Integration of Geophysical Data to improve Process Understanding of Hillslope Hydrology

Abstract:

Hydrological processes at the hillslope scale are of great importance when describing watershed processes, since the overall discharge of a catchment is strongly linked to the runoff generation processes at the hillslope scale (Tromp van Meerveld et al., 2008). However, hydrological processes at the hillslope are highly variable and complex (McDonnell et al., 2007). Different approaches exist to conceptualize and model these processes. Due to the complexity, it remains difficult to capture all driving factors and often therefore the predictions at the catchment scales remain poor. In the last years, several studies outlined that a new direction in hillslope hydrology is needed which improves modeling and predictions (McDonnell et al., 2007, Weiler and McDonnell, 2006). McDonnell et al. (2007) mentioned that enormous effort is spent into the description of the heterogeneity and complexity of hydrologic processes. To understand hydrological processes at the hillslope or even at the catchment scale, several studies exist concerning runoff processes which are based on multiple instrumentations covering large areas or the complete hillslope with the aim to gain a broad overview of the hydrological variety along the hillslope or catchment (Robinson et al., 2012, Sherlock and McDonnell, 2003, Tromp-van Meerveld et al., 2007). McDonnell et al. (2007) outlined, that rather the challenged question should start with 'why' something is happening instead of 'what' is happening. To answer these 'why' questions, a profound understanding about the underlying geology or subsurface structure is needed, which often plays a major role in determining hillslope hydrological processes.

Hydrologic measurements show the advantage that they can be carried out with high temporal resolution. However, the spatial resolution is limited. Commonly a spread of instruments covers a part or is widely spread along the catchment to indicate hydrological processes. These point measurements depict local processes and the task of the modeler is to bring the highly resolved point measurements in phase with each other. However, the interspatial volume between instrumented locations remains uncertain. Less is known about the structure and hydraulic characteristic of the subsurface between instrumented locations. Especially in highly heterogeneous areas this uncertainty can be a strong limiting factor. One possible way to link between investigated locations is the integration of geophysical data. Geophysical measurements capture differences in physical properties persistent in the subsurface layers like density, electrical conductivity or resistivity. These properties can be related to hydrological parameters. However, a direct quantitative relation between physical properties and hydrological parameters requires the establishment of pedo-transfer functions, which can be time consuming and costly. In general, the use of localized hydrologic point measurements as ground truth data can help to relate physical properties with hydrological parameters and allow the regionalization between point measurements using geophysical data. Commonly, geophysical methods are applied using lateral profiles resulting in 2D or 3D cross sections. These depth sections allow a qualitative interpretation of the subsurface structure which facilitates the evaluation of the subsurface structure with respect to their hydrologic data. One drawback of geophysical measurements inheres to their ambiguity in interpretation. The combination with different methods and/or ground truth data is essential if specific depth relations are needed. Often, solely a structural picture is sufficient. We have to remark at this point that geophysical methods are applicable if sufficient large contrast in the physical properties exists between the target (bedrock, subsurface layer, groundwater table) and the surrounding material. Otherwise it might be the case that structures exist but due to the lack of difference in physical properties it is not possible to resolve them. Several authors outlined the potential inherent to the combination between hydrologic and geophysical investigations to determine subsurface geologic structures and hydrologic parameters (Kleber and Terhorst, 2013, Vignoli et al., 2012, Brunet et al., 2010, Koch et al., 2009, Robinson et al., 2008). The pathways, infiltrating water chooses after drainage, often remain unrevealed. Geologic structures seldom translucence through the surface and the geologic and geomorphic structures within the first layers can tremendously differ from the surface topography. Due to their possible different hydraulic properties these structures can strongly influence the pathways of water during drainage processes (Koch et al., 2009).

