Der Einfluß der Landnutzung auf Verdunstung und
Grundwasserneubildung - Modellierungen und Folgerungen für das
Einzugsgebiet des Glan
D. Reichert (July 2001)
Two methods of scenario development as well as modelling spatially distributed simulations have been applied to determine the impact of landuse on evapotranspiration and groundwater recharge. The objective of this approach is to assess the impact of urban and regional planning on the human and natural environment. The question was whether or not these simulation techniques may be used as a Decision Support System.
First, landuse scenarios were developed for a sample catchment, the river Glan in Germany. The scenarios are developed by taking into account the characteristics of the catchment area, current planning and political conditions as well as spatial trends in landuse distribution. The catchment of the river Glan is mainly used for agricultural purposes, however both the relief of the low mountain range and the soil properties adversely affect cultivation. Whilst migration poses a problem in the Northern part of the catchment, suburbanisation increases steadily in the surroundings of Kaiserslautern. Despite of economical and political stimulus for intensive landuse within Europe, recent landuse trends expose a shift from natural grassland to settlement of forest in general. Moreover, a significant conversion from agriculture towards grassland can be noticed within the Glan valley.
Taking all that into account five scenarios for the landuse categories settlement, agriculture and forest were developed. In addition to these realistic scenarios, two extreme-scenarios for the landuse categories settlement and agriculture were included to emphasise hydrological effects. Under the assumption that previous trends and political conditions remain constant, the scenarios indicated an enlargement of the landuse categories settlement and forest as well as a reduction of agricultural land.
The spatial distribution of landuse changes were simulated with the LAnduse change DEvelopment MOdel (LADEMO). The model is capable to estimate the impact of the landuse on evapotranspiration and groundwater recharge with a high degree of accuracy. It provides a synthesis between the overall estimation of parameters, the current landuse and the user-based objectives. Modelling results indicate a significant spatial development of the landuse distribution within the catchment. Particularly, in the flood plain and on the slopes agricultural use has altered to grassland of forest, whereas settlements have enlarged along development axes, such as main roads. According to that the simulation with LADEMO provides decision-makers with information required for fundamental decision-making processes.
The spatially differentiated calculation of evapotranspiration and groundwater recharge was accomplished with the transpiration model TRAIN. The basic parameterisation of each landuse class is significant for the interpretation of results. The parameterisation in TRAIN consists of both temporally variable and altitude related parameter-sets.
Based upon the results of the modelling, it was possible to draw conclusions on the hydrological impact of every landuse class as well as the impact of scenarios. Furthermore, this can be used as basis for various spatial planning projects. Different properties of each landuse class have revealed a characteristic distribution of evapotranspiration and groundwater recharge both in the annual mean and throughout the course of the year. Evapotranspiration of forests was, for example, relatively high compared to that of settlements. In contrast, groundwater recharge of forests was lower than that of settlements. The reason for the latter phenomenon is that currently TRAIN does not account the rates of sealing and the effects of canalisation which both increase runoff.
The results of modelled landuse scenarios reveal a limited impact on the water balance of the entire catchment. The mean annual difference between the two extremes (extreme-agriculture- and extreme-settlement-scenario) was only 16 mm for evapotranspiration (1) and 5 mm for groundwater recharge (2). This is a deviation of less than 5 % from the reference condition and is, therefore, within the permitted error range of modelling. The observed differences refer to particular landuse modelling results. This can be seen as a trend, and thus, may be applied for further conclusions. For example, the impact of the extreme-agriculture-scenario on the local energy budget was calculated. According to this the increased evapotranspiration would require an additional energy amount of 1.17 W / m² which, in consideration of diverse winds, would lead to a local cooling.
The hydrological investigation provides a base for the analysis of landuse trends within the catchment. The entire analysis might be applied to other regions, and can moreover, provide, some new methods in local and regional planning. Thus spatially differentiated modelling allows planning institutions and authorities to implement particular management solutions at critical locations. The value of scenarios is being exposed whenever different (planning) alternatives result in different effects, and scenarios are then employed as a socalled Decision Support System. However, the application of models involves timeconsuming data acquisition, adaptation and processing of large datasets. Thus, limitations of human and financial resources have to be assessed properly prior to any further investigation.
In summary, the simulations in this study have demonstrated that these methods are sensible and suitable for regional planning. Especially the combination of scenarios and simulations should be used for achieving decisions.
(1) Reference condition (evapotranspiration): 517 mm
(2) Reference condition (groundwater recharge): 142 mm
