2.1.1 Sam pling Frequency of Radiation Measurem ents
The BSRN requires that all radiation variables be sam pled at 1 HZ with an averaging tim e of one m inute. The final output for each variable should consist of the one-m inute m ean, m inim um , m axim um and standard deviation. T his specification is based upon the typical 1/e response tim e of first class pyranom eters and pyrheliom eters being approxim ately one second. Although som e instrum ents require the m easurem ent of m ore than one signal for the calculation of a specific radiation elem ent, the archived data will consist only of the m ean, m inim um , m axim um and standard deviation of the radiation elem ent.
W hen an elem ent requires m ore than one signal to be m easured or the conversion from the signal to the final value of the elem ent is non-linear, difficulties arise in providing a single true sam ple standard deviation for the one-m inute m ean value. This can be accom plished if the m easurem ents are stored each second and the calculations done later or the data acquisition system is capable of calculating the param eter each second.
The m ost com m on radiation observation m ade in the BSRN that requires m ultiple signals is infrared irradiance, where between 2 and 5 m easurem ents are m ade each second, depending on instrum ent type. There are two m ethods of data handling that provide the exact standard deviation for the flux and two m ethods that provide an estim ate of the standard deviation if the standard deviation cannot be calculated.
(1) Observations can be m ade of each of the required signals once per second and stored.
Using this data, the one m inute average can be calculated by applying the appropriate instrum ent responsivity to each voltage m easurem ent and the appropriate effects of the case and dom e tem peratures. The standard deviation can then be calculated from the individually calculated flux values. The prim ary drawback of this m ethod of signal processing is storage requirem ent associated with collecting one-second data.
(2) W ith the increasing com putation power of data acquisition equipm ent, the determ ination of the infrared flux can be m ade following the m easurem ent of the appropriate signals.
This would require the conversion of the instrum ent therm opile signal into a flux, and between 1 and 4 therm istor resistance m easurem ents into tem peratures and then the equivalent blackbody fluxes. This m ethod requires the em bedding of the therm opile responsivity into the data acquisition system . Many scientists are unwilling to include such inform ation in the acquisition stage of an observation because of the risk of error and the difficulty of correcting the problem when discovered. To reduce the potential of this type of error, while m aintaining the capability of calculating the standard deviation, the m ean, m inim um , m axim um and standard deviation of each of the raw signals can be stored along with calculated infrared irradiance.
Alternatively, given the difficulty associated with observing, storing and calculating the exact standard deviation of the infrared flux, the standard deviation of the flux can be reasonably estim ated based by the standard deviation of the therm opile signal. This estim ate assum es that over one-m inute the tem peratures of the case and dom e rem ain nearly constant and therefore do not affect greatly the overall standard deviation of the flux. This assum ption is substantiated by reference to Figure 8.1 that illustrates that a 5% change in therm istor resistance alters the overall flux by less than ±1.6 % over an extended tem perature range.
The observation of tem perature using therm istor technology is illustrative of a non-linear conversion from resistance to tem perature using the Steinhart and Hart equation (see Sec. 9.2.3).
In cases where the one-second data is not stored or the conversion of resistance to tem perature is not accom plished within the data acquisition system each second, the standard deviation of the tem perature should be estim ated based on the positive standard deviation
U ncertainty is defined as a param eter associated with the result of a m easurem ent, that characterizes the
1
dispersion of the values that could reasonably be attributed to the m easurand. Form er BSR N publications have termed this accuracy. T he terminology has been changed to follow the ISO guidelines. Accuracy is a m ore general term inology that expresses a variety of ideas, m any of which cannot be quantified.
ISO , 1993: G uidelines for the Ex press ion of U ncertainty M easurem ent. First Edition.
2
nip (lowercase) is used as an acronym and is not to be confused with N IP™ of Eppley Laboratories.
3
2.1.2 Sam pling Frequency of Ancillary Measurem ents
At stations where the ancillary m easurem ents are under the control of an independent agency, such as a national weather service, the frequency of the various m easurem ents cannot often be altered. The higher the frequency the greater the usefulness of the data, up to the sam pling rate of the radiation m easurem ents. BSRN station scientists should encourage any independent collection agency to sam ple and record data following standard W MO procedures at the very m inim um .
W hen autom atic data logging is em ployed to record such variables as pressure, tem perature, hum idity, wind speed and wind direction, providing these data at the sam e frequency as the radiation data is beneficial. Stations are encouraged to obtain these observations coincide ntally with the radiation m easurem ents using a one-m inute sam ple rate to aid in understanding the energy balance of the radiation instrum ents and the infrared com ponent of the radiation balance.
