195greater than 1 bar and these waters degas spontaneously with
bubbles formation when they are brought to atmosphere
(0.88 bar at the elevation of the lake surface), CH 4 being
mainly responsible of this sparkling effect.
11.5.2 Gas Exchange with Atmosphere
Both CO 2 and CH 4 present in high concentrations in the
monimolimnion are greenhouse gases and it will be interest-
ing to check if these gases are emitted to atmosphere.
11.5.2.1 Methane
Figure 11.5 shows that methane concentrations present a
minimum close to 10 nM at 45–50 m depth. Thus, the upward
advective-diffusive flow of CH 4 vanishes at this depth. We
shall see in the following Sect. 11.5.3.3 that CH 4 is almost
completely oxidized into CO 2 at this depth.
The oversaturation of methane in the surface waters of the
lake may be related to a production in the sediment in shal-
low areas (Bastviken et al. 2004 , 2008 ). However, in Pavin
Lake, due to the shape of the basin, epilimnetic sediments
represent a weak part of the total sediment area. Alternatively,
another explanation for this oversaturation, also encountered
in ocean surface water and known as the methane paradox,
involves CH 4 production in oxygenated superficial water
(Carini et al. 2014 ). In this article, methane production may
proceed from bacterial degradation of methylphosphonic
acid, which is synthetized by marine archaea. Oxic
methanogenesis is also reported in lakes such as Lake
Stechlin (Germany), where the CH 4 production is performed
through an interspecific transfer of H 2 and/or acetate from
photoautotrophs such as cyanobacteria to attached archaea
(Grossart et al. 2011 ). Such syntrophic associations between
archaea and bacteria may be responsible of an important
methane emission from aquatic systems (Bogard et al. 2014 ).
In June 2010 a weak emission (about 58 micromoles.
md−−^21. on average, i.e. 0 .. 026 kmold−^1 for the whole sur-
face) was observed at the lake surface by direct measure-
ments using a floating chamber and laser sensor (Guimbaud
et al. 2011 ). Alternatively, a mean CH 4 escape ∼ 10 −−^31 kmol.d
can be derived from the model of the lake presented by Lopes
et al. ( 2011 ). This value is much lower than those obtained in
shallow lakes (Bastviken et al. 2008 ) for which values range
from 200 to 1000 micromoles. md−−^21. , but is very similar to
those from others meromictic lakes (Borges et al. 2011 ).
Despite their usually high methane content in the bottom
layer, meromictic lakes are then weak CH 4 emitters com-
pared to classical lakes. This may be related to the perma-
nently stratified water column that limits the diffusion
(through very low eddy diffusion coefficients in the chemo-
cline, close to molecular diffusion) and well established bac-
teria consortia in the redoxcline that oxidize almost all the
methane diffusing from the deep compartment.11.5.2.2 Carbon Dioxide
To quantify the gas exchange at the air water interface, it was
necessary to know (i) the partial pressure of the gas in the
surface water which depends on the photosynthetic activity,
(ii) the kinetic constant of the gas exchange with depends on
the wind velocity. A quantitative treatment of this question
needs a continuous determination of these two parameters
which is out the scope of the present paper and necessitates the
development of specific sensors (Lefèvre et al. 1993 ; Prévot
et al. 1998 ). Only some indicative results are given here.
p(CO 2 ) was determined by calculations from Alk and pH
measurements. pH measurements in natural, low ionic
strength waters have a limited precision of about ±01. unit
and accuracy on p(CO 2 ) is then about 25 %. Nevertheless, asFig. 11.10 Gas
pressures compared to
hydrostatic and
atmospheric pressures
11 Carbon Cycle in a Meromictic Crater Lake: Lake Pavin, France