Sa. Harris et Jh. Pedersen, COMPARISON OF 3 METHODS OF CALCULATING AIR-TEMPERATURE FROM ELECTRONIC MEASUREMENTS, Zeitschrift fur Geomorphologie, 39(2), 1995, pp. 203-210
Data was collected on air temperature every 20 minutes using a YSI 440
36 thermistor with a Lakewood data logger at three stations in differe
nt climatic and latitudinal situations. Mean daily air temperature was
calculated from the data in three ways, viz.: dividing the total of 7
2 readings by 72 (the integration method), adding the minimum and maxi
mum temperatures and dividing by 2 (the International Standard), and a
dding the 0800 and 1700 hour readings and dividing by 2. The integrati
on method provides a lower reading for most monthly and yearly data, a
nd there are substantial differences (up to 7 degrees C) for mean dail
y air temperatures calculated by different methods. These are largest
in winter and least in summer. Differences increase with continentalit
y and the integration method clearly gives a more accurate measure of
heat flux and climatic change, and should be calculated along with the
minimum-maximum average for automated stations.