Uni-Logo
You are here: Home Publications Abstract "Heft 35"
Document Actions

Abstract "Heft 35"


Freiburger Bodenkundliche Abhandlungen

Schriftenreihe des

Institut für Bodenkunde und Waldernährungslehre
der Albert-Ludwigs-Universität Freiburg i.Br.
Schriftleitung: F. Hädrich


Heft 35


Lothar Zimmermann

Der Bodenwasserhaushalt an einem
Hochlagenstandort im Südschwarzwald



Freiburg im Breisgau 1995

ISSN 0344-2691


Summary:

This thesis was conceived within the scope of the forest ecological project ARINUS. Parameters of the water cycle were recorded at the ARINUS-site Sohluchsee (Black forest, SW Germany) during the period 1987-1994, both on the scales of catchments and experimental plots in the forest stand.

The aim was to quantify and model each component of the water balance. For this purpose a physically deterministic soil water transport model was adapted to receive a higher temporal and spatial resolution of water fluxes in the experimental plots. The representativity of the calculated water fluxes on the experimental plots was to be checked by means of a comparison with the water balance of the catchment. After successful adaptation to the site the model was to be used to simulate the soil water balance retrospectively from weather serv-ice data for the years 1962 to 1987 to Interpret better occurrence and course of the montane needle yellowing of spruce and forest growth results.

The podsol at the examined mountain range site (mean annual precipitation 1800 mm, mean annual temperature 4.5°C) had developed itself from the saprolite of the base-poor Bärhalde granite. Due to the sandy-loamy texture and the high skeleton percentage of 60 to 80 %, the Proportion of pores ?<1O µm is 50% of total porosity. The saturated hydraulic conductivity is with 200 cm/d throughout the whole soil profile very high. The available field capacity is situated in a medium range between 130 and 160 mm/1 m soil depth. During all seven years of the measurement period matric potential in the mineral soil did not exceed pF=2.74 (= -550 cm WC) according to tensiometer recordings. Thus, the mountain range site is characterized by a good water supply which was expected according to the capacitive water balance terms. Nevertheless, noticeable drying Symptoms were observed in the O-layer during the dry summers 1991 and 1992 with TDR-probes. Peak values of volumetric water content around 9 vol% were measured in this compartment.

The model WHNSIM, which was used for the Simulation of the soil water fluxes, is based on an one-dimensional solution of the RICHARJDS-equation. Interflow during saturated conditions is taken account of by the source and sink term. As the model is designed for agricultural sites, the evapotranspiration part had to be changed for the forest Situation. Potential evapotranspiration was calculated by means of the formula by HAUDE including a forest and altitude factor. Interception was gained by the difference between open field precipitation and throughfall. From a comparison with daily values of potential evapotranspiration, rates of potential transpiration were derived which were reduced to actual transpiration using a soil moisture dependent function.

During the measurement period open field precipitation ranged from 1410 mm yr"1 to 2160 mm yr-1. Interception was with 16% of open field precipitation very low. During the measurement period actual evapotranspiration averaged 520 mm yr-1 , whereas 42% were actual transpiration. The high percentage of interception explains itself from the perhumid climate conditions of the mountain range site. Average runoff was 1390 mm yr-1, which came quite near to a seepage of 1338 mm-1 calculated for the plot scale. The deviation of the watershed water balance laid with 4.4% of open field precipitation in the range of the error of precipitation measurement, so that through the long-term balance the magnitude of the calculated evapotranspiration can be taken for sure.

Calibration of the model was realized by soil physical parameter functions. As the measurement of hydraulic conductivity K(?) under field conditions is linked with a high uncertainty and the tranfer error between laboratory and field Situation is very high, the functions measured in the laboratory were used to give starting points for the iterative calibration. The optimization criteria was the comparison between matric potentials simulated by the model with measured values. A sigmoid-shaped curve was found to be the best parametrization for the pF-curve ?M(T). Values for hydraulic conductivity lay for the upper soil in the range of the laboratory measurements, while for the deeper soil lower values had to be calibrated. In a second step, simulated water Contents were compared with measured values. With a simple transfer of the laboratory pF-curve, the simulated water contents could be adapted to TDR-measurements. Accordingly, the measured soil water in the organic layer can be used for physiological interpretation. The validation of both quantities ?M and T was achieved with an independent data set covering several years of highly different weather conditions. By the comparison of simulated seepage rates with results from a formerly applied soil water balance model and rates of watershed runoff the soil water model was additionally validated. The annual balances resulting from both models and the comparison of seepage rates with watershed runoff showed little, non-systematic deviation. In addition, the actual transpiration rates which were calculated from the model could be compared with measurements of xylem sap flow in the year 1994. As a result, the sums exhibited only a deviation of 10% from each other. Most of the time the transpiration rates followed the course of the measured xylem sap flow. Through multiple control with independently measured quantities the water transport model is very well validated for the site.

For retrospective Simulation, daily values of precipitation, relative air humidity and air temperature of adjacent weather Service stations had to be transferred to the experimental site. As a longer meteorological record existed for the site, regression models could be established. The assumptions for retrospective Simulation were tested by application on the measurement period and compared with measured values. The model assumptions for the retrospective Simulation period performed the measurement values very well.

For the whole Simulation period from 1962 to 1987 the retrospective model ran demonstrated a distinct drought in the upper soil only for the years 1976, 1983 and 1985. In the years 1976 and 1983 the drying period occurred already at the beginning of the Vegetation period. The drought in these years was discussed in literature as triggering and synchronizing factor for the montane needle yellowing of spruce. Nevertheless, similar peaks of dry-ness in the organic layer were also reached in the dry years 1990 till 1992 without resulting in yellowing or low Mg-needle levels. The only difference compared to the earlier years 1976 und 1983 laid in the later date of the occurence of the dryness relative to the Vegetation period. The montane needle-yellowing is causally combined with an insufficient supply of magnesium. At the base-poor study site Schluchsee, the main storage of available Mg is in the organic surface layer containing 60% of the fine root mass. During the early summer growth period with high nutrient demand the dry-out of the top soil reduces the nutrient supply and the uptake through the fine roots. Hence, early summer dry-out of the top soil is evident as triggering cause for the montane needle yellowing of spruce. Qualitative com-parisons of water balance quantities with radial growth during the Vegetation period gave no direct functional relationship. At the perhumid mountain range site Schluchsee even in dry years like 1983 there is a sufficient amount of plant available soil water in the rooted mineral soil. No drought stress Situation for the tree stand is reached. With the simulated soil water regime for past years a data base was created, which enables an extended analysis of tree growth on the ARINUS-experimental site.


Personal tools