Seasonal thermal signatures of heat transfer by water exchange in an underground vault
Résumé
The temperature of a 10-point vertical profile at the rock–atmosphere interface has been monitored
since 2000 September in an underground vault at Aburatsubo, Japan, where resistivity
variations have been reported in association with earthquakes. The non-ventilated vault is
characterized by an annual temperature variation of about 1.2 ◦C peak to peak, compatible
with thermal diffusion in the surrounding tuff rock, and by a long-term temperature increase
of about 0.1 ◦C per year, possibly due to a local or global climate change. Owing to a careful
relative calibration of the 10 thermistors used in this experiment, these data establish that the
ceiling temperature is higher than the floor temperature by 0.04 to 0.28 ◦C. Transient temperature
variations are observed in association with human presence or with typhoons, with a
characteristic spatial pattern revealing structural heterogeneity.Variations with periods ranging
from 1 day to 1 week, with an amplitude two time larger and a phase advance on the floor with
respect to the ceiling, are observed from November to May. Variations with periods larger than
1 week, with an amplitude two times smaller and a phase lag on the floor with respect to the
ceiling, are observed from June to October. These cycles are linked to the sign of the seasonal
heat flux. We propose an interpretation in which heat transfer in the cavity is dominated by
diffusion of water vapour from June to October (heat flux downwards, summer regime) and
by convective water transport from November to May (heat flux upwards, winter regime). The
water flow inferred from this model can be used to predict the water saturation of the rock as
a function of time. Because of a permanent radiative heat flux from top to bottom, the upward
water flow in the winter regime is larger than the downward water flow in the summer regime,
resulting in a slow depletion of water from the rock below the cavity. This unbalanced water
flow could contribute to an observed steady secular increase of rock resistivity, and possibly
also to the long-term temperature increase of 0.1 ◦C per year. It is important to understand
these processes in the context of underground geophysical observatories, underground waste
storage and contaminant transport, as well as for the preservation of cultural items such as cave
paintings.
Origine | Fichiers éditeurs autorisés sur une archive ouverte |
---|
Loading...