ISO/DIS 8932-3:2025(en)

ISO/TC 146/SC 5/WG 11

Secretariat: DIN

Date: 2025-01-29

Meteorology – Radiosonde – Part 3: Laboratory test method for solar radiation error of temperature sensor in radiosonde

© ISO 2025

All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below or ISO’s member body in the country of the requester.

ISO copyright office

CP 401 • Ch. de Blandonnet 8

CH-1214 Vernier, Geneva

Phone: +41 22 749 01 11

Email: copyright@iso.org

Website: www.iso.org

Published in Switzerland

Contents

Foreword

ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies (ISO member bodies). The work of preparing International Standards is normally carried out through ISO technical committees. Each member body interested in a subject for which a technical committee has been established has the right to be represented on that committee. International organizations, governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.

The procedures used to develop this document and those intended for its further maintenance are described in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the different types of ISO documents should be noted. This document was drafted in accordance with the editorial rules of the ISO/IEC Directives, Part 2 (see www.iso.org/directives).

ISO draws attention to the possibility that the implementation of this document may involve the use of (a) patent(s). ISO takes no position concerning the evidence, validity or applicability of any claimed patent rights in respect thereof. As of the date of publication of this document, ISO had received notice of (a) patent(s) which may be required to implement this document. However, implementers are cautioned that this may not represent the latest information, which may be obtained from the patent database available at www.iso.org/patents. ISO shall not be held responsible for identifying any or all such patent rights.

Any trade name used in this document is information given for the convenience of users and does not constitute an endorsement.

For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and expressions related to conformity assessment, as well as information about ISO's adherence to the World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT), see www.iso.org/iso/foreword.html.

This document was prepared by Technical Committee ISO/TC 146, Air quality, Subcommittee SC 5, Meteorology.

A list of all parts in the ISO 8932 series can be found on the ISO website.

Any feedback or questions on this document should be directed to the user’s national standards body. A complete listing of these bodies can be found at www.iso.org/members.html.

Introduction

Temperature and water vapour (i.e., humidity) are two of the basic atmospheric state variables and are important for developing weather and climate prediction models. To measure the temperature and humidity in upper-air, radiosondes are generally used. Radiosonde is an instrument intended to be carried by a balloon up through the atmosphere, equipped with devices to measure one or more meteorological variables such as pressure, temperature, humidity, and are provided with a radio transmitter for sending this information to an observing station.^[1] Radiosonde observations are often used in concert with other measurement techniques such as remote sensing using satellites to provide comparative data. For radiosonde-derived data to be useful, the measurement accuracy of radiosoundings is needed. From a metrological perspective, this measurement accuracy should be expressed in terms of uncertainty that is traceable to the International System of Units (SI).

Correction for solar radiation effects is critical for precisely measuring the actual temperature in the upper air. Since the temperature sensor is exposed to the outside air, the temperature measured by the sensor will be different from the air temperature due to solar radiation impacting the sensor. During daytime soundings, solar radiation heats the sensor. Therefore, all radiosonde temperature measurements show positive value radiation-induced errors in daytime. This is one of the main errors decreasing the accuracy of temperature measurements; thus, the measured temperature in soundings requires correction based on an algorithm or look-up table supplied by manufacturers.

There are many parameters that affect the radiation correction in soundings such as radiation flux, air temperature, air pressure, ventilation speed, solar elevation angle, and radiosonde orientation/motion with respect to sun. The radiative heating of sensors increases with a decrease in temperature (T), pressure (P) and ventilation speed (v), and, in general, the radiation correction increases as the radiosonde ascends. The decrease in these parameters (T, P, and v) results in a reduced heat transfer from the sensor to the air. With respect to the impact of ambient temperature on sensors, the reduction in thermal conductivity of air at low temperatures contributes to lowering the heat transfer coefficient; thereby, resulting in an increased radiation correction at such temperatures.^[2] The thermal conductivity of air plays a crucial role in influencing heat transfer at the boundary between the air and the sensor.

In radiosoundings, cloud cover and distribution as well as surface albedo also affect solar radiative heating. These conditions change in real time and are difficult to model in manufacturer’s correction algorithms; thus, introducing uncertainty. The radiation correction is a complicated process which depends on multiple factors and parameters as well as different radiation correction algorithm features as applied by various manufactures which are proprietary and rarely disclosed. This document provides a testing method for measuring the radiation error for integrated radiosonde temperature sensors, using a limited number of samples from a batch of mass produced sensors.

The Standing Committee on Measurements, Instrumentation and Traceability (SC-MINT), one of the Technical Commissions and Research Board at the World Meteorological Organization, –– offers advice, recommendations, and promotes studies on the effective and sustainable use of instruments, such as radiosondes as well as offering methods for upper air weather observations. The SC-MINT Guide recommends that for temperature correction, solar radiation correction should be applied using software during data processing.^[1] Consideration of additional heating sources, the temperature sensor and supporting hardware are designed such that solar heating does not vary significantly as the radiosonde rotates in flight relative to the sun.

Despite recognising the importance of correcting sensor based meteorological temperature measurement, there is limited information in the SC-MINT Guide or other technical documents on test methods, procedures and related instrument details. Therefore, there is a need to publish a consensus procedure for evaluating radiosonde temperature error induced by solar radiation in an experimentally controlled way.

The procedure presented in this document provides laboratory setup technical requirements for conducting solar radiation test, test procedure for determining solar radiation error on radiosondes with various environmental and geometrical parameters, and a method for assessing uncertainties within test results. Temperature, pressure, ventilation, and radiation flux, as well as geometric parameters including sensor boom tilt angle and light illumination angle are evaluated over ranges that are varied to mimic in-flight conditions as experienced by radiosondes.

Note that when considering uncertainty in soundings, other factors such as temperature spikes due to patches of warm air coming off the sensor housing and the balloon, time-lag, and albedo should also be considered, as summarized in Table 2 of Reference [3]. While all uncertainty terms affecting the results should be considered, this procedure specifically focuses on testing environmental and geometrical effects on the radiation correction.

