Kipp&Zonen BSRN Scientific Solar Monitoring System - Manual (Page 83)

D utton, E.G ., J.J. Michalsky, T . Stoffel, B.W . Forgan, J. H ickey, D . W . N elson, T .L. Alberta and I. R eda,
23
2001: M easurem ent of broadband diffuse solar irradiance using current com m ercial instrumentation with a
correction for therm al offset errors. Jour. Atmos. O cean. Tech., 18, 297 - 314.
W ardle, D .I. et al., 1996: Im proved m easurements of solar irradiance by m eans of detailed pyranometer
24
chracac terisation. Solar H eating and C ooling Programm e T ask 9, International Energy Agency R eport
IEA-S H C P-9C 2, Atm ospheric Environm ent Service, D ownsview, O ntario.
Philipona, R . C . Fröhlich, C h. Betz, 1995: C haracterization of pyrgeometers and the accuracy of
25
atmospheric long-wave radiation instrum ents. Applied O ptics, 34(9) 1598-1605.
71
changes by greater than 0.5% over the operating tem perature range of the instrum ent, a responsivity
correction factor should be applied.
The zero-offset due to therm al emittance should be corrected, but presently no agreem ent on the correction
m ethodology has been reached. Two com m on m ethods are presented below:
(1) An em pirical correction factor dependent on instrum ent type can be used of the form : 23
0 1 2
os = b + b NetIR + b DC
where: os = offset in W m -2
NetIR = the net IR irradiance m easured by the therm opile of
a shaded and ventilated pyrgeom eter
d om e case
DC = F [T - T ] 4 4
T = m easured pyrgeom eter tem peratures
F = Stefan-Boltzm ann constant
x
b = regression coefficients dependent on instrum ent
(2) A sim ilar correction using the IR irradiance has been used with several instrum ents when
they are mounted in the Canadian Meteorological Service venilated housing: This m ethodology
correlates night tim e offset with the incom ing infrared radiation. The relationship for Eppley
PSP’s and Kipp and Zonen CM11's is approxim ately :
24
Z = -1.5 + 0.025 P
where: Z = the zero offset
P = the infrared signal as m easured by the pyrgeom eter
As both of these m ethods are determ ined em pirically on a lim ited num ber of instrum ents, the coefficients
should not be applied without first testing the results on the individual instrum ents on which they are
to be used.
A less accurate method of determ ining the offset, is through the linear interpolation of the 60-m inute-m ean
value of the pyranom eter signal from before and after astronom ical twilight ($108°) for the sam e day.
This assum es that the infrared response of the instrum ent rem ains reasonably constant throughout
the solar day, which is not the case in conditions of scattered cloud.
9.2.3 Pyrgeom eters
The pyrgeom eter signal should be based upon the calibration constants directly traceable to the W RC
blackbody. The flux should be calculated as :
25
where = therm opile responsivity (:V W m ) -1 2
= instrum ent dependent calibration constants