Validation of reanalysis and satellite imagery data with four ground stations in Salta-Northern Argentina for a seven-year period

SARMIENTO, NILSA; DELLICOMPAGNI, PABLO; BELMONTE, SILVINA; FRANCO, JUDITH; LOPEZ, EMILCE

Viena

Conferencia; EGU-2018; 2018

Copernicus

The geographical area for this study is the province of Salta, located in the northwest of Argentina. The total areaof Salta is 155.488 km2 and has a total resident population of 1.214.441 inhabitants (population density is 7.8inhabitants/km2).Firstly, the importance of the solar energy potential in Salta has been highlighted by the government in itsProvincial Renewable energy Plan and by worldwide researchers. Yet solar radiation estimations are availablemainly in low spatiotemporal resolutions and studies that characterize the temporal variability are scarce.Secondly, to design and evaluate solar energy devices is highly-recommended to count with ground measuredhistorical data of global solar radiation from calibrated and maintained meteorological stations. However,high costs of installation, maintenance and communication to centralize the data have promoted the developmentof different models to estimate the global solar radiation from available data and the use of solar radiation dataderived from satellite imagery.For both reasons, in this work we evaluate global solar radiation from the ERA5 reanalysis data set providedby the European Centre for Medium-Range Weather Forecasts (ECMWF) and the Meteosat secondgeneration derived data set provided by Land Surface Analysis Satellite Applications Facility (LSA-SAF). Theseare compared with data from four pyranometers distributed around the region in a period of seven years.ERA5 is a climate reanalysis dataset developed by ECMWF and it covers the period 1950 to present.However, is still in production and is expected to be available for use by early 2019. The first batch of ERA5,covering the period 2010 to 2016, was released in July 2017.The variable downloaded from the ERA5-ECMWF is Surface Downwelling Shortwave Flux in Air, thevalues are saved as the amount of solar radiation that reaches the atmosphere, accumulated since the beginningof each prediction period (which corresponds to the calendar days), having a temporal resolution of one hour andapproximately 30 kilometers of spatial resolution.The variable downloaded from LSA-SAF is Down-welling Surface shortwave flux (DSSF),refers to the radiativeenergy in the wavelength interval [0.3µm, 4.0µm] reaching the Earth?s surface per time and surface unit, ithas a temporal resolution of 30 minutes and approximately 4 km of spatial resolution.The European organization for the Exploitation of Metrological Satellites (EUMETSAT) Satellite ApplicationFacility (SAF) on Land Surface Analysis (LSA) provides value-added products for the meteorological andenvironmental science communities with applications in the fields of land surface modelling, numerical weatherprediction, hydrology, climatology and renewable energy resources assessments.An analysis of variance (ANOVA) is performed to determine whether there are any statistically significantdifferences between the means of the three independent data sets. The comparison is performed for monthlyaccumulated radiation. The results show that there are no statistically significant differences between the sources.However the differences between ERA5 and ground measurements are closer than LSA-SAF data and groundstations.