Long‐term stability of meteorological temperature sensors

Abstract

This study presents the results of successive calibrations of sensors for meteorological applications consecutively exposed to particular environmental conditions. The determined sensor characteristics and measurement uncertainties were used to analyze how the environmental factors and the ageing of the sensor (impact of the time) results in the variation of characteristics in terms of the sensors’ stability and the reproducibility of the calibration curves obtained. The study consists of the regular calibrations of two temperature sensors: HMP 155 and Vaisala's HMP 45AC, after having been exposed to thermal shocks and exposure to extreme ranges of humidity. The selected measuring devices represent sensors commonly used in Stevenson Screens, compact solar shields and other meteorological applications. The investigation verifies the following impacts on characteristics’ change and reproducibility: time, high humidity, high and low temperatures, and rapid temperature changes (10 series) from −30 to −10°C typical for anticyclonal conditions. The result of the measurements is presented and discussed. The analysis is completed by fully understanding, evaluating and reporting uncertainties in the results.

Temperature extreme records: World Meteorological Organization metrological and meteorological evaluation of the 54.0°C observations in Mitribah, Kuwait and Turbat, Pakistan in 2016/2017

“…Their drift is normally very low but annual recalibration is recommended for reference climate observations. Such sensors are moreover also quite stable at temperature changes in the range −40 to +60°C and do not present significant hysteresis, as also studied in MeteoMet (Kowal et al, ). Despite such performance and due to the importance of the measurement under investigation, since the HMP 155 involved in this investigation arrived at INRiM for calibration about 1 year after having made the record measurement, it is important to make an evaluation of the drift of the sensor, if any, and the associated uncertainty.…”

Metrological evaluation of the effect of the presence of a road on near‐surface air temperatures

“…As a matter of fact, the uncertainty evaluation on such biases, at the current level of knowledge, would be basically impossible. From a data analysis point of view, this brings in also other effects not strictly related to the RSE, for example, errors related to fast temperature transients; in fact, each sensor responds with a different time constant to the temperature rise produced by a fast and strong variation in radiation (Kowal et al ., 2020). On the other hand, daily averages on a filtered dataset would yield inconsistent results given the non‐continuous and non‐homogeneous nature of the dataset.…”

Thermal metrology for climate: a review of projects, activities and open issues

“…The MeteoMet consortium has grouped close to 100 participants: almost all National Metrology Institutes (NMIs) in Europe took part as funded partners, more than 10 Universities, 20 among public and private Research Institutes, several private companies and manufacturers in the status of collaborators, International Institutions (the WMO, the GCOS, the Ny-Ålesund Science Committee, about 20 NMHSs and others). MeteoMet accounted for around 300 identified deliverables: new laboratory and transportable calibration systems [4][5][6][7], improved knowledge in scientific investigation of environmental quantities and their effects on instruments [8,9], on site campaigns also in remote areas such as high mountains and the Arctic [10,11], procedures and intercomparisons, technical advances in sensors and measurements methods [12][13][14], evaluation of uncertainties and instrumental in climate and meteorological measurements and data series [15][16][17], evaluation of climate records [18], discussion on reference grade measurements [19,20], training and dissemination.…”

“…This variability of characteristic affects a significant impact on the temperature measurement accuracy, and it becomes fundamental to evaluate the amplitude of this shift, the time of occurrence and try to determine causes of the change. A study was carried out over a two years period, to investigate the factors affecting meteorological air temperature thermometers during normal operational work, which have an influence on sensor characteristic variability and uncertainty [13]. The most of measurements were carried out at the Polish Institute of Low Temperature and Structure Research (INTiBS).…”