A hydrological programming language extension for integrated catchment models

Abstract

In the last decade, the application of classical hydrological rainfall runoff models is disputed as a valid method for understanding water transport in landscapes (eg. Beven 2002 and 2006, Seibert and McDonnell 2002, Sivapalan et al. 2003, Kirchner 2006, Tetzlaff et al. 2008), since current models lack the ability for formulation and rejection of hypotheses. Vache and McDonnell (2006) developed a framework for the rejection of model structures, where each model structure is understood as a hypothesis on runoff generation. This thesis presents a newly developed generalized form of the Vache-McDonnell rejectionist framework, called the Catchment Modelling Framework (CMF). It is an extension to the programming language Python, offering a toolkit for the set up of a wide range of hydrological models, following the finite volume approach by Qu and Duffy (2007). Buytaert et al. (2008) and Clark et al. (2011) call also for such modular frameworks as a tool for the formulation, implementation, test and rejection of process hypotheses. Buytaert et al. (2008) demand such frameworks to be portable, accessible and modular. While hydrologists debate the theoretical application limits of runoff models for solute transport in landscapes, a growing demand of integrated landscape models for the integration of lateral transport of matter by runoff arises in interdisciplinary projects, like eg. the NitroEurope IP (EC). With this background, modular hydrological frameworks need to be designed for simplified data exchange during the model runtime for interfacing the hydrology with models from different disciplines, like CMF. CMF is portable, accessible and modular as an open source extension to the Python language (see Chapter I and II) and can be used for the formulation of different hypotheses (Chapter III). The principle of the connection of CMF with models from different disciplines is shown in Chapter IV, and Chapter V shows the relevance of tightly connected models of transport and turnover for the emission of greenhouse gases from ecosystems.Beven, K., 2006. Searching for the Holy Grail of scientific hydrology. Hydrol.Earth Syst.Sci 10, 609-618.Beven, K.J., 2002. Towards an alternative blueprint for a physically-based digitally simulated hydrologic response modelling system. Hydrol.Proc. 16, 189-206.Buytaert, W., Reusser, D., Krause, S., Renaud, J.P., 2008. Why can t we do better than Topmodel? Hydrol.Proc. 22, 4175-4179.Clark, M.P., Kavetski, D., Fenicia, F., 2011. Pursuing the method of multiple working hypotheses for hydrological modeling. Water Resour.Res 47.Kirchner, J., 2006. Getting the right answers for the right reasons: Linking measurements, analyses, and models to advance the science of hydrology. Water Resour.Res. 42_ , W03S04, doi:10.1029/2005WR004362.Qu, Y.Z., Duffy, C.J., 2007. A semidiscrete finite volume formulation for multiprocess watershed simulation. Water Resour.Res. 43, W08419, doi:10.1029/2006WR005752.Seibert, J., McDonnell, J.J., 2002. On the dialog between experimentalist and modeler in catchment hydrology: Use of soft data for multicriteria model calibration. Water Resour.Res. 38, doi: 10.1029/2001WR000978.Sivapalan, M., 2003. Process complexity at hillslope scale, process simplicity at the watershed scale: is there a connection? Hydrol.Proc. 17 _, 1037-1041.Tetzlaff, D., McDonnell, J.J., Uhlenbrook, S., McGuire, K.J., Bogaart, P.W., Naef, F., Baird, A.J., Dunn, S.M., Soulsby, C., 2008. Conceptualizing catchment processes: simply too complex? Hydrol.Proc. 22, 1727-1730.Vache, K.B., McDonnell, J.J., 2006. A process-based rejectionist framework for evaluating catchment runoff model structure. Water Resour.Res. W02409, doi:10.1029/2005WR004247. Im letzten Jahrzehnt