Wintertime Surface Heat Exchange in a Boreal Forest:
From the Plot to the Stand
Manfred Sthli1, Mikaell Ottosson-Lfvenius2, Per-Erik Mellander2,
and Kevin Bishop2
The present study aims to quantify the heat exchange for a 10-km2 boreal forest in the north of
Sweden (64 N latitude) during wintertime. The forest is a composite of old spruce stands, heath, and
pine stands; it includes different soil types and open clearings. A GIS-database with extended infor-
mation on vegetation, soil type, and topography is available. Detailed physical measurements have
been conducted continuously at single selected points during the last three winters, and a spatial
snow and soil frost monitoring program within the different stands ran last winter. All these data are
used for the parameterization and validation of a numerical model (WINSOIL) to quantify the heat
balance. The model is one-dimensional with a single snow layer and a sophisticated description of
the soil frost. Our simulation strategy is to conduct multi-run simulations where some governing
parameters vary randomly within an (observed or estimated) reasonable interval. The results are
presented as means and standard deviations of a large number of simulated outputs, such as frost
depth or temperatures. This gives us an idea of the spatial variability within the forest investigated.
The first multi-run simulations confirm the considerable variability of surface temperature and frost
depth measured at different locations both within the stands and between different stands. A model
validation with ground measurements demands an enormous amount of work for such an extended
area. A direct validation of the simulated turbulent heat fluxes with eddy correlation systems is
planned for the coming winter. Finally, we will be discussing how the multi-run method could be
used to improve the parameterization of GCM models with a grid size of some km2 applied to
northern Scandinavian winter conditions.
1
ETH Zrich, Institute of Terrestrial Ecology, Grabenstrasse 3, CH-8952 Schlieren, Switzerland
2
Forest Research Station Svartberget, S-922 91 Vindeln, Sweden
42
Previous Page