With this study we present the results from two measurement campaigns during 2013 and 2014 which were carried out in a small headwater catchment about 5 km south of Freiburg, Southern Germany. The area evolved due to tectonic processes caused by the rifting the Rhine graben and is nowadays situated within the eastern main fault between Rhine Graben and Black Forest (Villinger, 1999). The bedrock consists of Gneiss, partially overprinted by granitic plutonites (Villinger, 1999). It is expected that the subsurface structure of the slopes is mainly determined by periglacial cover beds, which can have a significant influence on hydrologic processes (Kleber and Terhorst, 2013). Cover beds are divided into three layers: a basal layer, an intermediate layer and an upper layer. All layers can differ in their physical properties (Kleber and Terhorst, 2013). The existence, evolution and thickness of these layers are variable and site specific. The catchment is formed as a v-shaped valley with steep slopes (mean 27°) (Bachmair and Weiler, 2013). According to Kleber and Terhorst (2013), the basal layer is expected to develop within the valley and the flattened areas, the intermediate layer might be missing along the steep gradient and the upper layer is expected to cover the entire hillslope. For hydrologic investigations, the slope which is exposed to the northern direction was divided into three sectors for hydrologic due to its land use forms. All three sectors are similar in size and topography, but they possess different vegetation covers: grassland, coniferous forest and mixed forest. On each of the three sectors, three transects were chosen and equipped with 10 groundwater observation wells each. Additionally, trenches monitoring subsurface flow and a v-notch weir at the catchment outlet had been installed (Bachmair and Weiler, 2012). Long term hydrologic data was acquired during the last 3 years. A former study compared long well observations, hillslope topographic characteristics and discharge measurements of the catchment (Bachmair and Weiler, 2013, Bachmair et al., 2012). This led to the conclusion that soil properties and topography have a strong influence on the outflow dynamics. On the other hand, different amplitudes in the hydrographs were found during periods of similar hillslope hydrological dynamics. These findings implicated that additional factors which are not captured by the installation of the wells possess an influence on the hydrologic processes.

During this study we ask whether and how we can improve our process understanding of hydrologic dynamics at the hillslope scale when incorporating not only hydrologic methodologies but also geophysical measurements. In order to improve the understanding of the geologic structure forming the hillslope, two different geophysical methods were used: refraction seismic (RS) and electrical resistivity tomography (ERT). The choice of the geophysical method was based on the ability to apply the method on the steep slope. ERT profiles could be spread out over the entire hillslope, whereas refraction seismic was found to be less applicable along the steep slope and in the forest, when dense vegetation hindered the hammer stroke. Seismic profiles are therefore limited to the grassland site and elongated along equal contour lines. One of the main questions of our study is, to which extent additional geophysical data can help to improve hydrological process understanding of the hillslope. A competitive benefit of adding geophysical data to hydrologic investigations is the transfer of localized information towards larger scales. We compare well depths located at the grassland site to the inversion results received from electrical resistivity data. However care has to be taken, since inversion results from ERT measurements are not able to provide exact depth results. Different methods are tested to improve the reliability of ERT inversion by using either well depth or the complementary data of seismic refraction. In the following, we pose a set of questions concerning the hillslope, which we try to answer:

• Are there structural differences between two of the hillslopes, the grassland site and the mixed forest?
• Is it possible to relate the results from geoelectric profiles to the observations acquired at the wells?
• Are we able to detect the depth and structure of the bedrock?
• Are there structural differences between two sides (northern and southern slope) of the stream?
• Are we able to delineate subsurface flow patterns, with the help of geophysical data?

We ask whether and how the process understanding at this small hillslope can be improved when adding geophysical information. So far, preliminary results from two measurement campaign exist. The campaigns were both carried out in late spring and early summer months. Both methods (ERT and RS) were applied to refine underlying geological structural information. At the grassland site, geoelectric profiles were spread along observation well transects using an electrode spacing of 1m and a layout of 46 m. Three long profiles are acquired perpendicular to the slope using an electrode spacing of 2m and a layout of 118 m. Additionally two long profiles were elongated parallel to the stream, each having a spread of 142 m with an electrode spacing of 2m. Refractions seismic was carried out along these profiles using a geophone spacing of 2m and with offset shots placed up to 46 m. At the mixed forest hillslope, four geoelectric profiles were carried out, each with an electrode spacing of 2m. One profile is elongated perpendicular the hillslope spreading 94 m and crossing the other three profiles which are elongated parallel to the hill. The results at the grassland hillslope reveal a three-dimensional, highly heterogeneous structural pattern of the subsurface (Figure 1). Under the assumption that different electrical resistivity values come along with different hydraulic conductivities (Koch et al., 2009) a strongly heterogeneous subsurface flow paths can be delineated. We found that interpolating between observation wells in a common geostatistical manner is not feasible due to the rather fast changing subsurface structures. A distinct layering of the subsurface exists, thus supporting the assumption that the subsurface of the hillslope is dominated by periglacial deposits. Refraction seismic profiles are used to delineate the structure and depth to the bedrock. The outlook for further research aims towards the depth verification of the bedrock using direct push techniques and a detailed technical investigation of the subsurface layers and their geotechnical characteristics. A direct push campaign allows the in situ determination of technical parameter and the relation between seismic and geoelectric parameters. Furthermore it is suggested to investigate the opposing hillslope concerning the geomorphologic evolution of the periglacial deposits and their effect on hydrological processes.