The core procedure discussed in this document involves the SI-traceable generation of air ventilation speed in the test environment to simulate the ascendance speed of radiosondes. In addition, the solar irradiance, temperature, and pressure in the test cell should closely replicate those observed in upper air. To attain the required range variability of atmospheric parameters in the test setup, this testing method employs both an open suction-type wind tunnel and a closed-type wind tunnel as illustrative examples. These methods are chosen because of their traceability to the SI and validation by metrological and meteorological experts in testing radiosonde temperature sensors^[2,4].

Note that while the open suction-type wind tunnel using a combination of sonic nozzles and vacuum pump is based on Korean and US patents,^[5,6] the patent holder has granted a license, free of charge, to an unlimited number of applicants globally, without discrimination, and under reasonable terms and conditions. This license allows for the creation, use, and sale of implementations based on this ISO document. However, it's crucial to clarify that this ISO document does not detail the patented technique and is not an endorsement for use of patented material. Alternative equivalent procedues, employing different types of wind tunnel systems, can also be used to determine air ventilation speed at varying temperatures and pressures if they are SI-traceable and meet the data quality objectives of the application.

Meteorology – Radiosonde – Part 3: Laboratory test method for solar radiation error of temperature sensor in radiosonde

The document specifies a test method for estimating the magnitude of radiosonde temperature sensor warming, induced by direct solar radiation, based on variations in air pressure, temperature, ventilation speed, tilt angle of its supporting sensor boom, and light illumination angle on the boom through a laboratory evaluation. This document will describe the following:

a) Technical requirements for a laboratory setup to measure the effect of direct solar radiation on radiosonde temperature measurement under simulated sounding conditions;

b) A test procedure for estimating radiosonde temperature measurement errors due to direct solar radiation in the air pressure range of 3 hPa ~ 1,000 hPa, temperature range^[1] of −70 °C ~ 50 °C, ventilation speed range of 3 m∙s^-1 ~ 7 m∙s^-1 at a predefined irradiance (e.g. 1,000  or higher), sensor boom tilt range^[2] from 0° to 45° with respect to the air ventilation direction and the range of light illumination^[3] angle from 0° to 90° with respect to the sensor boom plane;

c) A method to evaluate uncertainty in the results under the test conditions.

The essential components of the laboratory setup are a climate chamber, wind-tunnel, a test cell, thermometers, pressure and vacuum gauges, a laser anemometer and a solar simulator. These components are summarised in Clause 5. Clauses 6–8 provide details on test preparation, the procedure for installing a radiosonde in the test cell, the operation of the laboratory setup, the experimental range and sequence, and data processing. In Clauses 9 and 10, a method to evaluate and report uncertainty of the determined radiation errors on the temperature using the uncertainty propagation law, based on a mathematical model, is proposed.

NOTE 1 Since the test method is limited by the use of ground-based facilities, radiative cooling effect of radiosonde temperature sensors observed in stratosphere may not be reproduced and represented in the test result.

NOTE 2 The light source spectrum may be limited in the infrared (IR) region compared to that of the visible light solar spectrum. As recommended by the WMO guide by SC-MINT,^[1] the heat exchange in infrared radiation (IR) needs to be avoided by using sensor coatings that have low emissivity in the IR. Otherwise, the effect of IR can be underestimated in the test.

NOTE 3 Due to potential limitations in the number of test setups or laboratories capable of conducting this test, peer-reviewed reports or papers published online or offline resulting from research activities conducted by academia or meteorological institutes can be utilized as a test report when following this test procedure.

The following documents are referred to in the text such that some or all of their content constitutes requirements under this document. For dated references, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies.

ISO/IEC 17025:2017, General requirements for the competence of testing and calibration laboratories

ISO/IEC Guide 98‑1:2009, Uncertainty of measurement – Part 1: Introduction to the expression of uncertainty in measurement

ISO/IEC GUIDE 98‑3:2008, Uncertainty of measurement — Part 3: Guide to the expression of uncertainty in measurement (GUM:1995)

ISO/IEC Guide 99:2007, International vocabulary of metrology — Basic and general concepts and associated terms (VIM)

WMO No.182, 1992, International Meteorological Vocabulary

ISO 17713‑1:2007, Meteorology — Wind measurements — Part 1: Wind tunnel test methods for rotating anemometer performance

ISO 15387:2005, Space systems — Single-junction solar cells — Measurements and calibration procedures

IEC 60050‑713:1998, International Electrotechnical Vocabulary (IEV) – Part 713: Radiocommunications: transmitters, receivers, networks and operation

IEC 60904‑9:2020, Photovoltaic devices – Part 9: Classification of solar simulator characteristics

ISO 9300:2022, Measurement of gas flow by means of critical flow nozzles

The following terms and definitions are specific to this standard.

ISO and IEC maintain terminology databases for use in standardization at the following addresses:

— ISO Online browsing platform: available at https://www.iso.org/obp

— IEC Electropedia: available at https://www.electropedia.org/

3.1

radiosonde

instrument intended to be carried by a balloon through the atmosphere, equipped with devices to measure one or several meteorological variables (such as pressure, temperature, humidity), and provided with a radio transmitter for sending this information to the observing station

Note 1 to entry: A series of battery-powered telemetry devices that are suspended high in the atmosphere using a weather balloon, which measure atmospheric parameters, such as temperature and humidity, and transmit data to a ground system using radio frequencies.

Note 2 to entry: To ensure the traceability of radiosondes as a product under WMO control, it is necessary to obtain the WMO code (BUFR code 0-02-011). This WMO code must be unique for each radiosonde, and a single code must not be used for more than one type of radiosonde.

Note 3 to entry: The current requirements for threshold, breakthrough, and goal uncertainty in upper-air temperature measurements are specified in Observing Systems Capability Analysis and Review tool (OSCAR),^[7] which is a resource developed by WMO.

3.2

radiosonde body

housing of a radiosonde comprising a circuit board with measurement chips and a data transmission section with antennae and batteries

3.3

radiosonde sensor boom

boom connected to the radiosonde body to which the temperature and humidity sensors are attached

3.4

radiosonde installation chamber

chamber in which the body of a radiosonde can be installed

3.5

test cell

chamber in which a radiosonde sensor boom with temperature sensors can be installed. The test cell must have one or more transparent window(s) allowing the transmission of simulated solar irradiance.

