Elsevier

Renewable Energy

Volume 75, March 2015, Pages 21-29
Renewable Energy

Analysis of a solarimetric database for Mexico and comparison with the CSR model

https://doi.org/10.1016/j.renene.2014.09.013Get rights and content

Highlights

  • A solar radiation database from the Mexican National Weather Service was analyzed.

  • Operation and calibration problems and missing data were identified.

  • A subset of 36 stations with reliable data was selected for comparison purposes.

  • Comparison against the CSR satellite model shows 6% RMSD on average.

  • Bias between −7.3% and 8.8% in monthly averaged daily global irradiation.

Abstract

An analysis of the solar radiation database from the network of meteorological stations of the Mexican National Weather Service was carried out. The database includes global irradiance measurements from the oldest 136 stations distributed in the Mexican territory. The consistency of data acquisition from the launch of the stations until 2010 was checked, and visual inspection of graphs of daily irradiance data was carried out, for the first three years operation, to ensure quality and reliability of the data. The results indicate that less than half of the stations have an adequate regularity for data records. With a limited number of selected stations that passed the applied quality criteria, evaluation of hourly and daily global irradiations was carried out. These results were compared to satellite derived data for Mexico, based on NREL's CSR model. The results of the comparison show a good agreement between measured and modeled daily global solar irradiation with an average RMSE of 6.6%. Based on the selected stations, a daily irradiation mean of 5.5 kWh/m²/day is estimated for the country.

Introduction

In recent years the solar industry is experiencing a growth not seen before. Solar energy technologies, both photovoltaic and thermal, are being deployed at accelerated rates around the world. Reliable solar resource information is a crucial input to the successful planning of solar projects, and therefore increased effort is necessary to evaluate this resource where enough quality data is lacking.

It is a common place to say that Mexico has a very strong potential for the utilization of solar technologies due to its high insolation, which is by far its largest renewable energy resource. The country is located in a privileged position in the northern hemisphere, between latitudes 14° 32′ and 32° 43′, including the line of the tropic of Cancer, having nearly 2 million square kilometers area. In early works, it was estimated that average solar irradiance over this territory is higher than the 4.6 KWh/m²/day [1]. However, to date there is a lack of precise knowledge about this resource. Therefore, in order to realize the very important potential of the country, it is essential to gain a better knowledge of the distribution of solar radiation on the territory.

Several maps and tables of solar radiation in Mexico have been published by different authors [1], [2], [3], [4], [5], [6]. However, all of them are based on estimating methods of different types, which have been compared to ground measured solarimetric data for only a handful of locations in the country. Therefore, it is presumed that uncertainties of these models are high [7]. As new and more precise methods are being developed for the indirect evaluation of the solar resource by several researchers [8], [9], [10], [11], [12], there is an increasing need of validated data that can be used for comparison purposes.

The Mexican National Weather Service (SMN; Servicio Meteorológico Nacional) has several networks of sensors to measure climate parameters. Two decades ago SMN initiated an automatic network of atmospheric monitoring stations (EMAS), each of them with a thermopile pyranometer for measuring global solar radiation. The location of these stations is not optimal for the needs of solar resource assessment, because their distribution was made with hydro-meteorological criteria. However, their large number (currently, 136 stations) provides a wide coverage of the different natural regions (Fig. 1) and could fulfill this purpose. Unfortunately, a periodic calibration program was not established for the radiometers in this network, and the equipment has not been calibrated since installation. However, it is to be expected that the first years of operation of each sensor represent reliable information that could be used for the calibration of empirical model parameters related to geographic and climatic conditions for each station [13].