ist die Anwendung klassischer hydrologischer Modelle als Mittel zum Prozessverständnis des Wassertransports in Landschaften in die Diskussion geraten, da diese Modelle nicht gut zur Formulierung und Zurückweisung von Hypothesen geeignet sind (z.B. Beven 2002 und 2006, Seibert und McDonnell 2002, Sivapalan et al. 2003, Kirchner 2006, Tetzlaff et al. 2008). Vaché und McDonnell (2006) entwickelten ein System, mit dem Modellstrukturen falsifiziert werden können, wobei jede Modellstruktur als Hypothese der Abflussbildung verstanden wird. Die vorliegende Doktorarbeit zeigt eine generalisierte Neuentwickelung des Vaché-McDonnell Systems, das Catchment Modelling Framework (Einzugsgebiets-Modell-System, CMF). Es ist als Erweiterung der Programmiersprache Python implementiert und bietet einen Baukasten für die Entwicklung einer Vielzahl verschiedener hydrologischer Modelle, basierend auf dem finiten Volumen Ansatz von Qu und Duffy (2007).Auch Buytaert et al. (2008) und Clark et al. (2011) fordern solche modularen Systeme als Mittel für die Formulierung, Programmierung, Prüfung und Falsifizierung von Prozesshypothesen. Buytaert et al. (2008) stellen fest, dass solche Systeme portabel, verfügbar und modular sein müssen. Gleichzeitig wächst, während Hydrologen die theoretischen Anwendungsgrenzen von Abflussmodellen für Stofftransport in Landschaften diskutieren, der Bedarf an integrierten Landschaftsmodellen für die Integration von lateralem Transport in interdisziplinären Projekten, wie z.B. dem NitroEurope IP (EU). Vor diesem Hintergrund sollten modulare hydrologische Systeme so konzeptioniert sein, dass Datenaustausch während der Modellaufzeit mit Modellen aus anderen Disziplinen möglich ist.CMF ist portabel, verfügbar und modular, da es eine quelloffene Erweiterung der Sprache Python (s. Kapitel I und II) ist, und kann für die Formulierung verschiedener Hypothesen genutzt werden (s. Kapitel III). Das Prinzip der Verbindung von CMF mit Modellen aus anderen Disziplinen wird in Kapitel IV gezeigt; Kapitel V zeigt das Potential solcher eng gekoppelter Transport- und Umsatzmodelle für die Berechnung der Emission von Treibhausgasen aus Ökosystemen.Beven, K., 2006. Searching for the Holy Grail of scientific hydrology. Hydrol.Earth Syst.Sci 10, 609-618.Beven, K.J., 2002. Towards an alternative blueprint for a physically-based digitally simulated hydrologic response modelling system. Hydrol.Proc. 16, 189-206.Buytaert, W., Reusser, D., Krause, S., Renaud, J.P., 2008. Why can t we do better than Topmodel? Hydrol.Proc. 22, 4175-4179.Clark, M.P., Kavetski, D., Fenicia, F., 2011. Pursuing the method of multiple working hypotheses for hydrological modeling. Water Resour.Res 47.Kirchner, J., 2006. Getting the right answers for the right reasons: Linking measurements, analyses, and models to advance the science of hydrology. Water Resour.Res. 42_ , W03S04, doi:10.1029/2005WR004362.Qu, Y.Z., Duffy, C.J., 2007. A semidiscrete finite volume formulation for multiprocess watershed simulation. Water Resour.Res. 43, W08419, doi:10.1029/2006WR005752.Seibert, J., McDonnell, J.J., 2002. On the dialog between experimentalist and modeler in catchment hydrology: Use of soft data for multicriteria model calibration. Water Resour.Res. 38, doi: 10.1029/2001WR000978.Sivapalan, M., 2003. Process complexity at hillslope scale, process simplicity at the watershed scale: is there a connection? Hydrol.Proc. 17 _, 1037-1041.Tetzlaff, D., McDonnell, J.J., Uhlenbrook, S., McGuire, K.J., Bogaart, P.W., Naef, F., Baird, A.J., Dunn, S.M., Soulsby, C., 2008. Conceptualizing catchment processes: simply too complex? Hydrol.Proc. 22, 1727-1730.Vache, K.B., McDonnell, J.J., 2006. A process-based rejectionist framework for evaluating catchment runoff model structure. Water Resour.Res. W02409, doi:10.1029/2005WR004247.