3.6

climate chamber

chamber or enclosed space where the internal temperature or temperature and humidity can be controlled within specified limits

Note 1 to entry: The climate chamber can be used to control the temperature of the radiosonde in a test cell.

Note 2 to entry: For huge setups, the laboratory itself, appropriately controlled at room temperature, can play the role of a climate chamber.

3.7

dry air generator

device that can generate dry air below the frost point of –80 °C. Dry air is used as a medium of ventilation in a test cell.

3.8

heat exchanger

device that efficiently transfers heat from a high-temperature object to a low-temperature one. A tubing or pipe with heat exchange pins can be used to adjust the dry air temperature to the target temperature for a test. The heat exchanger should be installed upstream of the test cell in the climate chamber.

3.9

liquid bath

equipment in which liquid is present in a container, with a specific volume, to maintain a constant temperature. When the test pressure and ventilation speed are high, the temperature of dry air should be preadjusted before entering the climate chamber. An additional heat exchanger in the liquid bath can be used to stabilise the temperature of the test cell in the climate chamber.

Note 1 to entry: Liquid whose freezing point is lower than the lowest temperature of the testing should be used for the bath. In general, ethanol, silicon oil or a halocarbon is used.

3.10

reference thermometer(s)

device used to measure the reference temperature in the test cell. Generally, calibrated platinum resistance thermometers are used with the resistance read by a calibrated digital multi-meter.

3.11

reference pressure gauge

instrument used as a reference to comparatively calibrate a pressure gauge. Generally, a capacitance diaphragm gauge for which standard retroactivity has been maintained is used. The reference pressure gauge is used to measure air pressure in the test cell.

3.12

Capacitance Diaphragm Gauge

CDG

pressure or vacuum sensor that measures gas pressure by directly measuring the force applied on the surface of a thin diaphragm in the sensor. The mechanical deflection of the elastic sensor diaphragm is a function of the applied pressure. The CDG can be used to measure air pressure in various areas of the system.

3.13

air-tight wind tunnel

pathway through which air ventilation is generated by a wind generator based on an electrical fan or a vacuum pump. The test cell should be located in the air flow generated by the wind tunnel. For simulating the ascent speed in radiosoundings, it is necessary to simultaneously control the air ventilation speed and pressure. For an open suction-type wind tunnel, this can be achieved by using a vacuum pump and sonic nozzles. For the continuous circulation of air flow an electrical fan in an air-tight closed-loop wind tunnel can be used (schematic diagram presented in Figs. 1 and 2).

3.14

flow straightener

device used to reduce disturbances in air flow within the test cell

3.15

vacuum pump

device that draws gas, such as air, into the test cell to control the pressure

3.16

sonic nozzle

sonic nozzle (also known as critical flow Venturi) is a flow device that can be used as a calibration standard for gas flow meters and used to achieve a specific maximum constant flow when the ratio of the downstream pressure to the upstream pressure is smaller than a certain critical point as specified in ISO 9300:2022. The test cell lies in the downstream region of the sonic nozzles, wherein the pressure is lowered using a vacuum pump to attain the critical condition.

Note 1 to entry: The critical point in a sonic nozzle occurs when the gas velocity reaches the speed of sound within the nozzle. The pressure ratio at this critical condition is about 0,528 (=P_t/P_o), where P_t is the downstream pressure and P_o is the upstream pressure, respectively.

3.17

electric fan

electrically powered control device used to create air flow

3.18

solar simulator

device that generates illumination with a spectrum like that of sunlight by using Xenon (Xe) arc lamps to simulate solar radiation effect on radiosondes. The range of wavelength emitted from Xe arc lamps are 200 nm – 2 500 nm with a relatively smooth emission curve in the UV to visible spectra.

3.19

reference pyranometer

radiometer is normally used to measure global sunlight irradiance on a horizontal plane. Typically, a pyranometer of which traceability to the International System of units (SI) has been established is used

Note 1 to entry: A pyranometer can also be used at an angle to measure the total sunlight irradiance on an inclined plane. In this case, this includes an element caused by radiation reflected from the foreground.

3.20

irradiance

radiant power incident upon a unit area of a surface

Note 1 to entry: Irradiance is expressed in watts per square metre (W·m⁻²).

3.21

irradiance uniformity tester

experimental apparatus comprising an X-Y translation stage using a photodiode or a thermoelectric pyranometer with a current/volt meter to evaluate the two-dimensional (2D) uniformity of the illumination field generated by the solar simulator. Pyranometers are more suitable for comprehensive light measurements across a wide range of wavelengths whereas photodiodes operate mostly within ranges from 400 nm to 1 100 nm.

3.22

quartz window

flat transparent plate that separates the radiosonde sensors in the test cell from the outside environment while allowing light transmission through a solar simulator located outside the test cell. One or two quartz plates are necessary in the test cell while one is needed for the climate chamber when the solar simulator is outside the climate chamber.

3.23

light absorbing plate

plate made of black materials that eliminates light reflectance

Note 1 to entry: The black plate can be inclined at an angle of 45 degree to prevent potential light reflection back to the test cell. If it is not possible, the reflected flux should be measured.

3.24

correction

compensation for an estimated systematic effect

Note 1 to entry: The compensation can take different forms, such as an addend or a factor, or can be deduced from a table.

3.25

measurement uncertainty

non-negative parameter characterizing the dispersion of the quantity values being attributed to a measurand, based on the information used

Note 1 to entry: Measurement uncertainty includes components arising from systematic effects, such as components associated with corrections and the assigned quantity values of measurement standards, as well as the definitional uncertainty. Sometimes estimated systematic effects are not corrected for but, instead, associated measurement uncertainty components are incorporated.

Note 2 to entry: The parameter may be, for example, a standard deviation called standard measurement uncertainty (or a specified multiple of it), or the half-width of an interval, having a stated coverage probability.

Note 3 to entry: Measurement uncertainty comprises, in general, many components. Some of these may be evaluated by Type A evaluation of measurement uncertainty from the statistical distribution of the quantity values from series of measurements and can be characterized by standard deviations. The other components, which may be evaluated by Type B evaluation of measurement uncertainty, can also be characterized by standard deviations, evaluated from probability density functions based on experience or other information.