At a very m inim um , all stations should record air tem perature at the sam e location and at the sam e sam pling frequency as the radiation m easurem ents.
2.2 Uncertainty of Measurements1
2.2.1 Uncertainty in Radiation Measurem ents
These accuracies are based upon state-of-the-art com m ercially available equipm ent. At the onset of the BSRN program m e, a table listing the uncertainties about individual flux m easurem ents was produced (Table 2.1) that included the uncertainties thought to be achievable by 1997. T hese uncertainty values have been achieved using new sensors and m ethods of observation, som e being surpassed. Nevertheless, new m ethods of observation are continuing to develop that will continue to decrease the overall uncertainty associated with instantaneous m easurem ents.
Even as instrum entation and m ethods of observation have im proved over the decade since the inception of the network, the estim ate of uncertainty has becom e m ore refined. The publication of the International Organization for Standardization (ISO) Guide to the Expression of Uncertainty Measurement 2 (GUM) provides a standard m ethod for the determ ination of uncertainty in m easurem ent. National Metrology Institutions (NMI) and industrial laboratories have adopted its m ethodology and the BSRN recom m ends that all uncertainty calculations follow the procedures of the guide.
To m eet and exceed these target accuracies, the m easurem ent of each quantity will require a particular m ethodology of m easurem ent. W hile these m ethodologies are not absolute in nature, they will ensure a given level of uncertainty in the m easurem ent if followed (assum ing appropriate on-site m aintenance etc.). The BSRN is concerned m ore with m eeting the target m easurem ent uncertainty however, than the m anner in which the uncertainty is m et. Methods of m easurem ent associated with these uncertainties were first published in W CRP-64, 1991. W hile m any m ethodologies have not changed significantly since the inception of the program m e, several m easurem ent techniques have im proved. Those that have not changed are repeated verbatim in this m anual.
2.2.1.1 Direct Solar Irradiance
The target uncertainty for m easurem ent of direct solar irradiance in the BSRN is 1% (or 2 W m-2 as the m inim um deviation from the "true" value as reflected in the uncertainty of the W orld Radiom etric Reference). For the continuous m easurem ents used in providing the m ean value over one-m inute, a norm al incidence pyrheliom eter (nip ) or sim ilar is recom m ended. 3
BSRN Measurement Uncertainty Table 2.1. Uncertainty requirem ents for the Baseline Surface Radiation Network radiation fluxes. W here values are given in percent and absolute, the latter are the m inim um deviation from the “true” value m easured by the instrum ent for any irradiance.
Experim ents have shown that for m any nip instrum ents the uncertainty associated with the noise of these instrum ents exceeds the uncertainty requirem ents for direct solar irradiance m easurem ents. Therefore, an absolute cavity radiom eter (ACR) should be used in parallel to
"calibrate" the norm al incident pyrheliom eter quasi-continuously (every 5-60 m inutes, if the norm al direct beam intensity (I) > 400 W m ). -2
Pyrheliom eters norm ally operate with a window that blocks part of the solar infrared signal.
Sim ilarly, m any ACRs when used as all-weather instrum ents also have a window to protect the instrum ent from the elem ents. These windows m ust be m ade of the sam e m aterial to ensure that differences in window transm ittance are not ‘calibrated’ into the m easured irradiance and thus increase the uncertainty of the m easurem ent. To obtain higher quality m easurem ents that include the signal from the infrared portion of the solar spectrum , the instrum ent can be operated without a window or with a window m ade of a m aterial that has flat transm ission characteristics from approxim ately 290 nm to 4000 nm (> 99% of the solar spectrum ). Recent advances in the construction of all-weather enclosures, both windowless and those using sapphire or calcium fluoride windows and special heating and ventilation system s have reduced the dependence on sim ple therm opile pyrheliom eters that require frequent com parison with fair-weather ACR instrum ents. It is recom m ended that an all-weather ACR be used continuously with a standard pyrheliom eter used to fill ‘data gaps’ during the period when the ACR is in calibration m ode.
Caution m ust be exercised if a windowless ACR is to be operated continuously. The m inim um protection required is to house the instrum ent in a ventilated housing. The opening aperture of the housing should be a m inim um of 10 radiom eter-opening-aperture diam eters distant from the entrance aperture of the enclosed ACR and have a diam eter no greater than twice the field of view of the ACR. Care m ust be taken when ventilating the instrum ent so that no venturi effects are created that m ight alter the therm al equilibrium of the instrum ent. In areas where severe weather conditions are prevalent, system s that include a m eans of closing the opening aperture are required.
