Fjölrit RALA - 15.12.2000, Síða 11

9
Fertilization
Table 2 shows the amount of fertilizer used in the experiments. In Kollaljorður the fertilizer
was split in two applications. The experiment in Narsarsuaq was fertilized at the same time as
the surrounding grass field.
Table 2. Amount of fertilizer used in the experiments (kg/ha) and time of application.
Location Year Date N P K S Mg Ca
Kollafjorður 1996 6.5, 3.6. 100 28 66
1997 2.5,26.5 140 44 105
1998 7.5. 120 34 84
Upemaviarsuk 1996 100 44 83 13 9 7
1997 30.4. 100 44 83 13 9 7
1998 100 44 83 13 9 7
Narsarsuaq 1996 Late May 98 43 81 13 9 7
1997 Late May 98 43 81 13 9 7
1998 6.6. 105 46 87 14 10 7
Korpa 1996 14.5. 80 17 52 9 16
1997 12.5. 83 18 35 8 15
1998 7.5. (lst cut) 90 19 37 9 18
1998 16.6. (2nd cut) 45 10 19 5 9
Measurements
Each plot was divided in two with one half harvested according to a harvest scheme. The
other half was cut once late in summer in the first year. The following year the treatments
were reversed on the two halves. This was done to avoid harvest time effects from the
previous year.
In 1996 and 1997 there were fíve harvest dates, the first one when P. pratense was 10
cm high and thereafter at two week intervals. However, on May 29 there was no yield at all of
Poapratensis in Kollafjorður and therefore an extra harvest was cut on July 24. In the first
four harvests, yield was measured by clipping two 2 m long and 10 cm wide strips in each
plot (one at each end). The clipped strips were always a few cm further in than the earlier
clipped ones. At the fifth harvest date the remainder of each plot was harvested with a
harvesting machine. Dates for onset of growth, green cover of field, heading and anthesis
were noted and at each harvest date height of the stand was measured. Furthermore, at each
harvest date samples of Engmo were collected for estimation of the developmental stage. The
samples from all locations were evaluated by the same person in Iceland to ensure uniform
results. All weeds were cleaned from the yield samples and these were subsequently used for
chemical analyses.
In the third experimental year, 1998, there were two harvest times of the first cut at
Korpa, but the regrowth was harvested at the same date for all plots. The experiments in
Greenland and Kollafjorður were harvested only once in 1998. In Upemaviarsuk the summer
was too cold and short for two cuts, but two cuts should be possible at the other locations.
Digestibility was measured in samples from all harvest dates in 1996 and 1997. Crude
protein was measured in all samples from 1996 and from three harvest dates in 1997. Samples
from two of the four blocks were analyzed.