Wolfgang Gossel, Peter Wycisk: Importance and effects of model - PowerPoint PPT Presentation

Wolfgang Gossel, Peter Wycisk: Importance and effects of model couplings in hydrogeology

Objectives Implementation Reasons Kinds Behaviour Calibration Two examples for solution of hydrogeological problems • How were the groundwater flow and transport conditions influenced by open pit lignite mining in the region Untere Mulde/Fuhne? • What was the impact of climate change during the last 140 000 years on the groundwater balances in the Nubian Aquifer System? ? 3/26

Objectives Implementation Reasons Kinds Behaviour Calibration Problem and model Problem solution with Pumpversuch_02102002: Darstellung der Messwerte mit einer logarithmischen Geraden zur Auswertung der Absenkung diverse methods: 0.000 B6A B6E B6C 0.010 logarithmische Gerade, B6C logarithmische Gerade, B6E logarithmische Gerade, B6A Absenkung s [m] 0.020  0.030 Analytical 2 . 3 Q t   2 0.040 T lg     methods 4 s t 0.050 1 0.060 Statistical 1 10 100 1000 10000 Zeit t [s] methods Numerical methods

Objectives Implementation Reasons Kinds Behaviour Calibration Modelling systems, -tools and models Changed after Bossel (1992) „data“

Objectives Implementation Reasons Kinds Behaviour Calibration Modelling systems, -tools and models Modelling systems Modelling tools Computer- implementation Geology (hardrocks, GSI3D unconsolidated rocks): static, constructive Groundwater recharge: GIS, Excel dynamic, GIS-based Surface water: Excel, own statistic, time series programming Groundwater: transient, Feflow numeric

Objectives Implementation Reasons Kinds Behaviour Calibration Suitability of modelling tools Objective modelling tool • Dynamic, variability in time domain, long time frames • 3D spatially variable resolution according to data availabilty and objectives

Objectives Implementation Reasons Kinds Behaviour Calibration Complexity of models and necessity of couplings Dependent on: • Number of compartments • Number of used modelling systems • Nonlinearity of the functions in modelling systems and the connections of the modelling systems • Number of parameters • Heterogeinity of parameters and boundary conditions in time and space Nubian Aquifer System Untere Mulde/Fuhne Surface Unsat. water Ground- Climatic Groundwater zone water changes flow and recharge transport, Ground- density driven water Geology flow and transport

Objectives Implementation Reasons Kinds Behaviour Calibration Pros and Cons of complex models Pros: • Substitution of weakly defined parameters (or b.c.) by better systematical approach • Range of objectives increases Cons: • High affinity to instabilities of the coupled modelling systems • Sensitivity difficult to measure • Scale dependency of parameters in different modelling systems • Technical implementation of interfaces -> solvable

Objectives Implementation Reasons Kinds Behaviour Calibration Kinds of coupling ModSys 1 ModSys 1 Horizontal coupling e.g. regional e.g. local model model Serial coupling ModSys 1 ModSys 2 (non iterative) Periodically ModSys 1 ModSys 2 synchronized e.g. flow e.g. transport modelling modelling coupling Parallel coupling ModSys 1 ModSys 2 (iterative) Integrated coupling ModSys 1 ModSys 2

Objectives Implementation Reasons Kinds Behaviour Calibration Coupling construktive geolog. mod. with hydrogeol. mod.

Objectives Implementation Reasons Kinds Behaviour Calibration Sequential and serial coupling, periodically synchronized Soil (substrate) Land use Jährliche Niederschlagssummen Bitterfeld Niederschlagssumme [mm] 1000 900 800 700 600 500 400 300 200 100 0 Stabw 1947 1949 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 Mittel Min Max Jahre Jährliche Summe der potentiellen Evapotranspiration (ETP) Bitterfeld 1000 Jahressumme ETP [mm] 900 800 700 600 500 400 300 200 100 0 1947 1949 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 Mittel Max Stabw Min Jahre Jährliche Klimatische Wasserbilanz (KWB) Bitterfeld 500 Jahressumme KWB [mm] 400 300 200 100 0 -100 -200 -300 -400 -500 1947 1949 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 Mittel Min Max Stabw Jahre Climatic data 145 hydrotopes Depth to groundwater

Objectives Implementation Reasons Kinds Behaviour Calibration Iterative, integrated coupling Nubian Aquifer System: Completely density driven groundwater modelling Density Transport Flow parameters parameters parameters Transient density Transient Transient coupled groundwater groundwater groundwater flow flow model transport model and transport model B.C. for flow model B.C. for transport model

Objectives Implementation Reasons Kinds Behaviour Calibration Interfaces of modelling systems Technical problem: Coupling of modelling systems for surface water (SW) and groundwater (GW) • Systematical approach: Total discharge of SW doesn‘t influence GW level but SW level • Level and discharge of SW are dominated by catchment abroad of investigation area • Realisation of the Transfer-/Leakage-coefficient in groundwater modelling system • Level-discharge-dependency is implemented in SW modelling system

Objectives Implementation Reasons Kinds Behaviour Calibration Stability • Convergence problems for numerical solutions • Over-/Undershooting at interpolations or in dynamical models between time steps • Oscillations in feedback between modelling systems • Sometimes step back to „effective“ parameters necessary

Objectives Implementation Reasons Kinds Behaviour Calibration Complex example

Objectives Implementation Reasons Kinds Behaviour Calibration Calibration of coupled models Impact of kind of coupling on calibration of models: Kind of coupling Work load Stability Transparency Sequential middle high high Periodically synchronized high high high Iterative low low low Integrated middle low low Dilemma of the geological model: • Can not be calibrated on itself but only proven (geometrical criteria). • Importance: „Database“ for unsaturated and saturated zone modelling • Solution: Indecies for the derived model and calibration of ranges of values

Objectives Implementation Reasons Kinds Behaviour Calibration Results of coupled models Transport model Untere Mulde/Fuhne 1895-2005

The future .... Trends of development of model couplings • Horizontal coupling needless or integrated • Vertical couplings applicable via integrated solutions, unstable • Defined programming interfaces of the tools are good solution, up to now not user friendly • Data base will dominate the applicability of model couplings • OpenSource Additional tools, e.g. • Open standards Modpath, MT3D, RT3D, HUF, UZF, Pre- and Postprocessors • Documented interfaces Integration of transport, density, • Closed Source heat, unsat. zone, discrete • Open and proprietary stand. features, development of self- made modules • Documented interfaces

Summary • Application of complex models increases • Complex models depend often on couplings of modelling systems • Different kinds of coupling have consequences for behaviour, calibration, results and prognoses of models • Systematical application after detailed analyses necessary • Trend in hydrogeology towards opening of central parts of modelling tools via defined interfaces and standards

Thanks • Thanks for your attention • Thanks to the department for the support • Projects for the examples: SAFIRA I and II: Prof. Dr. Peter Wycisk, Tobias Hubert, (Dr.) Tilmann Steinmetz, Andreas Wollmann, Karsten Sommerwerk, Steffi Schnabel, Gerd Fleck, Oliver Neef, Raik Richter und Holger Fabritius Ad-Hoc- Project „Mulde -Hochwasser 2002 and Austragsdynamik“ : Christian Neumann Nubian Aquifer System: Dr. Ahmed Sefelnasr,