A New Look at Calibration and Use of Eppley Precision Infrared Radiometers. Part II: Calibration and Use of the Woods Hole Oceanographic Institution Improved Meteorology Precision Infrared Radiometer*
For some years, investigators have made measurements of downwelling longwave irradiance with the Eppley Precision Infrared Radiometer (PIR), recording the values of thermopile voltage and body and dome thermistor resistances and combining them in data processing. Part I of this paper reviews previous work on the processing equation and presents an improved equation. It establishes that the standard single-output Eppley has an inherent uncertainty of 5%. By measuring the three possible outputs separately and comparing them in the improved equation, the inherent accuracy can be improved to 1.5%. Part II presents a method of calibrating the Eppley PIR for the three-output equation using an easily constructed blackbody cavity in a temperature bath capable of a 0°–50°C temperature range. Calibration of PIR thermistors is recommended since occasionally one is found out of specifications. An outdoor comparison of 15 PIRs calibrated with the technique was carried out in groups of four, with one PIR used in all of the groups as a standard of comparison. The mean differences and 1-min standard deviations between 12 individual PIRs and this standard over comparison periods of 10–22 days were less than 6.0 and 11 W m−2, respectively. Only two of the PIRs and a standard single-output Eppley PIR (calibrated by Eppley) had mean differences and standard deviations greater than 7 and 11 W m−2, respectively. Although the new calibration procedure yielded consistent results in the mean, at times the longwave measurements diverged by up to 45 W m−2 for several hours. Some of these events are attributable to confirmed pinholes in the dome filter, but others are left unexplained.