Note 4 to entry: In general, for a given set of information, it is understood that the measurement uncertainty is associated with a stated quantity value attributed to the measurand. A modification of this value results in a modification of the associated uncertainty.

3.26

standard uncertainty

measurement uncertainty expressed as a standard deviation

3.27

coverage factor

k

number larger than one by which a combined standard measurement uncertainty is multiplied to obtain an expanded measurement

Note 1 to entry: A coverage factor is usually symbolized.

3.28

expanded uncertainty

product of a combined standard measurement uncertainty and a factor larger than the number one

Note 1 to entry: The factor depends upon the type of probability distribution of the output quantity in a measurement model and on the selected coverage probability.

Note 2 to entry: The term “factor” in this definition refers to a coverage factor.

The laboratory setup can be described as the mechanism for generating air ventilation in the test cell such as by using an open suction-type or a closed-type wind tunnel.

Figure 1 shows the schematic diagram of an open suction-type setup, comprising sonic nozzles and a vacuum pump, to induce air ventilation at low pressures^[2].

A thermo-hygrostat chamber or space is needed to control and stabilise the temperature of the test chamber during the test. In general, the temperature range of commercially available climate chambers is −70 °C ~ 50 °C.

The heat exchanger, test cell and installation chamber should be contained within the climate chamber.

Key

Figure 1 — Schematic diagram of the laboratory setup comprising an open suction-type wind tunnel with a vacuum pump and sonic nozzles

Dry air should be used to test the radiation error of a radiosonde temperature sensor, especially at cold temperatures, to prevent frost/dew from forming on the inside of the windows. The outside of the windows should be blown using dry air.

It is recommended that the frost/dew point of dry air generated from a dry air generator should be lower than the test temperature to prevent frost/dew formation.

The heat exchangers should allow for fast thermal exchange; hence, the dry air temperature is adjusted to the target temperature of the test cell. The dry air temperature can be preadjusted by using a heat exchanger in a liquid bath before the air enters the climate chamber. In general, the temperature range of commercially available liquid baths ranges from −80 °C ~ 60 °C.

The test cell should be installed inside a climate chamber or liquid bath maintained at a constant temperature to control and stabilise the temperature. A climate chamber is more desirable over a bath for the handling of the test cell with radiosondes.

The test cell should allow light irradiation on the sensor through high light transmittance window(s) such as quartz.

The test cell should be located between the sonic nozzles and the vacuum pump to control the air pressure and ventilation speed.

The transmitted light beam should not be reflected back through the test cell. This can be achieved by placing a light-absorbing plate at the back end of the test cell.

The test cell should allow for the installation of at least one radiosonde sensor boom with temperature sensors.

If the test cell is big enough to test multiple sensors at the same time, the temperature gradient and irradiance inside the test cell should be evaluated and compensated for.

The temperature of the test cell can be varied within the range of −70 °C ~ 50 °C by using a commercially available climate chamber.

The temperature of the test cell should be measured around the radiosonde temperature sensor by using a calibrated thermometer to maintain traceability as per the International Temperature Scale of 1990^[7].

The temperature fluctuation of the test cell should be within ±0,05 °C over 10 minutes to enable measurement of a radiation error of 0,1 °C.

The operations of rotating and tilting the sensor boom are optionally and can be performed using stepper motors.

The test cell should be capable of mimicking the environmental conditions of air pressure encountered during a typical radiosonde ascent sounding as closely as possible. To fulfil this criteria, the pressure range of the gauge should be between 3 Pa and 1,000 hPa.

To mimic realistic air pressure, the entire system, including the sonic nozzles, heat exchangers, test cell, reference thermometer, pressure gauges and/or other ancillary devices, should be airtight.

The pressure in the test cell is measured using a calibrated pressure gauge to maintain SI traceability. The pressure in the test cell should stay within the larger of these two limits: ±5 % of the target pressure or ±0,6 hPa, for at least 10 minutes.

The ventilation speed in the test cell should mimic a typical radiosonde ascent speed. : Specifically, the range of air ventilation speed should be controlled within 3 m·s^-1 ~ 7 m·s^-1.

The reference air ventilation speed can be achieved using calibrated sonic nozzles.

Alternatively, a laser Doppler anemometry (LDA) can be used to measure the air ventilation reference speed by using the Doppler shift of the laser light scattered off aerosol particles. When employing LDA to measure ventilation speed, removal of the sensor boom from the setup is recommended. This is necessary since the presence of the sensor boom may obstruct the light measurement; thereby, potentially affecting the accuracy of the readings as shown in the inset of Figure 1. If both the laser and detector of the LDA are housed together, they should be positioned either at the solar simulator or the black surface as shown in the inset of Figure 1. Otherwise, the laser and the detector of the LDA should be placed in the location of the solar simulator and the black surface, respectively.

The stability of the air ventilation speed at a measurement spot should be within ±0,1 m·s^-1 for 10 min.

The air ventilation speed should be radially uniform in the sensor boom area of the test cell. The 2D uniformity can be measured by the LDA and should be within ±5 % in the location that includes the sensor boom.

Placing the sensor boom location more than 40 mm in the closed-type setup with a diameter of 150 mm ~ 200 mm and at least 10 mm from the wall in the open suction-type setup with a diameter of 50 mm from the test cell wall is recommended^[2,4].

A solar simulator comprises a Xe arc lamp as the light beam source, an optical collimator for focusing the beam light and a shutter to screen the light beam.

The optical collimator is set to focus the light beam on the radiosonde sensor boom, including the temperature sensor. The conductive heat transfer from the boom to the temperature sensor affects radiation error; thus, a part of the boom connected to the sensor must also be illuminated. It is recommended that the minimum size of the beam spot be 50 mm in diameter to acceptably illuminate the sensor and its connecting part of the boom.

The radiation flux (or irradiance) can be fixed because the radiation error is linearly proportional to the flux.^[4] Although the slope of the linear function depends on additional environmental parameters than the flux, the linearity itself is maintained. It is recommended that the radiation flux be fixed at 1,000 W·m⁻² or higher throughout the test.

