Samnrkin et al.: Studies ofMetlume Production and Emission
333
carbon concentrations in pore water. A more detailled description of all mentioned methods
above can be found in Samarkin et al (1997).
The isotopic composition (13C and D) of the methane gas samples is measured in the Institute
of Geology and Geophysics, University of Leipzig.
Other investigations
The cation contents of the pore water samples were analysed at the Alfred-Wegener-Institut
Potsdam with an ICP OES Optima 3000 (Perkin Elmer Co.) and NBR Standard SRM 1643d.
Results and discussion
Methane emission
Methane emissions from the polygon wall (site 8) and the polygon depression (site 7) differ
strongly from each other: while the mean emission rate of the wet polygon depression was
about 75 mg CH 4 *d- 1 *m- 2 , the mean emission rate of the higher and dryer polygon wall part
of the polygon was near zero. The emission rates showed no diurnal variations, they decreased
during the measurement period (compare Figures 2a and b).
In the polygon depression the soil temperatures in 5 cm depth decreased between July and
September from 8-IO°C to 2-3°C. The methane emissions of the investigated polygon
depression decreased over time with decreasing temperatures from about 120 mg CH 4 *d- 1 *m-2
at the end of July down to 40 mg CH 4 *d- 1 *m- 2 at the beginning of September. The water table
was always near or above the surface. Strong reductive conditions were indicated by high
amounts of reduced iron in pore water samples (70-90 mmolll) in comparison to the polygon
wall (0-5 mmolll). During heavy rainfalls the water table increase and the strong anoxic
conditions leaded to increased emission rates.
At the polygon wall the water table position was always below the surface. At the end of July
the soil temperatures at the polygon wall were lower than the comparable soil temperatures at
the polygon depression. One month later, at the end of August soil temperatures were nearly the
same at polygon wall and polygon depression. The contents of dissolved methane in the soil
pore water from permafrost boundary up to 10 cm below the surface were the same in polygon
wall and polygon depression (5-10 mg CH 4 /1), only in the dryer upper part of the polygon wall
the methane content of the pore water was lower than in the polygon depression part of the
polygon (> 1 mg CH 4 /1, compare Figures 3a and b).
The dependence of the methane emission rates on the water table and the temperature was
confirmed by a multiple regression analysis. A high response of emission rates on soil
temperatures can be shown in 5 cm soil depth (multiple regression coefficient ~: 7.5, error
probability p: 0.0014, so significance level < 1 %) and in 10 cm depth (~: -9.5, p: 0.0251) and
less on water table position (~: -2.0, p: 0.0490). The influence of temperature on the methane
emission rates is also reported by different authors (Harris et aI., 1993: Morrisey and
Livingstone, 1992; Svensson and Rosswall, 1984; Whalen et ai, 1991; Whalen and Reeburg,
1992). In opposite to the CO 2 -Emissions, which show a linear relationship, the CH 4 -emissions
show a logarithmic relationship concerning temperature changes (Moore and Knowles, 1989).
[n other studies the water table position (Funk et ai, 1994; Moore and Knowles, 1989; Moore
and Dalva, 1993; Moore and Roulet, 1993) or the thickness of active layer was more important
(Whalen and Reeburgh, 1992).
The influence of the water table position became obviously during the field experiment: after a
first phase of high gas fluxes the methane emission rate decreased as a result of artificially
lowering of the water table at a polygon depression (compare Figure 4). The high flux might be
333
carbon concentrations in pore water. A more detailled description of all mentioned methods
above can be found in Samarkin et al (1997).
The isotopic composition (13C and D) of the methane gas samples is measured in the Institute
of Geology and Geophysics, University of Leipzig.
Other investigations
The cation contents of the pore water samples were analysed at the Alfred-Wegener-Institut
Potsdam with an ICP OES Optima 3000 (Perkin Elmer Co.) and NBR Standard SRM 1643d.
Results and discussion
Methane emission
Methane emissions from the polygon wall (site 8) and the polygon depression (site 7) differ
strongly from each other: while the mean emission rate of the wet polygon depression was
about 75 mg CH 4 *d- 1 *m- 2 , the mean emission rate of the higher and dryer polygon wall part
of the polygon was near zero. The emission rates showed no diurnal variations, they decreased
during the measurement period (compare Figures 2a and b).
In the polygon depression the soil temperatures in 5 cm depth decreased between July and
September from 8-IO°C to 2-3°C. The methane emissions of the investigated polygon
depression decreased over time with decreasing temperatures from about 120 mg CH 4 *d- 1 *m-2
at the end of July down to 40 mg CH 4 *d- 1 *m- 2 at the beginning of September. The water table
was always near or above the surface. Strong reductive conditions were indicated by high
amounts of reduced iron in pore water samples (70-90 mmolll) in comparison to the polygon
wall (0-5 mmolll). During heavy rainfalls the water table increase and the strong anoxic
conditions leaded to increased emission rates.
At the polygon wall the water table position was always below the surface. At the end of July
the soil temperatures at the polygon wall were lower than the comparable soil temperatures at
the polygon depression. One month later, at the end of August soil temperatures were nearly the
same at polygon wall and polygon depression. The contents of dissolved methane in the soil
pore water from permafrost boundary up to 10 cm below the surface were the same in polygon
wall and polygon depression (5-10 mg CH 4 /1), only in the dryer upper part of the polygon wall
the methane content of the pore water was lower than in the polygon depression part of the
polygon (> 1 mg CH 4 /1, compare Figures 3a and b).
The dependence of the methane emission rates on the water table and the temperature was
confirmed by a multiple regression analysis. A high response of emission rates on soil
temperatures can be shown in 5 cm soil depth (multiple regression coefficient ~: 7.5, error
probability p: 0.0014, so significance level < 1 %) and in 10 cm depth (~: -9.5, p: 0.0251) and
less on water table position (~: -2.0, p: 0.0490). The influence of temperature on the methane
emission rates is also reported by different authors (Harris et aI., 1993: Morrisey and
Livingstone, 1992; Svensson and Rosswall, 1984; Whalen et ai, 1991; Whalen and Reeburg,
1992). In opposite to the CO 2 -Emissions, which show a linear relationship, the CH 4 -emissions
show a logarithmic relationship concerning temperature changes (Moore and Knowles, 1989).
[n other studies the water table position (Funk et ai, 1994; Moore and Knowles, 1989; Moore
and Dalva, 1993; Moore and Roulet, 1993) or the thickness of active layer was more important
(Whalen and Reeburgh, 1992).
The influence of the water table position became obviously during the field experiment: after a
first phase of high gas fluxes the methane emission rate decreased as a result of artificially
lowering of the water table at a polygon depression (compare Figure 4). The high flux might be