046 hPa), which is an extremely high pressure. As the authors do not mention any quality-control procedure for the meteorological data, this creates ambiguities about their precision. Furthermore, filtering the meteorological data by their 5-min averages would avoid extreme values. 4. All solar radiation quality-control tests take into account the pyranometers' cosine-effect problem. Therefore, radiation values of solar altitudes lower than 5o (or even 6o or 10o according to the criterion set) must be excluded. The present study accepted all DNI values from sunrise to sunset. This verifies the erroneous behavior of the models in Figs. 3, 4 for solar altitudes close to or less than 10o. 5. In Table 3, the minimum value of the Linke turbidity coefficient, TL, estimated during the experimental period, is indicated as 0.79-a physical impossibility since, by definition, TL must be larger than 1 6. In Table 3, the minimum value of both ß and AOD is 0, which is an abstraction because it would mean a total absence of aerosols. 7. In Table 4, model performance results are deemed "excellent" if their coefficient of determination (R2) is <2.5%, which is unsubstantiated. 8. In Table 5, many R2 values are found >100%, which is impossible. 9. The main problem of the paper is that the method used to obtain the input data to the models is circular, i.e., directly derived from the very same irradiance data then used to "validate" the models' irradiance predictions. This invalidates the method and the results presented. These affect the reliability of the analysis presented in the paper, so the paper is retracted.