The stability and the two dimensional uniformity of irradiance should be measured by using a calibrated pyranometer on a moving stage (irradiance uniformity tester). The stability and uniformity should be within ±2 % over at least 10 min and ±5 % within an area of 50 mm x 50 mm.

The radiation flux can be controlled by varying the distance from the light source to the sensor or by varying optical density by using neutral density filters. When using filters, the irradiance measurement in the test cell should include the effect of the filters.

A vacuum pump is used to provide air ventilation in the test cell and is to be located at the end of the air flow stream in the setup.

A valve is installed upstream of the pump to control the pressure within the test cell to satisfy criteria for acceptable performance of the sonic nozzle.

The maximum constant flow of a given sonic nozzle is determined by the desired up- and downstream conditions needed, especially for the ratio of downstream pressure ( in test cell) to upstream pressure (). At a certain critical ratio of /, a constant mass flow is generated. This mass flow is unaffected by downstream flow disturbances or pressure fluctuations as long as the pressure ratio remains lower than the critical pressure ratio (ISO 9300:2022).

Several sonic nozzles with different throat diameters are connected in parallel. The proper nozzle should be selected to provide the required ventilation speed and the pressure for the test.

Key

Figure 2 — Schematic diagram of a laboratory setup comprising a closed-type wind tunnel with an electric fan

Figure 2 shows the schematic of a closed-type wind tunnel setup comprising an electric fan and a vacuum pump used to induce air ventilation at low pressures.^[4] Only the technical requirements that differ from the open suction-type setup are discussed below.

If the entire system cannot be located within a climate chamber due to its dimensions, it should be located and tested in a more suitable location such as in a laboratory with controlled temperature. For these oversized systems, the temperature of the test cell can be adjusted to the room temperature of the laboratory. The effect of low temperature on radiation error has been previously studied and should be accounted for accordingly^[2,8].

A dry air generator is not required if the test is performed only at room temperature.

A liquid bath is not required if the test is performed only at room temperature.

The test cell must be included in the loop of the closed-type setup.

Rotation and tilting of the sensor boom can be added using stepper motors if desired.

The requirement is the same as 5.1.5.

The reference air ventilation speed should be decided by an experimental tool, such as the LDA.

The change in incident angle on the closed-type setup is more easily done than in the open-suction type.

Any type of vacuum pump can be used as long as it can set the desired pressure in the test cell within the required uncertainty.

The fan’s diameter is comparable to that of the wind tunnel setup in which it is mounted.

A rectifier between the fan and the test cell can be used to reduce turbulence.

When a desired pressure is set by a vacuum pump, the fan is set to generate the desired air ventilation speed.

The tests should be performed indoors under the environmental conditions of 25 ± 3 °C temperature and 50 ± 20 % relative humidity.

Radiosonde measurements can be conducted using either wired or wireless communication. The preferred method can be selected based on the setup's best support capabilities. The communication capability of the radiosonde should be verified as operating acceptably before conducting the test.

The data acquisition software recording the measurement data from the radiosonde should be verified as operating acceptably.

If measurements are to be taken using wired communication, the measurement signals, such as the voltage, resistance, current and capacitance, should be verified for compatibility with the operational characteristics of the radiosonde sensor.

The test cell windows and climate chamber should be checked to confirm that they are acceptably clean.

Confirm that the liquid bath and the climate chamber (if used) are operating normally and that the temperature remains stable within ±0,1 °C.

Confirm the dry gas generator (if used) is generating dry gas at the lowest frost point of –80 °C or lower.

Confirm the measurement system is operating acceptably during normal operation and verify the calibration validity for all measuring devices including but not limited to thermometers, pressure gauges, laser or sonic anemometers, and digital multi-meters.

Verify the measurement and control program is operating normally and that data is being collected continuously.

Verify the vacuum pump under normal operation is generating the target air ventilation and pressure.

Verify the solar simulator in normal operation is producing a radiation flux that is temporally and spatially stable at the target irradiance.

Examine the setup for leakage by applying vacuum to the system for a minimum of 10 min and verify there is no significant increase in air pressure. If there is leakage, perform investigation and corrective action to eliminate the leak(s) before starting the test.

If a fan used, confirm that under normal operation the rotating speed is sufficient to induce the desired ventilation speed within a stability of ±0,1 m·s^-1.

6.4.1 The simulator should be turned on in advance and allowed to warm up the light source to the temperature required for testing.

6.4.2 The radiation flux is measured at the sensor location in the test cell by using the reference pyranometer to confirm that the measured value is close to the target irradiance. The temporal stability should be ±2 % or better, and the two dimensional uniformity measured using the irradiance uniformity tester should be ±5 % or better. After measurement, the reference pyranometer should be removed from the test cell with the setting of the simulator intensity kept the same during the test.

6.4.3 After completing the procedures in 6.4.1 and 6.4.2, the beam is blocked by the screen shutter.

6.4.4 The stability of the light source should occasionally be verified using a calibrated pyranometer.

6.5.1 The radiosonde body is placed in the chamber, and the boom with temperature sensors placed between the two windows of the test cell.

6.5.2 A long-term power supply for the radiosonde is required since the test may take several hours or days to perform all tests of the radiosonde temperature sensor under all intended conditions.

6.5.3 The illumination area (beam spot) should be smaller than the window aperture to ensure that the light beam passes through the test cell, as shown in Figure 3. If this is not done, the test cell may be heated due to illumination.

6.5.4 The entire temperature sensor and a part of the boom must be illuminated, as shown in Figure 3.

6.5.5 The area for air ventilation and radiation flux stability and uniformity testing must cover the temperature sensor as shown in Figure 3. The test should be conducted using at least a (30 x 30) mm² area for the airflow and a(50 x 50) mm² are for the radiation flux.

Key

Figure 3 — Basic installation of a radiosonde temperature sensor in the test cell

6.5.6 The basic configuration of the sensor installation should ensure the sensor boom is illuminated in a perpendicular direction and is parallel to the air ventilation, as shown in Figure 3. This results in the highest degree of radiative sensor warming because the effective area of the irradiance on the sensor is maximised while the effect of air ventilation is minimised. The geometrical parameters affecting the effective irradiance on the sensor are discussed in Annex C.

