INTERCOMPARISON BETWEEN PYRANOMETERS TO ESTABLISH A LOCAL REFERENCE STANDARD

Abstract

Continuous solarimetric monitoring stations often experience data interruptions due to operational issues. Some of these stations use multiple sensors to monitor the same parameter, ensuring redundancy and data continuity. However, sensor replacement is often done without considering their differences. This study aims to comparatively analyze irradiance sensors in a precision solarimetric station to evaluate the equivalence of their measurements. We will use sensors classified as Class A and B, according to ISO 9060 standards, with the more accurate sensors serving as the reference. These reference sensors will be validated following the criteria of the World Meteorological Organization's Baseline Surface Radiation Network (WMO/BSRN). Data were collected under clear, cloudy, and mixed sky conditions for six days for each scenario in 2019. Subsequently, these sensors were compared considering three components of solar radiation: Global, Direct, and Diffuse. The results indicated that Class A thermopile sensors have significantly lower uncertainty and variance compared to semiconductor quantum sensors. Differences were observed even among different thermopile sensors, with and without ventilation units. Simple averaging between sensors is insufficient for accurate assessment, as variance showed differences of up to two orders of magnitude. The replacement of one sensor with another should consider variance, especially when choosing the temporal scale. Observation periods with gradually increasing averages at 10, 20, 60, 1440 minutes, or daily tend to progressively smooth out these differences when applied.