Two years ago, a project to analyze the existing database from the EMAS and to calibrate the radiometers was funded by the Secretaría de Energía (SENER) and the Consejo Nacional de Ciencia y Tecnología (CONACYT). This is a joint project between SMN and the Instituto de Geofísica of Universidad Nacional Autónoma de México (IGF-UNAM), which is recognized as a Regional Radiation Center for the region RA-IV by the World Meteorological Organization, and keeps an absolute radiometer calibrated with the set of radiometers from the world radiation center [14]. One of the main conclusions to date is that the database as a whole shows many problems and inconsistencies. A second project, financed by UNAM is in course, to identify a subset of good quality data from this database that can be used for the purpose of preliminary validation of solar radiation models. In the present work preliminary results from the assessment of the SMN solarimetric database are presented, as well as a comparison with data calculated with the Climatological Solar Radiation model (CSR), which available to the public from the SWERA web site, maintained by NREL [15]. The comparison is made for monthly and yearly averages of daily global horizontal irradiation, for each station.

Section snippets

Analysis of SMN data

The database (from 1998 to 2011) [16] was obtained from a total of 136 solarimetric stations that record global solar irradiance at 10 min intervals. The installed radiometers sensors are mainly Kipp & Zonen CMP11 pyranometers, each one, factory calibrated at the moment of installation. Table 1 shows the names and locations of the stations considered in this paper and Fig. 1 shows the distribution of stations in the Mexican territory. A first step was to identify inconsistent data such as out

Validation of data from SMN

In the initial review of the data contained in the database from SMN, approximately 40 million records of solar radiation were found, from which 45,871 records were found with inconsistent date stamps (0.11%).

Fig. 2 shows the results for 21 stations from the compiled tables. Stations shown are those that had a higher number of data for the time considered in the database, from 1999 to 2011. The graph shows an overall decline in the number of records from the period 2004–2007.

A similar process

Conclusions

A preliminary analysis of the database from the EMAS stations of the SMN was carried out. Large gaps were identified in the database in the period 2004–2007. This problem is generalized in most of stations. From the analysis of daily number of records, for the first three years of operation of every station, those with higher than 80% complete data were identified. The number of stations that fulfilled this requirement was 56 (41%), of a total of 136. After examining the daily irradiance graphs

Acknowledgments

Our thanks to Universidad Nacional Autónoma de México for their financial support through grant PAPITT A100312-2.

Cited by (17)

  • Life cycle assessment of a solar absorption air-conditioning system

    2019, Journal of Cleaner Production
    Citation Excerpt :

    A plot showing the variability of some meteorological parameters is shown in Fig. 4. A full record of ambient temperature and solar irradiance data is available in the solarimetric and meteorological database of the IER-UNAM (ESOLMET-IER, 2018), which is in full agreement with the Mexican National Weather Service (Riveros-Rosas et al., 2015). Because the SAACS is an experimental prototype powered by solar energy, it was not connected to the local electricity grid.

  • A new methodology to extend the validity of the Hargreaves-Samani model to estimate global solar radiation in different climates: Case study Mexico

    2017, Renewable Energy
    Citation Excerpt :

    It is clear that a precise estimation on either case has economical, environmental and technological implications. Unfortunately, although solar radiation is a very important variable, it is only recorded in meteorological and solarimetric stations that are widely dispersed, specially in developing countries [5,6]. The amount of solar energy that reach the earth's surface depends on local conditions, as temperature, humidity, air pollution, visibility, etc., and thus the study of solar radiation under local conditions [7] is required.

  • Estimating daily global solar radiation by day of the year in six cities located in the Yucatán Peninsula, Mexico

    2017, Journal of Cleaner Production
    Citation Excerpt :

    However, understanding daily variations of solar radiation is important for assessing the renewable energy produced by solar energy. Riveros-Rosas et al. (2015) analyzed the solar radiation database of the Mexican National Weather Service and evaluated the daily global solar radiation for 36 select ground stations and compared the data to satellite-derived data for Mexico from the National Renewable Energy Laboratory's (NREL) Climatological Solar Radiation model. He estimated an average daily irradiation of 20 MJ/m2/day for Mexico, but noted that obtaining more conclusive results requires a larger amount of reliable solarimetric data over Mexico (Riveros-Rosas et al., 2015).

View all citing articles on Scopus
View full text