6.5.7 After installing the radiosonde body and sensor boom, the doors of the installation chamber and climate chamber are closed. The installation chamber is then maintained airtight even when in a vacuum state.

The radiosonde temperature sensor radiation error depends on several environmental parameters including but not limited to temperature, air pressure, air ventilation speed, and irradiance (radiation flux) as well as geometrical factors such as the tilt angle of the sensor boom, motion of the radiosonde during ascent, and the solar elevation angle. In the case of environmental parameters, the magnitude of radiation error positively correlates to solar irradiance but negatively correlates to temperature, air pressure and ventilation speed.^[2,8] Geometrical factors are related to the effective solar irradiance (S_eff) on the sensor and the boom. In this International Standard, light illumination on the maximum effective area of the sensor boom, as shown in Figure 3, is considered as the default configuration for the test.

The tilt angle of the sensor boom with respect to the air ventilation direction can be adjusted within the range of 0° to 45° depending on the volume of the test cell.^[4] If tilting is not possible in the test cell, a default angle of 0° can be used.

Light illumination angle shown in Figure 2 can be varied within the range of 0° to 90° emulating the solar elevation angle (Annex C) with respect to the horizontal plane depending on the ability of moving the light source or waveguide tilting.^[4] If changing the illumination angle is not possible, a default angle of 90° can be used.

At lower temperatures, the thermal conductivity of air decreases, leading to a reduction in heat transfer from the sensor to the surrounding air under solar irradiation. This leads to an increase of radiation correction at lower temperatures.^[2,8] Therefore, conducting tests to examine this temperature effect is encouraged.

The temperature of the test cell and the ventilated air can be varied within the range −70 °C ~ 50 °C by using a commercially available climate chamber and/or a liquid bath. For example, the test can be performed at a few selected points, such as −70 °C, −55 °C, −40 °C, −20 °C, 0 °C, 20 °C and 40 °C.

When the temperature variation testing is not feasible, the test can be conducted at only room temperature. In this is case, the effect of temperature on radiation error should be properly compensated for^[2,8].

If the radiosonde humidity sensor is of the heating type, both the reference thermometer and the radiosonde temperature sensor could be affected by the humidity sensor heater. The temperature of the reference thermometer should not increase due to heat from the radiosonde heater.

Note 1 It is advisable to minimize the impact of the heater by increasing the air ventilation speed in the test cell to around (5~6) m∙s^-1 which will facilitate the convective cooling of sensors during soundings.

The air pressure and ventilation speed in the test cell can be varied within the range 3 hPa ~ 1,000 hPa and 2 m·s^-1 ~ 7 m·s^-1, respectively. For example, the test can be performed at a few selected air pressures of 3 hPa, 5 hPa, 10 hPa, 20 hPa, 50 hPa, 100 hPa, 300 hPa, 500 hPa and 1,000 hPa and at ventilation speeds of 3 m·s^-1, 5 m·s^-1 and 7 m·s^-1.

If temperature variation testing is possible, 7.1.2.1 can be repeated at each temperature.

The radiation flux can be fixed between 1,000  and 1,500  throughout the test because it is linearly proportional to radiation error^[4,9].

After installing a radiosonde temperature sensor in the test cell, the cell door is closed while carefully ensuring airtightness.

The solar simulator is then turned on. The light beam spot should be verified as illuminating the sensor properly, as shown in Figure 3.

The vacuum pump is operated to discharge the air in the test cell.

The rest of the operation sequence is shown in Figure 4.

Ideally, the measurement is conducted using the screen shutter of the solar simulator as shown in Figure 5. The illumination time and screen time should last until the temperature variation of the sensor reaches equilibrium.

The test sequence in Figure 4 can be repeated according to the desired test conditions in 6.6.

All measurements are initiated after the reference devices, such as the thermometer and the pressure gauges, are equilibrated for at least 10 min. The required conditions for the individual parameters before the first light illumination, as depicted in Figure 5, are listed below.

— The temperature variation range should be within ±0,05 °C for 10 min.

— The pressure variation range should be within ±5 % of the target pressure for 10 min.

— The ventilation speed variation range should be within ±0,1 m·s^-1 for 10 min.

It is recommended that the individual measurement values be obtained at least every 1 s during measurements.

To measure the reference temperature, the resistance of the reference thermometer is measured by using a digital multi-meter while employing a four-point method to minimise the effect of contact resistance.^[10] The result is then converted into temperature by using the coefficients in the calibration report.

Depending on the pressure range, multiple reference gauges can be used to minimise measurement uncertainty. For example, CDG 10 can be used when the pressure measured is in the range of 1 hPa ~ 10 hPa, CDG 100 in the pressure range of 10 hPa ~ 100 hPa and CDG 1 000 for pressure ranges higher than 100 hPa.

The air ventilation speed can be calculated by measuring the temperature, pressure and air mass flow rate based on their respective conversion equations, provided there are no experimental tools, such as an LDA, to confirm it.

The raw temperature of the radiosonde temperature sensor should be recorded and used.

Figure 4 — Flowchart of the sequential operation

The temperature of the setup is set to room temperature.

Air ventilation is stopped.

The air discharge line valve is closed, and the vacuum pump is stopped.

The radiosonde is removed from the test cell.

In principle, radiation error should be solely the temperature difference between the sensor and air under irradiation. In actuality, the air temperature measured in the test cell does not represent temperature since the air is heated by irradiation for a short time while passing through the test cell.

In this Standard, the radiation error value (ΔT_rad) is obtained by the difference in temperatures with irradiation () and without irradiation ( and ) as shown in Figure 5. The temperature of the test cell tends to rise (T₃ > T₁) during the measurements due to irradiation. Hence, the mean temperature between T₁ and T₃ can be used to determine the radiation error value as follows:

(1)

, , and are the mean of each temperature. The reference environmental parameters should be recorded throughout the experiments and an averaged value should be used to report the experimental condition, as shown in Table 1. To ensure statistical significance, it is recommended to have at least ten data points (measurements).

