eLearning

accumulation in the trees was not measured but
should be on the order of less than 10 kg N ha-1
y-1
,
which would be similar to wet N deposition in this
region. Overall, this was a strong chronosequence
design, and it showed no evidence of occult N ac-
cretion.
REPEATED SAMPLING
The Broadbalk “Wilderness”
In the classic Broadbalk experiment in England, a
very small abandoned field regenerated to forest,
and accumulated N at a rate of 48 kg ha-1
y-1
over
100 years in the top 23 cm of mineral soil (1865–
1966; Day and others 1975; see also Jenkinson and
others 1994). These values were the minimum an-
nual increase after accounting for expected N inputs
(1.4 kg N ha-1
y-1
rainfall, 13 kg N ha-1
y-1
dry
deposition, and 1.5 kg N ha-1
y-1
organic nitrogen in
rain, dust, and bird droppings). The occult N source
was thought to be nonsymbiotic N fixation, and
Day and others (1975) concluded they found high
rates of acetylene reduction in the field. These con-
clusions do not warrant high confidence because
the small plot size led to a large edge effect. A large
proportion of the dominant trees had canopies that
extended substantially beyond the plot boundaries
over adjacent agricultural fields. Only a few mea-
surements were made for the acetylene reduction
assays, and these were far below the estimated rates
of occult N accumulation. It would be necessary to
estimate the quantity of N scavenged by the trees
from the surrounding agricultural fields before an
assessment could be made of the sources of the high
N accumulation in the small forest plot.
The Calhoun Forest
The most thorough resampling of forest soils over
time occurred in the Calhoun Experimental Forest
in South Carolina, USA (Richter and Markewitz
2000). Mineral soils were sampled intensively in
eight replicate plots of a loblolly pine (Pinus taeda L.)
plantation, beginning at age 5 and continuing at
roughly 5-year intervals over the next 35 years. The
plantation developed on an old agricultural field, so
there was no initial O-horizon or vegetation. Most
of the N that accumulated in the O-horizon and
vegetation appeared to come from the pool of N
formerly in the mineral soil (Figure 1). The appar-
ent net increment for the whole forest was approx-
imately 200 kg N/ha, giving an average N accretion
of6kgNha-1
y-1
. The 95% confidence interval
around this mean ranged from –0.3 kg N ha-1
y-1
(a
net loss) to a gain of 14 kg N ha-1
y-1
. This thorough