Schwerpunktleiter: Jan Fleckenstein
Research InterestsMy research aims at the characterization, simulation and measurement of hydrologic processes in linked surface-subsurface systems. A better understanding of the spatial patterns and temporal dynamics of exchange fluxes between surface waters and groundwater (e.g. river-groundwater, lake-groundwater or wetland-groundwater) on scales ranging from the reach or plot scale to catchment scales, is of great importance for environmetal management and questions of global change. Currently research in my group is focused on three main areas, surface-water groundwater interactions and hyporheic zone processes, heterogeneity and scale, and heat as a natural tracer. In our work we use numerical models of complex surface-subsurface systems, geostatistical techniques and simulation as well as field experiments and measurements.
Surface Water Groundwater InteractionsUnderstanding the interaction between surface water and groundwater is essential for water managers and hydrologists. In most environments the two components are in continuous interaction and development of or changes to one component inherently affects the other. Important nutrient exchange and transformation processes take place in the hyporheic zone, at the interface between surface water and groundwater. Exchange can be highly variable over a range of spatial scales and highly dynamic in time (see figure 1). Understanding those patterns and dynamics can be crucial for efforts to restore ecological functions in aquatic systems. Surface water groundwater interactions are important in water supply and management (e.g. river recharge to aquifers, bank filtration), health of aquatic systems (e.g. nutrient cycling, minimum flows) and water quality (e.g. mixing of waters of different quality, well capture of contaminated surface water).
The emphasis of our research is on the characterization and quantification of spatial and temporal patterns. We combine field work and empirical work with numerical modeling. In addition to standard field techniques (e.g. transducers, TDR, seepage meters, coring, etc.) we have started to use heat as a natural tracer (see below). To characterize geologic heterogeneity and to assess associated uncertainties we frequently use transition probability and Markov chain based geostatistical simulations (see below). For our numerical modeling we use codes for saturatd and variably saturated flow and transport (e.g. MODFLOW, MODHMS, HydroGeoSphere, VS2DH, HYDRUS, PARFLOW) as well as conceptual rainfall-runoff models (e.g. TOPMODEL, HBV).
Projects: Effects of grid-scale and subgrid-scale geologic heterogeneity on river-aquifer exchange; Spatial and temporal patterns of exchange between a peat bog and streams and their impact on nutrient export; Structurally optimized modeling of non-linear effects of meteorological extreme events on water and solute dynamics (DFG FL613/6-2).
Heterogeneity and Scale A common problem in hydrology is an apparent discrepancy between the scale of most field measurements (point) and the scale of a simulation or the scale of the problem (river reach, catchment) respectively. Heterogeneities in hydraulic properties or in initial conditions caused by spatial variability of subsurface materials, topography, vegetation etc. exist at different, nested scales and affect flow and transport processes. Hence it is important to evaluate and quantify the effects of heterogeneities on model parameterizations and systems response over a range of scales.
To date our work has been mainly concerned with subsurface heterogeneities and how they affect surface water groundwater interactions. We quantified the effects of heterogeneities on river-aquifer exchange on the scale of hydrofacies (subsurface materials deposited under specific depositional conditions, which are assumed to have distinct hydraulic properties). The inherent inaccessibility of the subsurface makes a complete deterministic description of these hydrofacies impossible. Information is limited to point data. Geostatistical methods allow a stochastic description of the subsurface based on the spatial correlation of the point data. Using transition probabilities and Markov chain models in contrast to traditional variogram or covariance models, measures like mean facies length and global facies proportions can be incorporated into the model. This way a more realistic description of the subsurface is possible. Multiple realizations of hydrofacies distributions were created conditioned with available core data (see figure 2). Often the scale of those geostatistical models is too fine for an efficient simulation of flow. Therefore we developed simple up-scaling techniques to up-scale parameter values from the detailed geostatistical model to a scale amenable to numerical flow models. We have looked at effects of heterogeneities at the sub-hydrofacies scale on river aquifer exchange using nested geostatisical simulations and Monte Carlo analyses. Flow in this work is simulated with the parallel, variably-saturated, surface-subsurface code PARLFOW, running on a Linux cluster.
Projects: Effects of grid-scale and subgrid-scale geologic heterogeneity on river-aquifer exchange.
Figure 2: Different geostatistical realizations of the subsurface
Heat as a Natural Tracer
Temperature distributions at the interface between surface water and groundwater are indicative of heat transfer between the two systems. Measurements of temperature can be used to qualitatively describe patterns of exchange and to quantify exchange fluxes by inversing. With current sensors temperatures can be easily measured with high accuracy. Compared to common sensors for hydraulic parameters such as pressure or moisture content (e.g. transducers, TDR) temperature sensors are also relatively inexpensive. Whereas in recent years temperature measurements have increasingly been used to characterize and quantify interactions between rivers and groundwater, applications in groundwater-lake and groundwater-wetland systems are still rare.
We are testing the use of heat as a natural tracer in a groundwater-lake system. Problems are the missing diurnal temperature signal in lakes and the small magnitude of exchange fluxes. Preliminary simulations, however, suggest that vertical temperature profiles beneath lakes are sensitive to small changes in flux within the accuracy of common temperature sensors (figure 3). Exchange fluxes inverted from measured vertical temperature profiles can be checked against seepage meter measurements and estimates based on measured hydraulic gradients. We have used fiber-optics based distributed temperature sensing (DTS) to map temperature anomalies at the groundwater-lake interface. In a small peat-bog system drained by small streams we use temperature data to characterize the temporal dynamics of exchange. Onset and cessation of flow in small drainage canals as a response to precipitation events as well as the magnitude and depth of exchange can be inferred from the temperature data. Results will be used to recalibrate an existing numerical model of the system, which will then be used to quantify export of nutrients (N, P) from the system.
Projects: Quantifying the spatial and temporal patterns of exchange between a peat bog and streams using heat as a tracer; Quantifying the spatial and temporal variability of lake-groundwater exchange (DFG FL613/5-1); Using fiber optics for distributed temperature Sensing in a groundwater-lake system (BaCaTeC).