Assessment of monthly global solar irradiation estimates using air temperature in different climates of the state of Rio de Janeiro, Southeastern Brazil

Abstract

The number of weather stations that measure solar global irradiance (I g ) is scarce, and when it is available, it does not present long-time series, without gaps and high quality. When I g is unavailable, it is possible to estimate its integral over time-solar global irradiation (H g )-using empirical methods. However, for a better performance these methods need to be fitted to the local climatic conditions. The aim of this study was to assess the Hargreaves-Samani (HS) and Bristow-Campbell (BC) methods to estimate monthly average daily H g in the state of Rio de Janeiro, Southeastern Brazil, and to propose a simple approach to determine the empirical coefficients in function of the climate. The methods are based on maximum and minimum air temperature and on extraterrestrial solar irradiation. Series of air temperature extremes and H g from 15 automatic weather stations between 2000 until 2013 were used. The methods were evaluated by the statistical indexes: determination coefficient (r 2 ) of the linear regression between observed and estimated monthly H g , root mean square error (RMSE), Willmott's index (d) and performance index (c). The methods (BC-r 2 > 0.60, d > 0.85 and RMSE < 2.99 MJ m −2 d −1 and HS-r 2 > 0.55, d > 0.75 and RMSE < 3.85 MJ m −2 d −1 ) had satisfactory performance in the estimation of monthly H g for the state of Rio de Janeiro, when their coefficients were fitted to local climatic conditions. The BC presented performance classified as "optimal" (c > 0.85) in approximately 80% of the stations analyzed, while for Hargreaves-Samani, only 55% of the stations were classified as "optimal. " The highest HS coefficients (k r ) occurred in Semiarid (0.246 ± 0,023) and Dry Sub-humid (0.181 ± 0.011) climates and were associated with coastal regions (< 20 km), while the stations in Humid (0.146 ± 0.008), Sub-humid (0.1524 ± 0.003) and Dry Sub-humid (0.162 ± 0.011) climates located in interior regions presented the lowest k r . Thus, it is possible to determine the k r coefficient based only on the climatic classification of the site and distance of the coastal environment. In general, the highest atmospheric transmittance (β 0 -BC method) was observed in Semiarid and Dry Sub-humid climate regions. β 1 and β 2 coefficients did not present a distribution pattern with the local climatology and with the proximity of large water bodies. The methods presented a better performance in Dry Sub-humid and Semiarid climates, due to the lower variability of cloudiness and greater thermal amplitude.

Global Solar Radiation Models in Minas Gerais, Southeastern Brazil

“…e meteorological data sets were available simultaneously on both the BDMEP and SINDA platforms for the period from 1999 to 2017. e meteorological data were first submitted to quality control by two steps: basic validation and temporal validation [9,13]. In both steps, when the observation was considered spurious, it was removed from the series.…”

“…Although H s is an important variable, recorded data have been restricted to automatic weather stations (AWSs) [1,3,8], and the number of AWSs capable of registering these data has fallen due to the high costs associated with acquiring, maintaining, and calibrating the instruments [9][10][11][12]. Furthermore, when H s is available in the AWS, it does not present a long time series [13]. ere is a general lack of data coming from most Brazilian AWSs [1,13] even in the most developed regions of the country [14].…”

“…Furthermore, when H s is available in the AWS, it does not present a long time series [13]. ere is a general lack of data coming from most Brazilian AWSs [1,13] even in the most developed regions of the country [14]. For example, according to the Integrated Environmental Data System (Sistema Integrado de Dados Ambientais (SINDA)), of the 61 AWSs in operation in the Minas Gerais (MG) state (Figure 1), only 32 have H s measurement capabilities, and of these 32, only 10 have H s records available (Figure 1).…”

“…is technique has not, however, demonstrated its superiority in estimating H s when compared to traditional EMs [12]. EMs use different functional relationships and include easily measurable variables, such as air temperature, precipitation, and insolation [2,8,[13][14][15]. is makes EMs attractive since these variables are frequently recorded in weather stations.…”

“…It is not new that several studies of this nature have been recently carried out in various locations around the world, including some regions in Brazil, and have been conducted using different models [2,6,7,9,11,13,30]. However, the superiority of one model over the other has not been verified.…”