Influences of Soil Frost on Groundwater Flowpaths during Spring Flood in Boreal Spruce Forest
Lars Nyberg, Kevin Bishop, and Goran Lindstrom

To improve the process knowledge on how soil frost effects spring runoff in forest soils at the hillslope scale, as well as to evaluate the influences of soil frost on tree growth in the boreal forest ecosystem, a measurement program was established at the Svartberget Research Park at 64o in northern Sweden. The measurement program includes continuous monitoring of unfrozenwater content and bulk soil conductivity with TDR, water potential by gypsum blocks and soil temperature, as well as groundwater levels and snow depth. Four soil profiles are located in a transect at parallel to the direction of lateral subsurface flow to a small stream. The profiles cover the transition from upslope, mineral soils, to organic-rich riparian soils adjacent to the stream. The dynamics of water and heat flow are being modelled using the physically based SOIL model, which basically simulates one-dimensional vertical flows of water and heat, but also allows for lateral groundwater flow. The model results and the knowledge gained on the processes involved will be used to assist a conceptual formulation of soil frost in the HBV runoff model which is used extensively in Sweden and other countries for operational forecasting of spring runoff for the hydropower industry.

The measurements from the first winter revealed the potential of extending TDR for studies of soil frost from the profile to the hillslope scale. The start of the measurements in October 1995 were followed by an unnormally cold winter with a thin snow layer, which resulted in deeper soil frost than normal. The soil fost depth ranged from 30 cm in organic soil at 4 m from the stream to 80 cm in a well-drained podsol profile at 22 m distance. The freezing started in the superficial layers and continuously affected deeper layers. The unfrozen water storage in the top 30 cm in the profile 4 m from the stream decreased from 130 mm in october to 40 mm in March when the soil frost had its maximum. In the 30-70 cm layer, the unfrozen storage decreased from 180 mm to 145 mm. In June -96 the soil frost was gone and the unfrozen water storage was similar to that in the previous autumn. In the podsol profile at 22 m distance from the stream, the soil frost decreased the unfrozen water storage in the 0-30 cm layer from 50 to 20 mm, and from 65 to 25 mm in the 30-70 cm layer. The unfrozen residual, which was adsorbed to solid surfaces or held in the smallest pores, was close to 5 vol-% in the mineral soil while it was around 15 vol-% in the organic soil.

Lars Nyberg
Institute of Earth Sciences, Hydrology
Uppsala University
c/o MFG
Sjoleden 9
S-665  91 Kil
SWEDEN

Kevin Bishop
Sweden Department of Forest Ecology 
Swedish University of Agricultural Sciences
S-901 83 Umea
SWEDEN

Goran Lindstrom
Swedish Meteorological and Hydrological Institute 
S-601 76 Norrkoping
SWEDEN