Impact of land-use and land-management on the water infi ltration capacity of soils on a catchment scale

Abstract

Agriculture is the largest user of the resource soil. So, even small changes in certain soil properties can lead to huge effects on a regional scale. The infiltration capacity is as such an important soil parameter, and also a good indicator of soil quality and soil fertility. Silent sealing, as a result of a negative change in the infiltration capacity due to unfavourable landuse and management, will results in severe effects like faster runoff production and flooding on regional scale,. The assessment of impacts due to land-use or land-management changes on a regional scale is difficult, because detailed information on soil properties and land-management are rarely available. The awareness of the effects of interference in ecosystems is extremely important to supply landscape planners and politicians with information about the impacts of their proposed plans. The aim of this work is the assessment of the maximum water storage capacity of soils under different land-use and land-management situations in a real river catchment (Schunter). Based on field measurements of infiltration under several land-use and land-management situations, a modelling approach has been developed to determine the maximum potential water storagecapacity (S_max). This maximum water storage capacity is closely related to the saturated hydraulic conductivity (K_s), and also a suitable indicator, which can be used to compare different land-use/land-management scenarios. S_max is a theoretical value describing the maximum potential of a given soil/land-use unit. Although, in reality, the water storage is highly variable due to different soils and land-uses on a catchment scale, S_max allows the direct comparison of different soil/land-use units. Since the required input parameters for detailed process models are often not available at a regional scale, general assumptions and simplifications have to be applied in order to provide meaningful statements. In this special case an integrated measure is needed which takes the soil properties in combination to the land-use and the land-management into account. Such an integrated measure can be found as a part in the Curve Number (CN) from the “Curve Number Model" of the National Resource Conservation Service. The CN is a dimensionless value which has been experimentally identified for a variety of different soil, land-use and land-management situations for small scale catchments in the US. The CN is related to the water retention potential (S), and S was originally used to compute the direct runoff from a precipitation event. Since this work addresses only the agricultural viewpoint of impacts of land-use and land-management, the main focus is on the relation of CN to the water retention potential and the computation of S_max. Final runoff computations were not the aim of this work. Knowing the limitations of the Curve Number Model for hydrological questions, runoff has been computed for the year 2002for demonstration purposes only. The CN-Model has often been criticized for its obscure determination of the CN from precipitation/runoff relations, which have not been properly published, not even in the official handbooks. In this work new methods for the determination of the CN have been developed. Now the CN can be directly measured (CN_m) based on field infiltration measurements. Use of the saturated hydraulic conductivity allows the computation of the maximum water storage capacity (S_max) for a given soil, land-use and land-management combination. Since the maximum storage capacity is used, the prevailing wetting status of the soil can also be neglected. On a catchment scale, only a subset of all soil, land-use and land-management situations can be covered by measurements. The remaining situations have to be estimated or be adapted from literature values. The use of pedotransfer functions allow the computation of soil properties (e.g., K_s) based on their textural composition. The performance of the pedotransfer functions in comparison to the field measurements have been tested, resulting in a poor capability of predicting correct values from the pedotransfer functions. The comparison of measured CN_m with published values of the CN performed very well. Based on the CN_m-Model, scenarios of historic (1950), current (2009) and future (2070) land-uses for the Schunter catchment have been computed, showing the direct impact of different land-use situations to the maximum water storage capacity on a regional scale. Although the scenarios are just snapshots, not taking the temporal dimension of land-use changes into account, this method is useful to detect the impacts of land-use and landmanagement changes. This work examined a new method to derive the CNm by infiltration measurements in the field. The experimental determination of the CNm allows the update of existing curve numbers for special situations not covered in the handbook. Also, the application of the CN concept to German soils is now possible. The computation of the maximum potential storage capacity (S_max) is a useful measure to identify the impacts and to compare land-use and landmanagement scenarios. The impact of land-use and land-management changes on a catchment scale has been clearly demonstrated. Compared to the situation in 1950, in the year 2009 the maximum water storage capacity has decreased by 17 %. Projecting a similar land-use change of the past 60 years into the future will result in a loss of water storage capacity of 19 % compared to 1950. The model approach offers a useful tool for landscape analysis. Due to the manifold different landmanagement practices in agriculture, additional measurements should be performed in the future.

Auf Wunsch des Autors / der Autorin ist diese Dissertation nur als Druckausgabe verfügbar.