Zusammenfassung
Simultaneous but independent pairs of observations of solar radiation
made in close proximity to each other for extended time periods at
both Vancouver and Toronto have permitted an objective determination
of the uncertainty in measurements of solar radiation. The widely
quoted observation error of -.+ 5 per cent for an individual observation
is appropriate to an hourly time interval, with the uncertainty showing
substantial reduction with increased time averaging. The absolute
values of the systematic differences were of similar magnitude (2-2.5
per cent) at the two locations. Both the individual and systematic
differences showed marked seasonal and interannual variability,w
ith all relativee rrors being greatest in the winter months. However,
at both locations the absolute values of these same statistics showed
a strong dependence on the observed irradiance. The variability of
these discrepancies illustrates the need to specify the basic time
interval and irradiance when selecting or quoting a measurement error.
Nutzer