The Importance of Pyranometer Temperature Response

The Importance of Pyranometer Temperature Responsearticle picture
Published: Monday, March 9, 2015 One of the main parameters affecting the real-world measurement of solar radiation by pyranometers is temperature response. ISO 9060:1990 defines temperature response and the limits for the pyranometer classifications, but it does not define how to calculate the values. There is more than one way to do this and the results may differ considerably, as this article explains.

Pyranometers and Their Classification

High quality solar radiation measurements are essential in many fields such as meteorology, climatology, material testing and solar energy. Global Horizontal Irradiance (GHI) is the most frequently measured quantity and is composed of Diffuse Horizontal Irradiance (DHI) and Direct Normal Irradiance (DNI). The relationship of these three components of solar radiation is:

GHI = DHI + DNI*cos(θ),  where θ is the solar zenith angle (vertically above a location is 0°, horizontal is 90°).

ISO 9060:1990 is the International Standard for the “specification and classification of instruments for measuring hemispherical solar and direct solar radiation”. It defines a pyranometer as the instrument for the measurement of GHI and specifies the parameters that have the largest impact on the quality of the measurement.

Pyranometer quality is classified by ISO 9060 according to the performance specifications. The three categories of increasing quality are; second class, first class and secondary standard. Today, many pyranometers significantly exceed the requirements of secondary standard, but there is no higher quality class. There is no primary standard pyranometer for measuring GHI, in fact it is calculated from the highest quality DHI and DNI measurements.

Defining Temperature Response

Assuming that the dome is kept clean, one of the largest impacts on the uncertainties in measuring GHI is the effect of ambient temperature on the sensitivity of the pyranometer compared to its calibration temperature. In desert areas the ambient (shade) temperature may be up to +50°C and in arctic regions down to -40°C, whereas the calibration temperature is usually around +20°C.

ISO 9060 defines pyranometer temperature response as the deviation in sensitivity over a 50 Kelvin temperature interval as a percentage of the calibration sensitivity at a specified temperature, . The maximum values for the pyranometer classifications are given in the table below.

ISO 9060:1990 Temperature Response

Pyranometer classification

Secondary Standard

First Class

Second Class

Deviation over 50K interval




Measuring Temperature Response

ISO 9060 defines temperature response, but not how to measure it in a standardised way. One issue is that the temperature response of a thermopile pyranometer is non-linear and largely depends upon the heat-flow characteristics of the mechanical construction. Typically the function is a third-order polynomial.

A temperature range from -10°C to +40°C is commonly used as being representative of non-extreme climates. However, different methods to determine the pyranometer temperature response are used in the market.

The Kipp & Zonen method uses the largest deviation from the calibration sensitivity over an interval of 50 K:

This gives the worst-case change in pyranometer sensitivity with respect to the calibration value. An alternative method uses ½ the difference between the maximum and minimum sensitivity deviations over a temperature interval:

where ∆S(T) is the sensitivity deviation, at temperature T, relative to the calibration sensitivity.

But, this does not give a realistic temperature response when the deviations are mostly in the same direction.

The difference between the two methods is clarified by the following example.

Example of pyranometer sensitivity change with temperature

T [ºC]

Sensitivity deviation








0% (calibration temperature)





In this example the sensitivity reduces at both lower and higher temperatures, which is not an unusual effect, so the minimum deviation is -1.4% and the maximum deviation is 0%. By convention, the sign of the deviation is ignored and the temperature response is given as ± x%, or < x%

The temperature response calculated with the two methods is shown below.


Temperature response

Kipp & Zonen method


Alternative method

(0% +1.4%)/2 = 0.7%

This alternative method used in the market shows a temperature response of only 0.7%, but it is not realistic. If the pyranometer would be used at -10 ºC the sensitivity would deviate by -1.4% from the calibrated sensitivity.

The Kipp & Zonen method, showing a temperature response of 1.4% gives a realistic indication for the temperature response error in measurement.

Know the calculation of the temperature response

To know that you have a high quality pyranometer for measuring hemispherical solar irradiance you need to know the temperature response. However, you also need to know how the temperature response was calculated, as this is not defined within ISO 9060:1990.

For a comparison of Kipp & Zonen pyranometer specifications with the specifications of other manufacturers it is important to realize that there are different methods to determine the temperature response. Some manufacturers use a method which underestimates the temperature response by up to a factor of 2.

Below table shows the list of Kipp & Zonen pyranometers and their classification and temperature response. 


ISO 9060:1990 classification

Temperature response

CMP 22

Secondary Standard

< 0.5 % (-20 °C to +50 °C)

CMP 21

Secondary Standard

< 1 % (-20 °C to +50 °C)

CMP 10 & 11

Secondary Standard

< 1 % (-10 °C to +40 °C)

SMP10 & 11

Secondary Standard

< 1 % (-20 °C to +50 °C)

< 2 % (-40 °C to +70 °C)


First Class

< 4 % (-10 °C to +40 °C)


Second Class

< 5 % (-10 °C to +40 °C)


Second Class

< 3 % (-20 °C to +50 °C)

< 5 % (-40 °C to +70 °C)


Article by Marc Korevaar, Physicist at the R&D department of Kipp & Zonen.

Share this page