Rit (Vísindafélag Íslendinga) - 01.06.1976, Qupperneq 28

Comparison of the results obtained for samples from these two
rain gauges might be considered as an indicator of the accuracy in
sampling of precipitation for deuterium measurements.
The weighted mean deuterium concentration of the samples for
the nine months in question is practically the same for both rain
gauges. The monthly values show some difference. The root mean
square (rms) of the differences o = 1.7%0, which is considerably
larger than the accuracy of the measurement (0.7%0), may be con-
sidered as the accuracy in sampling of precipitation.
3. 2 Cold groundwater and rivers.
Sampling of water from cold springs needs no description. The
samples are usually collected directly at the outflow of a spring
without any possibility of isotopic enrichment due to evaporation.
Changes in isotopic composition of rivers or small streams due
to fractional evaporation from the river surface must also be very
small. Therefore a sample collected from a river will be representa-
tive for the sources feeding the respective river. These can be seen
by considering the following two extreme examples.
The river Þjórsá is one of the biggest and longest rivers in Ice-
land. Þjórsá has a total length of 230 km and a mean discharge of
approximately 400 m3/sec. "VVe assume the average width of the
river to be approximately 200 m, thus giving a total water surface
of 46 km2. According to Einarsson (1972) the highest possible eva-
poration from an open surface in Iceland is found to be approx.
80 mm during the month of July. Using these data it turns out that
at the time of the year when the evaporation is greatest, less than
4%0 of the water in the river Þjórsá has evaporated on its course.
In the case of a small stream with a relatively greater surface the
evaporating rate can be up to an order of magnitude higher. Con-
sider a small stream with approximately 0.1 m3/sec discharge, total
length of approximately 10 km and an average width of 10 m. In
this case about 3% of the water could have evaporated on its course.
This evaporation wordd not have changed the deuterium content
of the river significantly.