The World Infrared Standard Group (WISG) of pyrgeometers was established in 2004 as an interim reference for atmospheric downwelling longwave radiation. Along with the PMOD/WRC blackbody cavity, calibrations of research and commerical longwave radiometers are conducted in order to provide coherent measurements of longwave radiation. Calibration procedures for pyrgeometers at PMOD/WRC are described in a WMO document (WMO, 2015).
In order to provide a stable reference for atmospheric downwelling longwave irradiance, the WMO decided that an interim World Infrared Standard Group of pyrgeometers should be established (WMO, 2006).
The WISG was originally calibrated with respect to the Absolute Sky-Scanning Radiometer (ASR; Philipona, 2001a) during the International Pyrgeometer and Absolute Sky-scanning Radiometer comparison (IPASRC-I) in 1999 (Philipona et al., 2001b) and has been operated since then without modifications to its calibration constants.
The WISG provides internal consistency to atmospheric downwelling longwave radiation measurements, while its traceability to the International System of Units (SI) is necessary to assimilate longwave radiation measurements from ground-based networks with shortwave radiation fluxes, in order to establish a coherent global radiation budget estimate.
Figure 1. The WISG (four pyrgeometers on the right) on the PMOD/WRC roof platform.
The WISG consists of four pyrgeometers, two Eppley Precision Infrared Radiometers (PIR-31463 and 31464), modified with three dome thermistors and two Kipp and Zonen pyrgeometers with one dome thermistor each (CG4-FT004 and CG4-010535). Two WISG pyrgeometers (PIR-31463 and CG4-FT004) were calibrated during the IPASRC-I in 1999, while the other two were calibrated relative to this initial set. Longwave radiation is calculated by PMOD/WRC using the so-called extended Albrecht & Cox equation (e.g. Philipona et al., 1995):
where E is the longwave radiation in watts per square metre (W.m−2); U is the measured voltage of the pyrgeometer thermopile in volts; C is the pyrgeometer sensitivity in V.W−1.m2; sigma is the Stefan–Boltzmann constant (5.6704 x 10−8 W.m−2.K−4); TB and TD are the measured body and dome temperatures of the pyrgeometer in Kelvin, respectively; and ki are the instrument constants. In a standard pyrgeometer calibration procedure at PMOD/WRC, ki are determined in the laboratory using a reference blackbody, while C is retrieved relative to the WISG average from outdoor nighttime measurements during clear-sky conditions (Gröbner and Wacker, 2015).
The WISG is operated continuously on the PMOD/WRC roof platform, and its stability is monitored by internal consistency checks of the four pyrgeometers comprising the WISG. Measurements show that the WISG has been stable to within ±1 W.m−2 since 2004 (Gröbner and Wacker, 2013), which is much lower than its associated uncertainty of ±4 W.m−2. This confirms its role as a suitable long-term reference for atmospheric longwave irradiance measurements.
As mentioned above, the WISG has never been recalibrated since IPASRC-I, and thus its traceability to SI units using the ASR has never been subsequently re-established and verified. While a pyrgeometer calibration with respect to the WISG is possible with a relative expanded uncertainty (95% coverage probability) of 0.9%, the WISG absolute uncertainty of ±4 W.m-2; is limited by the traceability of the WISG to SI units. The realisation of a more accurate standard group to determine irradiance (Reda et al., 2012) and a revision of the WISG reference scale have been ongoing issues (e.g. Gröbner et al., 2014, 2015; Philipona, 2015). Recently, two independent attempts to demonstrate traceability of atmospheric longwave radiation measurements to SI were realised: The Infrared Integrated Sphere (IRIS) radiometer developed by PMOD/WRC (Gröbner, 2012) and the Absolute Cavity Pyrgeometer (ACP) from NREL (Reda et al., 2012). While IRIS is calibrated relative to the well-characterised PMOD/WRC blackbody cavity, thus providing traceability to SI through temperature measurements of the cavity (Gröbner, 2008), the ACP uses a self-calibrating technique for in situ calibration of the instrument. Both radiometers are operated as windowless devices in order to minimize spectral inhomogeneities of their spectral responsivities, which has been shown to be the main cause for the observed discrepancies of commercial pyrgeometers (Gröbner and Los, 2007; Gröbner and Wacker, 2013).
The evidence for a revision of the reference scale comes from concurrent operation of the WISG alongside IRIS during night-time clear-sky conditions since 2008 which yielded an underestimation of the WISG clear-sky longwave irradiance by 2 – 6 W.m−2, depending on the amount of integrated water vapour (IWV) (Gröbner et al., 2014) in the atmosphere. These results have been confirmed in two intercomparison campaigns with the ACP (Gröbner et al., 2014) where the ACP and IRIS measurements were consistent to within ±1 W.m−2 during both campaigns (which is within the instrumental uncertainties of ±4 and ±2 W.m−2, respectively), while the WISG measured lower values by an average of 5.6 W.m−2 (Gröbner et al., 2014). Further support comes from measurements during the second International Pyrgeometer Comparison in 2015 at PMOD/WRC (unpublished data). However, a replacement of the WISG as a transfer standard is not foreseen due to its all-weather and hence continuous measurement capabilities which is not the case for the IRIS and ACP radiometers.
A new blackbody cavity for pyrgeometer characterisations was built at PMOD/WRC in 2007 to replace the cavity in use since 1995 (Gröbner, 2008). The calculated effective emissivity of 0.99993 ± 0.00033 was obtained from Monte Carlo simulations taking into account the geometry and the measured temperature distributions of the cavity. The cavity is operated over the −30°C to +30 °C temperature range, and is initially flushed with nitrogen to reduce the relative humidity. The estimated uncertainties of retrieved pyrgeometer parameters k1, k2, and k3 are ±0.024, ±0.0008, and ±0.03, respectively. The relative uncertainty of pyrgeometer sensitivity C is 0.8%. The comparison with the old blackbody gave average differences of 0.005, 0.00026, and 0.08 for k1, k2, and k3, respectively. Measurements of the pyrgeometer sensitivity retrieved with the new cavity were found to be higher by ~1.0% than with the original cavity.
Calibration procedures for pyrgeometers at PMOD/WRC are described in a WMO document (WMO, 2015).