Physics-based numerical modeling of surface-groundwater flow and transport at catchment scale

DSpace Repositorium (Manakin basiert)


Dateien:

Zitierfähiger Link (URI): http://nbn-resolving.de/urn:nbn:de:bsz:21-opus-57135
http://hdl.handle.net/10900/49547
Dokumentart: Dissertation
Erscheinungsdatum: 2011
Sprache: Englisch
Fakultät: 7 Mathematisch-Naturwissenschaftliche Fakultät
Fachbereich: Geographie, Geoökologie, Geowissenschaft
Gutachter: Cirpka, Olaf (Prof. Dr.)
Tag der mündl. Prüfung: 2011-06-24
DDC-Klassifikation: 550 - Geowissenschaften
Schlagworte: Angewandte Geowissenschaften , Geowissenschaften , Grundwasser , Oberflächenwasser , Landnutzung
Freie Schlagwörter:
Hydrologcial Modeling , Surface-Subsurface , Agricultural Catchment , Groundwater , Land-Use Changes
Lizenz: http://tobias-lib.uni-tuebingen.de/doku/lic_ohne_pod.php?la=de http://tobias-lib.uni-tuebingen.de/doku/lic_ohne_pod.php?la=en
Zur Langanzeige

Inhaltszusammenfassung:

Die Veränderung von Landnutzung und die Intensivierung von Bewässerung und Düngung bedingen räumliche Wechsel, die das Oberflächenwasser und das Grundwasser stark beeinflussen. Daher benötigt die Darstellung dieser Transformation eines Modells, welches die Komplexität sowohl von Oberflächen- als auch von Grundwasser erfasst. Empirische und konzeptionelle hydrologische Modelle berücksichtigen physische Prozesse im Einzugsgebiet generell nicht explizit. Daher sind ihre Fähigkeiten Auswirkungen von veränderter Landnutzung und Entwicklungen von Bewässerung und Düngung vorauszusehen fraglich. Außerdem fehlt es meist an einer detaillierten Beschreibung gelöster Stoffe in Oberflächen- und Grundwasser. Physics-based models sind daher die beste Variante, um den Übergang in einen Wassereinzugsgebiet zur Bewässerungslandwirtschaft zu simulieren. Diese Art der Modelle verwendet einen Satz von gekoppelten partiellen Differentialgleichungen aus den Grundsätzen der Erhaltung der Masse und des Impuls von Wasser, um Strömung und Transportprozesse in den Einzugsgebieten zu beschreiben. Aus der Überprüfung von bisherigen Studien geht hervor, dass sich diese entweder nur mit natürlichen Wassereinzugsgebieten ohne anthropogenen Einfluss befassen oder es zu keiner Berücksichtigung von Änderungen der Landnutzung kommt. Erstere beziehen sich nur auf den Oberflächenabfluss. Ihre Methoden, welche sich teilweise stark unterscheiden, lassen sich nur auf ihre besonderen Fallstudien anwenden und können nicht auf andere Studien übertragen werden. In dieser Untersuchung schlage ich deshalb eine allgemeine Methodik für den Einsatz von Physics-based models vor. Die Methodik wird angewandt, um die Auswirkungen von Änderungen in der Landnutzung in dem semiariden Einzugsgebiet von Lerma (Tal de Ebro, Spanien) zu untersuchen. Hier vollzog sich in den letzten Jahren ein Wandlungsprozess vom Regenfeldbau zur Bewässerungswirtschaft. Seit vier Jahren wurde dieser Übergang des früher unkultivierten Landes zum bewässerten Ackerboden intensiv überwacht. Hierzu wurde ein rechnerisch effizienter Hang Äquivalent eingeführt, um die das Einzugsgebiet kontrollierenden Faktoren zu identifizieren, eine ausreichendes Kalibrierungsproblem aufzustellen und die Bewertung von komplexen Oberflächen-Untergrund-Modellen zu beschleunigen. Die Methode ist erfolgreich geprüft und wurde angewandt in der Kalibrierung des Physics-based model des Lerma-Bassins. Das kalibrierte und validierte Physics-based 3-D gekoppelte Modell wird verwendet, um die Wirkung des Übergangs in der Landwirtschaft anhand der hydrologischen Eigenschaften des Einzugsgebiets zu untersuchen. Die Fallstudie wurde in einem Gebiet durchgeführt, wo die räumlich-zeitlichen Daten von Anbaustrukturen, Bewässerung und Düngergaben in hohem Maße verfügbar sind. Die Bewertung der Ergebnisse zeigt, dass das Physics-based model die Auswirkung der Transformation von Landnutzung und Bewässerung auf die Strömungsdynamik des Grundwassers mit hoher Genauigkeit simulieren und vorhersagen kann. Anfälligkeit und Korrelationsanalysen über dem kalibrierten Modell Parametervektor zeigen, dass die kalibrierten Parameter identifizierbar sind und lokal einzigartig für die Parameter-Zonierung. Definiert wird diese durch Informationen über die lithologischen Einheiten und Textur-Daten. Um Veränderungen in der Entwicklung des Abflussverlaufes und der Funktionen des Einzugsgebietes zu zeigen, wurden die gesamte Stromlänge und die durchschnittliche Infiltrations Kapazität durch das Modell untersucht. Die Ergebnisse zeigen, dass die Bewässerung der Landwirtschaft das Ausgangsniveau des Grundwasserspiegels des Lerma Aquifers gehoben hat, was eine Vergrößerung und die ganzjährige Anwendung des Drainage- Netzwerkes notwendig macht. Eine ungefähre Infiltrationskapazität wird hier eingeführt, um die Entwicklung des Abflusses bei der Transformation zu studieren. Und doch werden räumlich verteilte Werte von exfiltrierenden Abflüssen beim Modell direkt erhalten. Sowohl die ungefähre Infiltrationskapazitätskurve als auch der Beitrag der Exfiltration zu Strömung sind konsistent. Sie geben eine Verschiebung vom Hortonschen- zum Dunneschen-Abflussprozess an, die aufgrund der Änderung der Flächennutzung und der Intensivierung der Bewässerung ausgelöst wird. Ein konservatives Transport Modell ist in dem erstellten Abflussmodell des Lerma Bassin integriert, um anhand der Auswirkungen der Transformation auch auf die Wasserqualität schließen zu können und um Konzentrationen unter verschiedenen Wasserwirtschafts -Szenarien vorherzusagen. Die Ergebnisse zeigen auf, dass es möglich ist trotz des Fehlens von Daten - rund um den Nitrogen Transformationsprozess - und vieler vereinfachter Annahmen mit einem schematisierten Physics-based Approach den Transport in den Einzugsgebieten zu simulieren.

