Abstract of the lecture.

Reconstruction of the past climate changes by inversion of the present-day subsurface
temperature-depth profiles measured in deep boreholes has established as one of the proxy
methods in paleoclimate research in the last two decades.

The temperature-depth profiles obtained by precise temperature logging in boreholes several
hundred metres to 2 - 3 km deep can be used to reconstruct the ground surface temperature
history on the time scale of several past centuries up to the last glacial/Holocene transition,
respectively.

Heat conduction, considered in the reconstruction technique as the only mechanism of

the heat transport in rocks, is definitely not the only heat transfer mechanism in soil. The
reconstructed ground surface temperature histories are therefore temporal changes of the
ground temperature at the upper boundary of the heat conduction domain, which begins
somewhere in the soil - bedrock transition zone. The long-term relationship between the soil
and surface air temperatures is therefore the key issue in interpreting the results in terms
of the long-term climatic variability. Meteorological factors like precipitation, cloudiness,
wind, soil moisture, thickness and duration of the snow cover and environmental factors like
a type of a vegetation cover are suspected to influence appreciably the air-soil temperature
difference.

In order to explore this relationship, the long-term monitoring of the air-ground
temperature coupling has been launched at the campus of the Institute of
Geophysics in Prague (50_ 02’ 27" N, 14_ 28’ 39" E, 274 m a.s.l.) in 1993. The air-
soil-bedrock temperature observations were complemented by the soil temperature
monitoring under different types of surface (grass, sand, bare soil, asphalt) at the
depths of 2, 5, 10, 20 and 50 cm, as well as by air temperatures at 5 cm above each
of the surface types and at 2 m above the background grass surface in 2003.

The results of this eight year monitoring indicate that the annual mean of the soil –
air temperature difference depends strongly on the type of the surface – grass, sand,
bare soil, asphalt – and amounts up to 4 – 5 °C. The inter-annual variations of the
difference for the given surface are smaller, typically about 1 °C. This variance is of
the same order as the amplitude of the long-term observed and/or estimated surface
temperature changes. However, the observational period of eight years is too short
to detect any statistically relevant trend and to prove or falsify the assumption that on
the decadal scale the annual mean of the difference is constant.

Jan Safanda, Institute of Geophysics Prague, Czech Academy of Sciences

Local : CLAV, Anfiteatro 1 - 23 de Novembro 2011 - 14:30h