Figure 1:Upper left side shows a sketch of the area. Contour lines are shown in black. Black dots indicate observation wells and orange lines present geoelectric profiles which were acquired during the campaign 2013. Resistivity inversion results from geoelectric profiles are shown on the right side. Lower left side presents a picture from the area.

References:

Bachmair S., Weiler M. 2012. Hillslope characteristics as controls of subsurface flow variability. Hydrology and Earth System Sciences Discussions 9: 6889 - 6934.

Bachmair S., Weiler M. 2013. Interactions and connectivity between runoff generation processes of different spatial scales. Hydrological Processes.

Bachmair S., Weiler M., Troch P.A. 2012. Intercomparing hillslope hydrological dynamics: Spatio-temporal variability and vegetation cover effects. Water Resour. Res., 48, W05537, doi:10.1029/2011wr011196.

Brunet P., Clement R., Bouvier C. 2010. Monitoring soil water content and deficit using Electrical Resistivity Tomography (ERT)- A case study in the Cevennes area, FranceJournal of Hydrology, 380, 146-153.

Kleber A., Terhorst B. 2013. Mid-latitude slope deposits (cover beds), Developments in Sedimentology 66, ISBN 978-0-444-53118-6.

Koch K., Wenniger J., Uhlenbrock S., Bonell M. 2009. Joint interpretation of hydrologigal and geophyscial data: Electrical resistivity tomography results from a process hydrological research site in the Black Forest Mountains, Germany. Hydrologigal Processes, 23, 1501-1513.

McDonnell J. J., Sivapalan M., Vache K., Dunn S. , Grant G., Haggerty R., Hinz C., Hooper R., Kirchner J., Roderick M. L., Selker J., Weiler M. 2007. Moving beyond heterogeneity and process complexity: A new vision for watershed hydrology. Water Resour. Res., 43, W07301, doi:10.1029/2006WR005467.

Robinson D. A., Binley A., Crook N., Day-Lewis F. D., Ferre T. P. A., Grauch V. J. S., Knight R., Knoll M., Lakshmi V., Miller R., Nyquist J., Pellerin L., Singha K., Slater L. 2008. Advancing process-based watershed hydrological research using near-surface geophysics: a vision for, and review of, electrical and magnetic geophysical methods. Hydrological Processes, 22, 3604-3635.

Robinson D. A., Abdu H., Lebron I., Jones S. B. 2012. Imaging of hill-slope soil moisture wetting patterns in a semi-arid oak savanna catchment using time-lapse electromagnetic induction. Journal of Hydrology, 416-417, 39-49.

Rosenbaum U., Bogena H.R., Herbst M., Huisman J. A., Peterson T. J., Weuthen A., Western A. W. Verecken H. 2012. Seasonal and event dynamics of spatial soil moisture patterns at the small catchment scale. Water Resour. Res., 48, W10544, doi:10.1029/2011WR011518, 2012

Sherlock M. D., McDonnell J. J. 2003. A new tool for hillslope hydrologists: spatially distributed groundwater level and soil water content measured using electromagnetic induction, Hydrological Processes, 17, 1965-1977.

Tromp-van Meerveld I., Peters N. E., McDonnell J.J. 2007. Effect of bedrock permeability on subsurface stormflow and the water balance of a trenched hillslope at the Panola Mountain Research Watershed,Georgia, USA. Hydrological Processes, 21, 750-769.

Tromp-van Meerveld I., Weiler M. 2008. Hillslope dynamics modeled with increasing complexity. Journal of Hydrology, 361, 24-40.

Vignoli G., Cassiani G., Rossi M., Deiana R., Boaga J., Fabbri, P. 2012. Geophysical characterization of a small pre-Alpine catchment. Journal of Applied Geophysics, 80, 32-42.

Villinger, E. 1999. Freiburg im Breisgau- Geologie und Stadtgeschichte, Landesamt für Geologie, Rohstoffe und Bergbau Baden-Württemberg, Informationen 12.

Weiler M., McDonnell J. J. 2006. Conceptualizing lateral preferential flow and flow networks and simulating the effects on gauged and ungauged hillslopes. Water Resour. Res., 43, W03403, doi:10.1029/2006WR004867.