THÈSE - Station Alpine Joseph Fourier - Université Joseph Fourier

Chapitre II Article 2A
decay (yr -1 ). Decay constants were determined for each species in each topographical location
by fitting a linear-regression of the (LN-transformed) litter remaining against time. An
ANCOVA was applied within topographical location to test the significance of the Mass
remaining × Species interaction.
Finally, an ANCOVA was applied to test the regression between (1) yearly mass loss of each
substrate against mean wintertime soil temperature and (2) litter N content (proportion of
initial N content) of each species against mass loss. Analysed data were the mean of each
harvest, in each topographical location. Species were considered as the qualitative factor
whereas mean wintertime soil temperature (1) or mass loss (2) were the continuous factors.
In Experiment II, we estimated decomposition rate for each period, each species and for each
topographical situation by applying the following formula :
1
T
[ remaining mass − remaining mass ]
Decomposition rate ( mg / d)
= ⋅
t
t−1
where T is the period between t-1 and t (in days). We did not estimate the exponential decay
constant (as done in experiment I) as the aims of this experiment were to precise for each
period the decomposition rate of each substrate rather than the mean decomposition rate over
all the period of the experiment. We used a two-way analysis of variance (ANOVA) to
compare period and site effects for each species separately and to test differences in litter N
content for each species (proportion of initial N content) after the first winter, summer and
second winter. Finally, a Student’s t-test was used to assess the initial value (=1) litter N
content (in proportion to initial) and that after winter I, summer and winter II. Statistical
analyses were performed using R software (R Development Core Team 2006).
Characteristics
Winter duration (d)
Mean winter soil temperature (°C)
Mean summer soil temperature (°C)
Gravimetric summer soil moisture (gH2O g -1 soil)
Productivity (g m -2 d -1 )
Total biomass (g m -2 )
Late snowmelt
location
240 (12)
-0.15 (0.01)
7.7 (1.55)
0.35 (0.03)
7.23 (0.45)
325.67 (22.04)
Early snowmelt
location
206 (15)
-3.05 (0.45)
7.6 (1.62)
0.42 (0.03)
2.18 (0.14)
185.90 (11.72)
Table 1 Winter duration, soil temperature records and soil moisture are an average of three years records from
the three sites A, B and C within each topographical location (from the end of 2003 to the end of 2006). Values
are the estimated parameter ± standard error except for winter duration where it is standard deviation.
141