Kipp&Zonen BSRN Scientific Solar Monitoring System User Manual

Page 80

To maintain the traceability of pyrgeom eter m easurem ents the following procedure has been established:

(1)

Each BSRN station requires a m inim um of two pyrgeom eters, initially calibrated at the W RC.
One of these instrum ents is to be declared a site reference instrum ent and used only during
tim es of com parison. The other instrum ent(s) will be classified as the field instrum ent(s). An
initial com parison of the two instrum ents should be m ade im m ediately upon deploym ent at
the station to determ ine the relationships between the therm istors and therm opiles of the
instrum ents.

(2)

Com parisons between instrum ents should occur a m inim um of once every 4 m onths at sites
where no significant seasonal variations occur and once each season (nom inally every 3
m onths) at locations with significant change. IT SHOULD NOT BE NECESSARY FOR THE
FIELD INSTRUMENTS TO BE TAKEN OUT OF SERVICE DURING THE PERIOD OF THE
COMPARISON.

(3)

W herever possible, the instrum ents being com pared should use the sam e tracking shade
device and data acquisition system to reduce system atic biases. Instrum ent ventilation m ust
be with the sam e style ventilator. In cases where separate tracking shade devices are used,
care m ust be taken to ensure that both the field and reference instrum ents are shaded correctly.
W hen the sam e data acquisition system cannot be used, a com parison between data acquisition
system s m ust be perform ed. Com parison data should not be collected until the reference
instrum ent has com e to therm al equilibrium with its surroundings.

(4)

Norm al station sam pling protocols should be used during the com parison.

(5)

The com parison should last a m inim um of 2 days and a m axim um of 5 days. It is desirable
that data be obtained which corresponds to a typical range of irradiances for the period. This
m ay be accom plished by acquiring m easurem ents under a variety of non-precipitating weather
conditions during both daylight and nighttim e hours.

(6)

From the continuous data set collected during the com parison period, only steady-state
conditions should be used for the com parison. This can be som ewhat arbitrarily defined as
those periods where the standard deviations of the therm opile and therm istor signals are
less than 0.25% of the m agnitude of the respective m ean signal for the averaging period.

(7)

Analyses should be perform ed to determ ine if there are any changes in the following:

(i)

the ratio of a given reference instrum ent therm istor to the respective field instrum ent
therm istor (dom e and body).

(ii)

the ratio of the reference instrum ent therm opile output to the field instrum ent therm opile
output.

(iii)

the ratio of the calculated irradiance of the reference instrum ent to the calculated
irradiance from the the field instrum ent.

Note: For typical irradiances, the tem perature of the case accounts for between 75 and 100%
of the signal for clear to isotherm al conditions. The Eppley PIR uses Yellow Springs Instrum ents
(YSI) therm istor YSI 44031 for these m easurem ents. The therm istor is specified to have an
interchangeability of ±0.1 °C between 0 °C and 70 °C and is nom inally 10 KS at 25 °C.
Conversion from resistance to tem perature is accom plished through the Steinhart and Hart
equation,

T = a + b(lnR) + c(lnR)

-1

where

T = tem perature in Kelvin
R = resistance in ohm s

The coefficients for the YSI 44031 therm istor are: a = 0.0010295, b = 0.0002391 and
c = 1.568e-07.