For radiosondes, the radiation error should be calculated to the second decimal place.

Steps should be repeated for all the selected test conditions.

Key

Figure 5 — Determination of the sensor’s radiation error

Table 1 — Example of the radiation error test at a specific environmental condition

The data set for estimated measurement errors in direct radiation can be obtained from experiments in predefined ranges of temperature (T), pressure (P) and ventilation speed (v), at a fixed solar irradiance (S₀), and can be represented using the following parameters.

(2)

Examples of analytical functional forms for data sets are described in Annex A.

If the irradiance (S₀) is fixed, the radiation error with effective irradiance (S_eff) can be obtained based on the following linear relationship.

(3)

The standard uncertainty for radiation error with normal illumination (S₀) can be derived from Formula (2), according to the uncertainty propagation law. Thereafter, the general form of standard uncertainty can be expressed by Formula (4). Note, the correlation among the parameters is assumed to be negligible because control over each parameter is possible in the laboratory setup.

(4)

where

In Annex B, the sub-components of each uncertainty parameter are explained.

When the tilting and motions of the radiosonde sensor with respect to the solar irradiation direction are known, the standard uncertainty of radiation error with effective irradiance (S_eff) can be expressed as follows:

(5)

where is the standard uncertainty of effective irradiance.

The geometrical factors affecting are described in Annex C. At the simulated sounding conditions, not all environmental factors present in atmospheric sounding conditions can be fully accounted for. Therefore, the uncertainty terms related to the direction of light illumination and the ventilation with respect to the sensor are neglected here.

The expanded uncertainty (about 95 % of the level of confidence) is calculated by multiplying the combined standard uncertainty by the coverage factor = 2 using Formula (6).

(6)

The measurement uncertainty should be calculated to one or two significant digits and reported according to the order of magnitude marked on the thermometer used in the test.

If necessary, the significant digits of the measurement uncertainty may be increased to the order of magnitude of the reference thermometer scale.

The measurement uncertainty should be calculated for all test temperature, pressure and ventilation speeds.

The test reports should include the environmental conditions, radiation error and uncertainty.

After performing the test according to this standard, a report is prepared with the results following the project format. Items that should generally be included in a test report:

— Client (name and address of institution) and report number;

— Instruments tested with the manufacturer and instrument models;

— Date test request was submitted and the date of test;

— Test environment (temperature and humidity);

— Name of test procedure or method;

— Test results;

— Test uncertainty;

— Description on measurement traceability;

— Signature of person performing the test and the person in charge;

— Comments and remarks.

An example of reporting the test results with uncertainty is presented in Table 2.

Table 2 — Example of test results of temperature measurement errors in direct radiation at various conditions

1. 
   (informative)
   
   Analytical functions of radiation error data set
   1. Analytical functions of radiation error data set

A.1.1 There can be many types of mathematical formulae for data error assessment. For example, a polynomial function^[4] can be applied to the pressure and ventilation speed ranges at room temperature as follows:

, where represent the fitting parameters (A1)

A.1.2 Application of acombination of the polynomial of the logarithm of pressure and a linear function of ventilation speed has been proposed in an independent study^[2] as follows:

(A2)

where , , and are the fitting parameters, which can be further processed using a linear function of T.

A.1.3 A combination of exponential decay functions for pressure and a linear function for ventilation speed has also been proposed^[2]:

(A3)

where , , , , and are the fitting parameters, which can be further parameterised by a linear function of T.

1. 
   (informative)
   
   Evaluation of uncertainty of environmental parameters
   1. Estimation of standard uncertainty
      1. Temperature,

B.1.1.1 The standard uncertainty for temperature in a test cell is based on the calibration uncertainty of the reference thermometer used in the test.

B.1.1.2 In addition to , the uncertainty budget for temperature should include the uncertainties due to the resolution of the thermometer , repeatability of the temperature , spatial temperature gradient of the test cell and the digital multimeter accuracy as follows:

(B1)

B.1.1.3 Further, reference thermometer long-term stability should be evaluated and incorporated if required. The combined standard uncertainty should ideally be within 0,05 °C.

• • 1. Pressure,

B.1.2.1 The standard uncertainty for pressure in a test cell is based on the calibration uncertainty of the reference pressure gauges used in the test.

B.1.2.2 In addition to , the uncertainty budget of the reference pressure should include the uncertainties due to the resolution of the gauge , repeatability of the pressure measurement , spatial pressure gradient of the test cell and the digital multimeter accuracy as follows:

(B2)

B.1.2.3 The long-term stability of the reference pressure gauges should also be evaluated and incorporated if required. The combined standard uncertainty should ideally be within 2 % of the reading value.

• • 1. Ventilation speed,

B.1.3.1 The standard uncertainty for the ventilation speed in a test cell is based on the calibration uncertainty of the sonic nozzles or the LDA used in the test.

B.1.3.2 In addition to , the uncertainty budget of the ventilation speed should include the uncertainties due to the repeatability of the ventilation speed and the spatial gradient of the ventilation speed .

(B3)

B.1.3.3 The combined standard uncertainty should ideally be within 2 % of the set value.

• • 1. Simulated irradiance,

B.1.4.1 The standard uncertainty for the simulated irradiance (radiation flux) in a test cell is based on the calibration uncertainty of the pyranometer used in the test.

B.1.4.2 In addition to , the uncertainty budget of irradiance should include the uncertainties due to the repeatability of irradiance and the spatial gradient of irradiance .

(B4)

B.1.4.3 The combined standard uncertainty should ideally be within 3 % of the set value.

• • 1. Fitting residuals,

B.1.5.1 To acquire the mathematical model for uncertainty evaluation, it is necessary to fit the data with the appropriate analytical function.

B.1.5.2 The standard uncertainty due to fitting can be obtained by the standard deviation of the residuals of the data set.