W hen using a calcium fluoride window, yearly inspections are recom m ended to ensure the integrity of the flat because of the anhydrous nature of the m aterial. In very hum id or wet environm ents, inspections of the flat should be m ade m onthly. T o protect the instrum ent from precipitation, an autom atic cover triggered by a rain sensor can be installed.
Experim ents have also shown that m aintaining the tem perature of the therm opile on certain ACR instrum ents, when used in an all-weather m ode, further enhance perform ance.
M ajor, G ., 1992: Estim ation of the error caused by the circ um solar radiation when m easuring global
4
radiation as a sum of direct and diffuse radiation. Solar Energy, 48. (S ee T able 4.1 for comm on combinations of pyrheliometers and pyranometers.)
Bush, B.C ., F.P.J. Valero and A.S. Sim posn, 2000: C haracterization of therm al effects in pyranometers: A
5
data correction algorithm for improved m easurem ent of surface insolation. Jour. Atm os. O cean. Tech., 17, 165 - 175.
A solar tracker with an accuracy of ±0.10° or better, is needed to accom m odate the pyrheliom eter, the ACR and, during calibrations, a second ACR. It is recom m ended that the tracker pointing be m onitored using a four-quadrant sensor because pointing accuracy is im portant in determ ining the quality of the direct beam m easurem ent. The sam pling rate should be the sam e as that of the instrum ents attached to the tracker. A variety of high-quality trackers are now available that use four-quadrant sensors for actively positioning the tracker on the solar beam during periods of high irradiance and a solar position algorithm during low intensity conditions.
The param eters to be m onitored are: output of the pyrheliom eter therm opile; outputs of the ACR (U, I or therm opile signals for a passive instrum ent); body tem peratures of the pyrheliom eter and ACR; output of the four-quadrant sensor. O nly the values associated with the calculated irradiance (m ean, m axim um , m inim um and standard deviation) are required by the BSR N archive.
All other raw data should be archived at the centre responsible for the m easurem ents.
2.2.1.2 Diffuse Radiation
The original instrum ent configuration associated with the BSRN target uncertainty of 4% (5 W m )-2 was a ventilated pyranom eter with a sensor to m onitor the instrum ent therm opile tem perature (to be used to correct for tem perature-related changes in therm opile responsivity and therm al offset). Shading of the instrum ent from the direct sun was to be by a shading disk. The shade geom etry of the com bination of the sensor and the disk was to replicate the geom etry of the direct beam sensor when pointing toward the zenith (5° full-angle from the centre of the detector) . The4 instrum ent’s sensor and dom e m ust be com pletely shaded. Incorrect geom etry alone can lead to errors of up to 5 W m depending on instrum entation and atm ospheric conditions.-2
A relationship has been shown between pyranom eter therm al offsets and diffuse irradiance that can significantly affect the quality of the m easurem ent . A possible solution to overcom e the offset5 problem is to use a ‘black and white’ type pyranom eter (B&W ) for the m easurem ent of diffuse radiation. This type of instrum ent does not exhibit the therm al offset of ‘black’ therm opile instrum ents because both hot and cold junctions are exposed to the sam e therm al regim e. B&W instrum ents, currently, do not have the sam e optical qualities (spectral and directional) as the black therm opile instrum ents recom m ended for use at BSRN stations and cannot be used for the m easurem ent of global radiation. Using a B&W instrum ent for the m easurem ent of diffuse irradiance would therefore m ean a second type of instrum ent would have to be used for the global irradiance m easurem ent, which m ay increase the overall uncertainty of the three-com ponent m easurem ents.
Research is presently ongoing to determ ine whether a correction factor can be applied to the
‘black’ pyranom eter therm al offset. In Section 9.2.2 experim ental m ethods of correcting this offset are presented. Further research continues into the design of a ‘black’ pyranom eter that does not exhibit therm al offset. Several instrum ents that use other technologies (e.g., PRT) are available that do not exhibit offset problem s associated with radiative cooling.
The BSRN has yet to recom m end a standard m ethod of correcting for therm al offset or selecting a particular type of instrum ent for m easuring diffuse irradiance. A careful uncertainty analysis of any diffuse system will determ ine the quality of the m easurem ent if each com ponent is properly addressed.
The param eters to be acquired are: output of pyranom eter therm opile; pyranom eter body tem perature. Only the values associated with the calculated irradiance (m ean, m axim um , m inim um and standard deviation) are required by BSRN archive. All other raw data should be archived at the centre responsible for the m easurem ents.