Abstract:

Agriculture representing the main user of water and one of the largest anthropogenic sources of nitrate pollutants, understanding irrigation and fertilization impacts has become a priority issue in water management. The land-use transformation and the intensification of irrigation and fertilization involves spatially distributed changes affecting the hydrological fluxes across the surface and subsurface interface. Hence, the simulation of this transformation requires a model that can deal with complexity in both surface and subsurface. Empirical and conceptual hydrological models do not explicitly consider physical processes taking place within the catchment so that their ability to evaluate the impact of land-use changes and the evolution of irrigation and fertilization may be questionable. Furthermore, they also fail to provide detailed description of fluxes of solutes in the surface and subsurface. Physics-based models arises as the best alternative to simulate catchment transition into irrigation agriculture. This type of models use a set of coupled partial differential equations derived from the principles of conserving mass and momentum of water to describe the flow and transport processes at catchment scale. A review of previous studies reveals that they are either associated with natural watersheds without anthropogenic influence, only investigate surface response, or do not account for land-use changes. Their methodologies, that commonly widely differ from each other, typically hold only for their particular case studies and cannot be extrapolated to other studies. In this investigation, I propose a general methodology for the use of physics-based models. The methodology is applied to investigate the effects of land-use change in the semi-arid Lerma basin (Ebro valley, Spain), which underwent a transition from rainfed towards irrigation agriculture. For four consecutive years, this transition of formerly uncultivated land to irrigated farmland was intensively monitored. A computationally efficient hillslope equivalent is introduced to identify catchment controlling factors, set up an adequate calibration problem and speed-up the validation of complex surface-subsurface models. The method is successfully tested, and used in the calibration of the physics-based model of the Lerma basin. The calibrated and validated, physics-based, 3-D fully-coupled model is used to study the effect of the transition into agriculture on the hydrological functioning of the basin for this case study, where spatio-temporal data on cropping patterns, irrigation and fertilizer amounts, and the associated catchment response are available with considerably high resolution. Validation results show that the physics-based model is able to simulate and predict the impact of the land-use transformation and irrigation on surface and subsurface flow dynamics with high accuracy. Sensitivity and correlation analyses about the calibrated model parameter vector indicate that the set of calibrated parameters is identifiable and locally unique for the parameter zonation defined using information on lithological units and texture data. In order to indicate changes in the runoff generation process and catchment functioning, the evolution of the total stream length and the average infiltration capacity provided by the model were studied. The results show that irrigation agriculture has raised the base level of the water table of the Lerma aquifer causing new portions of the drainage network to become perennial. An approximate infiltration capacity is introduced to study the evolution of the runoff-generation processes during the transformation. Yet, spatially distributed values of exfiltrating fluxes are directly obtained from the model. Both the approximate infiltration capacity curve and the contribution of exfiltration to stream flow are consistent in indicating a shift from Hortonian towards Dunne flow runoff generating processes triggered by the land-use change and the intensification of irrigation. A conservative transport model is integrated to the validated flow model of the Lerma basin in order to simulate the impact of the transformation also on water quality, and to predict concentrations under different water management scenarios. Validation results indicates that despite the lack of data to describe the nitrogen transformation processes and many simplifying assumptions, it is possible to simulate transport at catchment systems with a parsimonious physics-based approach.

Das Dokument erscheint in: