Numerical Analysis of Transpiration Influence with Reforested Trees on Fluctuation of Groundwater Level in Mu Us Desert, China OHTE, Nobuhito*, KoBASHI, Sumizi*, Yu, Yi** and YAO, Hong-Ling**
Fig. 1 Location of Mu Us Desert in China (Yamamoto"), 1989) Notes: An arrow indicates Mu Us Shamo Research Center.
Fig. 2 A schematic diagram of the model DSLM: Dry Sand Layer Model, R: Rainfall, ES: Evaporation from sand surface, Re: Recharge into the infiltration zone GLM: Groundwater Level simulation Model, Lr: Absorption with the roots, Tr: Transpiration, Si: Groundwater runoff to the outside of the system, d: Dry sand zone, r: Root zone, i: Infiltration zone, g: Groundwater level
Fig. 3 A scheme of GLM d: Dry sand layer*, r: Root zone, i: Infiltration zone, g: Groundwater level, Ca: Crown area, Lai: Leaf area index, Ra: Crosssectional area of the root zone *Assumed to be constant as the ma ximum thickness of the dry sand layer dmax in this model. Fig. 4 Relationship between the declining coefficient of absorption Res and pf value
Table 1 Several parameters for the boundary conditions Fig. 5 Modeling of the soil layer at Point A and B d: Dry sand layer, r: Root zone, i: Infiltration zone Notes: The hatched area at Point A indicates the silty sand layer. In this figure, the ground water levels of Point A and B are the initial conditions.
Table 2 Parameters of the soil moisture properties for BROOKS and COREY type formula Fig. 6 The soil moisture characteristics of the dune sand and silty sand Table 3 Roots absorption-transpiration parameters of trees at Point B A ratio of the rate of the transpiration to the evaporation rate from free water under the condition without water stress on the trees.
Fig. 7 The calculated results of the evaporation rate from the sand surface, the thickness of the dry sand layer and the infiltration rate a: April 1st October 15th 1988 b: June 1st July 31 st 1988 R: Rainfall, ES: Evaporation from sand surface, s: Thickness of the dry sand layer, Re: Recharge into the infiltration zone
Table 4 Estimated water budget on the desert surface with DSLM R: Rainfall, EP: Evaporation from free water, ES: Evaporation from sand surface infiltration zone., Re: Recharge into the Fig. 8 The calculated result of the groundwater level at point A G. W. L: Groundwater level(depth below the surface) Notes: The fine line indicates the observed values. The bold line is the calculated values. Ca ~l and Ca ~ 3 from highest line respective- ly. Fig. 9 The calculated result of the groundwater level at Point B G.W. L: Groundwater level(depth below the surface) Notes: The lines indicate the calculated fluctuations under the conditions of non-transpiration,
Fig. 10 The calculated fluctuations of the transpiration rate Tr and the declining coefficient of absorption Res Notes: Ca ~ 3 and Ca ~5 indicate three and five times of the crown area value in Table 3.
12) MOLTZ, F. J.: Models of water transport in the soilplant system: A review. Water Resources Res. 7: 1245-1260, 1981 1) BROOKS, R. H. and COREY, A. T.: Hydraulic properties of porous media. Hydrology Papers, Cololado State Univ. 3: 1-27, 1964 2) FREEZE, R. A.: Three-dimensional, Transient, saturated-unsaturated flow in a groundwater basin. Water Resources Res. 7: 347-366, 1971 16) SUZUKI, M., KOBASHI, S., Y U, Y. and Y AO, H.: Observations of gruondwater level in the Mu Us Shamo desert. Proc. of 1989 International Symposium for Japan and China Joint Research,"Analysis of Mechanism and Movement of Desertification of the Arid Land Areas in China": 62-65. 1989 15) RICHARDS, L. A.: Capillary conduction of liquids through porous mediums. Physics 1: 318-333, 1931 Summary In order to evaluate the influence of water consumption by the planted trees on the fluctuation of groundwater level in Mu Us desert in Inner Mongolia, China, a numerical simulation model was developed. The model consists of two parts: 1) Dry Sand Layer Model(DSLM); 2) Groundwater Level simulation Model(GLM) with the sub model which represents the absorbing distribution of the trees. Inputting some meteorological data, the timesequential changes of the soil moisture, the roots absorbing characteristics and the groundwater level could be simulated by GLM using the infiltration rate below the dry sand layer calculated by DSLM. Using three factors
observed and surveyed that were the meteorological data, the soil moisture properties and the transpiration characteristics of planted trees, we tried to reproduce the fluctuation of groundwater level observed at the non-tree condition and to simulate the changes of ground water level at the semi-fixed dune. According to the former calculation result, this model could reproduce the observed ground water level by relevant evaluations of the soil moisture properties, the infiltration rate and the sink intensity from the groundwater zone. The simulated results suggest that even without direct absorption form the ground water body, the trees planted excessively high density would decline the groundwater level from year to year through the water consumption from unsaturated zone.