Tuning Sky Temperature Parameters


The amount of clouds in the sky is directly related to the infrared radiation emitted by the clouds in the wavelength range of 8-15 µm.


The infrared sensor measures the total downward radiation in this range and it contains several components:

·      The radiation emitted by the clouds;

·      The radiation emitted by the atmosphere ;

·      The radiation reflected by the ground;

·      Etc....


To setup the AAG_CloudWatcher one should define temperature limit below which the sky is considered CLEAR and the temperature limit which distinguishes a CLOUDY  sky from an OVERCAST (very cloudy) sky.


From empirical observation, the limit between CLEAR and CLOUDY conditions corresponds to a value between -6°C and -3°C (the program default value = -5°C) whereas the limit between CLOUDY and OVERCAST conditions is a value between 0°C and 2 °C (the program default value = 0°C).


However, during warm days / evenings, one notices that the measured sky temperature values reflect a large component due to other atmospheric radiation.  In order to cope with this effect a temperature correction model has been introduced (please refer to Sky Temperature Correction Coefficients). This model calculates a correction value for the measured sky temperature as a function of the surface temperature.


The model is a combination of a linear and an exponential relationship where the linear relationship predominates for ambient temperatures below 20°C whereas the exponential becomes more pronounced for ambient temperatures above this value.


Tuning up the sky temperature correction model


To tune up this model, one should observe the Cloud conditions graph from sunup to sunset for a clear day.


As the ambient temperature changes during the day the Cloud conditions graph should remain horizontal, provided the sky conditions remain stable and cloudless.


If one notices that there is a downward trend in the graph line as the ambient temperature increases – this means that the calculated sky temperature correction factor must be increased.


If this trend occurs for temperatures below 25°C, one should try to increase the coefficient K1 by a small amount. (Suggestion: try a value from 33 to 38)


If the trend is more noticeable for values above 30°C, then one should adjust coefficient s K3, K4 and K5. (Suggestion: try the following combination - K3 a value between 4 and 10 with K4=100 and K5=100)


On the other hand, if one notices that the graph is horizontal but it is too low, one may adjust coefficient K2. This coefficient shifts upwards the calculated correction value as the coefficient gets smaller and vice-versa. (Note that negative values are allowed for this coefficient.)


The default values for coefficients K1, K2, K3, K4 and K5 are 33, 0, 0, 0 and 0 and this corresponds to a simple linear relationship.


The following coefficients have proved to yield good results too: K1=33, K2=0, K3=8, K4=100 and K5=100. This combination is more nonlinear for ambient temperatures above 30°C.




When adjusting these coefficients, use the graphical display provided (see – Sky Temperature Correction Coefficients).