1. 
   (informative)
   
   Application of radiation correction to radiosoundings
   1. Effective solar irradiance
      1. Boom tilting, solar angle and radiosonde rotation

C.1.1.1 For applying the radiation correction obtained in this Standard to radiosoundings, the effective irradiance () on the sensor needs to be determined. Radiosondes keep changing positions with respect to the direction of solar radiation because they rotate and/or swing during soundings. The effective irradiance can be obtained by the mean over the radiosonde positions. The parameters affecting the are schematically illustrated in Figure C.1. The effective irradiance on the sensor can be calculated as follows:

(C1)

where

Key

Figure C.1 — Schematic diagram for estimating to the radiosonde sensor

C.1.1.2 In Formula (C1), the tilt angle of the sensor boom can be set in advance, and the solar elevation angle can be calculated by using the information on the location and the time of soundings. The effective illumination area obtained by the mean over a rotating radiosonde with a tilting angle of 45° is plotted as a function of the solar elevation angle in Figure C.2. In the calculation of the mean effective area, the pendulum motion of the radiosonde is not considered.

Key

Figure C.2 — Effective illumination area calculated by the mean over a rotating radiosonde with a tilting angle of 45° as a function of solar elevation angle

C.1.1.3 Since the in-situ measurement of the solar direct irradiance is not available during soundings, a proper radiation model that includes a cloud scenario and the surface albedo should be selected.^[11,12] The corresponding uncertainty of direct irradiance can then be evaluated and incorporated with the radiation correction in Formula (5) by using Formula (C1).

Key

Figure C.3 — The oscillating line represents the temperature of the rotating radiosonde

• 1. Temperature oscillation by the rotation of radiosondes
     1. Handling of temperature oscillation

C.2.1.1 The rotation of and any other semi-periodic motions of radiosondes introduce an oscillation during temperature measurement because of the change in effective solar irradiance. The approach of using the mean of the rotation in calculation of uncertainty ignores temperature oscillations although they can be a potential source of additional uncertainty.

C.2.1.2 The effect of rotation can be investigated through a laboratory setup by adding a spinning cycle for a radiosonde using a stepper motor. Figure C.3 shows an example of the temperature response for an irradiated rotating radiosonde.

C.2.1.3 The temperature oscillation during soundings can be suppressed by applying a low-pass filter or applying a moving average using an appropriate size filter window on the raw data.

C.2.1.4 Alternatively, the standard uncertainty due to the temperature oscillation of the rotating radiosonde can be incorporated when calculating the uncertainty of data fitting in Formula (5). The transient behaviour of oscillation can then be fitted with time, as shown in Formula (C2).

(C2)

where , and are the fitting parameters.

Note, represents the equilibrium temperature, as indicated through the horizontal line in Figure C.3.

C.2.1.5 Using the fitting parameter, the radiation correction () is obtained by using Eq. (C3). By using the fitting parameter, the radiation correction () is obtained using Eq. (C3).

(C3)

The second term represents the mean of the baselines due to the offset between the two measured temperatures in the same shutter state.

C.2.1.6 The standard uncertainty due to fitting in B.1.5 can be modified to include the uncertainty due to temperature oscillation. The uncertainty due to oscillation can be obtained by the standard deviation of the residuals, in the fitting of Eq. (C2) to the oscillated temperature.

Bibliography

[1] World Meteorological Organization. Guide to Instruments and Methods of Observation Volume I− Measurement of Meteorological Variables, Chapeter 12, 2021 edition, WMO-No. 8

[2] Lee, S.-W., Kim, S., Lee, Y.-S., Choi, B. I., Kang, W., Oh, Y. K., Park, S., Yoo, J.-K., Lee, J., Lee, S., Kwon, S., Kim, Y-G. (2022) Radiation correction and uncertainty evaluation of RS41 temperature sensors by using an upper-air simulator, Atmospheric Measurement Techniques, 15, 1107-1121.

[3] Dirksen, R. J., Sommer, M., Immler, F. J., Hurst, D. F., Kivi, R. and Vömel, H. (2014) Reference quality upper-air measurements: GRUAN data processing for the Vaisala RS92 radiosonde, Atmospheric Measurement Techniques, 7, 4463–4490.

[4] von Rohden, C., Sommer, M., Naebert, T., Motuz, V., and Dirksen, R. J. (2022) Laboratory characterisation of the radiation temperature error of radiosondes and its application to the GRUAN data processing for the Vaisala RS41. Atmospheric Measurement Techniques, 15, 383-405.

[5] Korea Patent 102021704 (2019) Apparatus and method for radio-sonde temperature and humidity calibration using upper air simulation technology

[6] US Patent US11287549B2 (2022) Apparatus and method for radio-sonde temperature and humidity calibration using upper air simulation technology

[7] Preston-Thomas, H. (1990) The International Temperature Scale of 1990 (ITS-90). Metrologia, 27, 3-10.

[8] Lee, S.-W., Kim, S., Lee, Y.-S., Yoo, J.-K., Lee, S., Kwon, S., Choi, B. I., So, J., Kim, Y-G. (2022) Laboratory characterisation and intercomparison sounding test of dual thermistor radiosondes for radiation correction, Atmospheric Measurement Techniques, 15, 2531-2545.

[9] Lee et al. (2018) Correction of solar irradiation effects on air temperature measurement using a dual-thermistor radiosonde at low temperature and low pressure, Meteorological Applications, 25, 283–291.

[10] Nicholas J. V. and White D. R. (2001) Traceable Temperatures, 2nd edition, pp 215-217, Wiley, Chichester.

[11] Key, J. R. (2002) Streamer User’s Guide, NOAA/NESDIS, Madison, Wisconsin, USA, available at: https://stratus.ssec.wisc.edu/streamer/documentation.html (last access: 19 July 2023).

[12] Key, J. R. and Schweiger, A. J. (1998) Tools for atmospheric radiative transfer: Streamer and FluxNet, Computers and Geosciences, 24, 443–451.

1. Currently, the lowest possible temperature of commercially-available climate chambers is approximately -75 °C. The temperature range can be adjusted base on the capability of the climate chamber used. ↑

2. The tilt angle of the sensor boom can be adjusted depending on the space of the test cell. If tilting is not possible, a default angle of 0° can be used. ↑

3. The light illumination angle can be adjusted depending on the ability of moving the light source or tilting the waveguide. If changing the illumination angle is not possible, a default angle of 90° can be used. ↑