Alados-A rboledas, L., J. Vida and J.I. Jim éniz, 1988: Effects of solar radiation on the perform ance of
6
pyrgeometers with silicon domes. Jour. Atm os. O cean. Tech., 5, 666 - 670.
U do, S.O ., 2000: Q uantification of solar heating of the dom e of a pyrgeom eter for a tropical location: Ilorin, N igeria. Jour. Atm os. O cean. Tech., 17, 995 - 1000.
Philipona, R . C . Fröhlich and C h. Betz, 1995: C haracterization of pyrgeometers and the accuracy of
7
atmospheric long-wave radiation m easurem ents. Appl. O ptics, 34, 1598 - 1605.
2.2.1.3 Global Radiation
BSRN target uncertainty is 2% (5 W m ). Although the global radiation m ay be determ ined as a-2 sum of direct and diffuse irradiance, a direct m easurem ent will be m ade with a ventilated pyranom eter (the sam e instrum ent type as for diffuse radiation) to provide a basis for quality control; including instrum ent characterisation and calibration (see Section 8.3 - Calibration procedures).
T he sam e therm al offset issues associated with the m easurem ent of diffuse irradiance m ust also be considered for global irradiance. The difference in the m agnitude of the irradiance signals (global vs. diffuse) reduces the overall relative uncertainty associated with therm al offset for global irradiance m easurem ents.
Param eters to be acquired are: output of pyranom eter therm opile; pyranom eter body tem perature. Only the values associated with the calculated irradiance (m ean, m axim um , m inim um and standard deviation) are required by BSRN archive. All other raw data should be archived at the centre responsible for the m easurem ents.
2.2.1.4 Reflected Solar Radiation
This m easurem ent, required at BSRN stations undertaking the "expanded m easurem ent"
program m e, will be done with the sam e type of ventilated pyranom eter as for diffuse and global radiation. It is suggested that a horizontal shadowband be used to protect the instrum ent dom e from reflecting direct solar radiation onto the therm opile at low solar elevation. The angle sustained should be less than 5° (i.e., covering nadir angles 85° to 90°). W ith the exception of frost on other m aterial on the dom e that would enhance the internal reflection problem , the error due to internal reflection or the direct beam grazing the therm opile on a level instrum ent is estim ated to be <1 W m . The m inim um height above the surface for the m easurem ent is 30 m-2 so that the observations represents the relfectance of the surrounding area. The actual height of the downfacing centre should be reported to the archive.
Param eters to be acquired are: output of pyranom eter therm opile; pyranom eter body tem perature. Only the values associated with the calculated irradiance (m ean, m axim um , m inim um and standard deviation) are required by BSRN archive. All other raw data should be archived at the centre responsible for the m easurem ents.
2.2.1.5 Downwelling Infrared Radiation
BSRN target uncertainty is 5% or 10 W m , whichever is greater. Significant evidence suggests-2 that a pyrgeom eter with a hem ispheric silicon dom e is negatively im pacted by solar radiation through dom e heating . The BSRN recognizing this fact determ ined that downward infrared6 irradiance should be m easured with a shaded and ventilated pyrgeom eter. Furtherm ore, it noted that, if using an Eppley PIR, the battery circuit m ust be disconnected and the therm istor tem peratures directly m easured. It was determ ined that a “m odified PIR" pyrgeom eter (Eppley) with three dom e tem perature sensors at 45° (but without a battery circuit) was capable of7 m easuring downwelling infrared radiation to the target uncertainty. Although not norm ally used in an unshaded m ode, this m odified instrum ent is designed to m easure infrared radiation in full sunlight. Furtherm ore, it was recognized that a shaded and ventilated unm odified Eppley
BSRN target uncertainty is 5% or 10 W m , whichever is greater. Significant evidence suggests-2 that a pyrgeom eter with a hem ispheric silicon dom e is negatively im pacted by solar radiation through dom e heating . The BSRN recognizing this fact determ ined that downward infrared6 irradiance should be m easured with a shaded and ventilated pyrgeom eter. Furtherm ore, it noted that, if using an Eppley PIR, the battery circuit m ust be disconnected and the therm istor tem peratures directly m easured. It was determ ined that a “m odified PIR" pyrgeom eter (Eppley) with three dom e tem perature sensors at 45° (but without a battery circuit) was capable of7 m easuring downwelling infrared radiation to the target uncertainty. Although not norm ally used in an unshaded m ode, this m odified instrum ent is designed to m easure infrared radiation in full sunlight. Furtherm ore, it was recognized that a shaded and ventilated unm odified Eppley