It. Kukkonen et C. Clauser, SIMULATION OF HEAT-TRANSFER AT THE KOLA DEEP-HOLE SITE - IMPLICATIONSFOR ADVECTION, HEAT REFRACTION AND PALEOCLIMATIC EFFECTS, Geophysical journal international, 116(2), 1994, pp. 409-420
The drill hole SG-3, 12 261 m deep in the Pechenga-Zapolyarny area, Ko
la Peninsula, Russia, is currently the deepest drill hole in the world
. Geothermal measurements in the here reveal a considerable variation
(30-68 m W m(-2)) with depth in the vertical component of heat-flow de
nsity (HFD). We simulate heat and fluid flow in the bedrock structure
of the Kola deep-hole site. Various potential sources for the observed
HFD variation are discussed, with special emphasis on advective heat
transfer, palaeoclimatic ground surface-temperature changes and refrac
tion of heat flow due to thermal conductivity contrasts. A 2-D finite-
difference (FD) porous-medium model of the Kola structure, constructed
from all available data on lithology, hydrogeology, topography, therm
al conductivity and heat-production rate in the deep-drilling area, is
the basis of all forward-model calculations. A conductive, steady-sta
te simulation indicates that heat production and refraction create a v
ariation of about 15 m W m(-2) in the uppermost 15 km, but are insuffi
cient to reproduce the measured HFD-depth curve in the uppermost 2-4 k
m. However, if topography-driven groundwater how is considered in the
model, the measured HFD variation is easily explained. The most sensit
ive parameters in fitting the model results to the observed HFD-depth
curve are the permeability of the top 4 km (10(-14)-10(-15) m(2)) and
the (constant) HFD applied at the base of the model at 15 km depth (40
-50 m W m(-2)). The palaeoclimatic effect for the Kola structure was c
alculated with a conductive transient simulation. A simplified ground
surface-temperature history (GTH) of the Kola area was simulated by va
rying the surface temperatures of the model during different intervals
of the simulation. Our results indicate that the measured variation i
n the vertical HFD cannot be explained by the palaeoclimatic effect al
one, because its amplitude decreases rapidly from about 16 m W m(-2) n
ear the surface to less than 2 m W m(-2) at depths in excess of 1.5 km
.