Íslenskar landbúnaðarrannsóknir - 01.03.1980, Page 17

TEST OF A HAND-HELD RADIOMETER 15
tember is 9.4°C. .Vlean annual precipta-
tion is 800 mm of which 270 mm fall
through the growing season. The average
length of the frost-free period is 140 days/
year which is common for areas near sea
level in Iceland. The average annual
cloud cover of this area is 75 to 80 percent.
Judging from the natural vegatation, this
area as well as most parts of the Icelandic
lowlands (below 300 m elevation) belongs
to the cold temperate zone.
The measurements were done on a cul-
tiv'ated permanent hayfield. The soil is a
well drained, mineral soil of loessial origin
mixed with volcanic ashes from nearby
volcanoes. Due to their origin, Icelandic
soils are dark colored which enables them
to absorb more heat than light colored
soils during the short season of growth.
Due to slow chemical weathering of the
minerals and slow decomposition of the
organic matter, most Icelandic soils are
infertile but give excellent response to the
use result of fertilizers. The yields were
quite comparable in quantity and quality
with yields from grasslands in more
favorables climates.
The vegatation of the study site was
dominated by typical temperate or cold
climate grasses, mainly diíferent species of
Fesíuca, Poa, and Agrostis. The difference in
yield of the measured plots was primarily
due to the difierences in density of the
grasses. The basal cover of the plots that
gave the lowest yield was approximately
40 percent and around 90 percent for the
highest plots.
Data used
In situ measurements oí the red and in-
l'rared spectral radiance were obtained
with the previously described hand-held
radiometer. Six series of spectral meas-
urements were made on July 13 and 14,
1976 for 29 V4 m2grass plots. Immediately
after the fifth series of radiance measure-
ments, the plots were clipped of all stand-
ing vegetation. The sixth series of spectral
measurements were made after the clipp-
ing was completed. The clipped vegata-
tion was transferred to paper bags and
force airdried at 50°C for 48 hrs. Upon
completion of the drying cycle, the total
dry biomass weight measurement was
made.
Regression analysis
Regression analysis was used to quantify
the statistical relationship between the
measured radiances and the total dry
biomass for each series of spectral meas-
urements. A standardized regression
program, Biomedical Computer Prog-
ram‘s BMD02R, was used for the
analysis (DlXON 1974). Standard regres-
sion notation will be used after Draper
and Smith (1966) to descrobe the four
regression models used.
Canopy radianceX = /30 ^ + /3l ^ X TD/3
(1) where
Canopy radiance^ = canopy radiance at
wavelengt «
/30^ = regression derived estintate of /30 at
wavelength X
= regression derived estinrate of /31 at
wavelength X
TD/3 = total dry biomass
Canopy radiance^= /3o^+ /3l^ X TD/3_j (2)
Canopy radiance^ = /30^ X EXP(/3l ^ X TD/3)
(3) and
Canopv radianceX= (1-EXP(/301 + /3m X
TD/3_,))
(4) where
S* = asymptotic radiance at wavelength •.
Equations 2, 3, and 4 were transformed
into linear models prior to regression