Abstract

In this paper a short introduction is given to the Danish hydrological observatory—HOBE. We describe characteristics of the catchment, which is subject to experimental and modeling investigations. An overview is given of the research reported in this special section of the journal, which includes 11 papers of original research covering precipitation, evapotranspiration, emission of greenhouse gasses, unsaturated flow, groundwater–surface water interaction, and climate change impacts on hydrology.

The guest editors introduce the contributions to this special section devoted to work at the Danish hydrological observatory—HOBE. The research projects at this site are part of a global effort to refine our understanding of hydrological processes in the face of land-use impacts and climate change.

There is a growing need to improve our scientific understanding of hydrological processes at catchment scale to assess the hydrological consequences of natural and anthropogenic changes. The consequences of such changes are typically manifested at large time and spatial scales, with implications for the sustainability of water as a resource.

The catchment scale has also been adopted as the most appropriate scale for water legislation and management. The Water Framework Directive of the European Union implemented in 2000 prescribes that water management strategies must be developed at the catchment scale—the natural geographical and hydrological unit—instead of according to administrative or political boundaries (Refsgaard et al., 2005). The catchment or basin scale is also used in integrated water resources management, which is widely used as the management principle in developing countries (Jonch-Clausen and Fugl, 2001).

Several investigations have documented that our knowledge of the in- and outgoing water fluxes and the exchange of water between the different hydrological compartments in a catchment is insufficient due to both measurement and theoretical limitations. As a result, difficulties with closure of the water budget at the catchment scale have been reported (Henriksen and Sonnenborg, 2003; Henriksen et al., 2003), and this is perhaps one of the most fundamental issues in hydrological sciences.

Addressing the hydrological consequences of, for example, climate or land-use change or undertaking proper water management according to new management principles must be underpinned by an accurate scientific understanding of the hydrological processes occurring at the catchment scale. In the past, most of the hydrological process understanding was obtained at small scales (plot, field, and small research catchments), and the issue of scale has not been considered in the design of experiments and in theoretical analyses. Catchments are subject to a time–space variability of landscape characteristics, and the hydrological processes occur and interact across a multitude of spatial and temporal scales. Therefore, the process understanding and associated parameters obtained at one scale cannot directly be applied at another scale. In consequence, there is a disparity between the scale at which the current process understanding is established and the scale at which future hydrological consequences need to be addressed and water management strategies need to be developed.

In response to the need of society to cope with the effects of climate and land-use changes and to perform integrated water management at the catchment scale, the international hydrological science community has proposed that large-scale hydrological observatories be established for long-term hydrological measurements of in- and outgoing fluxes, as well as fluxes between the different hydrological compartments. Such observatories are necessary instruments for obtaining a better understanding of the catchment behavior and for improving the reliability of predictions of how catchments respond to external stresses. The U.S. Consortium of Universities for the Advancement of Hydrologic Science (CUASHI; http://www.cuahsi.org [verified 26 Jan. 2011]), has been a promoter for such observatories and the U.S. National Science Foundation has provided funding for the establishment of six critical zone observatories (http://www.criticalzone.org/ [verified 26 Jan. 2011]). In Germany the TERENO project has been initiated, which involves the establishment of four observatories across Germany (Bogena et al., 2006). In Denmark a hydrological observatory, HOBE, was established in 2007 based on funding by the private fund the Villum Foundation (http://www.hobecenter.dk [verified 26 Jan. 2011]).

Fundamental to such observatories is that fluxes, state variables, and parameters are measured at multiple temporal and spatial scales. Recent developments in ground-based, air-borne, and satellite-borne sensor technologies have provided unprecedented opportunities to meet these requirements. Integrated high quality data collected at a hierarchy of nested scales, ranging from detailed studies and observations at plot scale to spatially distributed observations at catchment scale, provides an avenue for improving our understanding of the processes and their interaction at different scales. On this basis, new concepts and scaling theories can be developed and tested and thereby can help reduce the gap between the scale at which we understand the processes and the scale at which water management is requested.

Hydrological Observatory—

Skjern Catchment (HOBE)

The Skjern catchment located in the western part of Denmark was selected as the site for the hydrological observatory (Fig. 1 ). The river emerges in the eastern part of the catchment and flows into a brackish inland fjord adjacent to the North Sea. The area of the entire catchment is 2500 km2.

The climate in western Denmark is typical of the maritime regime, dominated by westerly winds and frequent passages of extratropical cyclones. Maximum precipitation is in autumn and minimum precipitation in spring. The dominant westerly wind results in mild winters and relatively cold summers with highly variable weather conditions characterized by frequent rain and showers. The mean annual precipitation is 990 mm, the mean annual reference evapotranspiration is 575 mm, and the mean annual temperature is 8.2°C.

The topography slopes gently from east to west, with land surface elevations from 125 m above sea level in the eastern part to sea level at the coast. The landscape is traversed by the Skjern river and its tributaries (Fig. 2 ). The river system runs in glacial outwash plains consisting of sand and gravel of Quaternary age with isolated islands of Saalian sandy till between the plains (Fig. 3 ). Alternating layers of marine, lacustrine, and fluvial deposits of Miocene age underlie the Quaternary deposits and thick clay layers from Paleogene underlie the Miocene deposits (Fig. 4 ).

The mean annual river discharge is 475 mm. Due to the highly permeable soils, all water outside the wetlands infiltrates, and the discharge to the streams is therefore dominated by groundwater flow (Van Roosmalen et. al., 2007).

Groundwater abstraction varies over the season and from year to year due to variable demands for irrigation. Abstraction for domestic and industrial purposes amounts to approximately 10 mm yr−1 while the average groundwater abstraction for irrigation is approximately 20 mm yr−1 (Henriksen and Sonnenborg, 2003). In dry summers the irrigation demand can be up to 50 mm yr−1 at the catchment scale. Approximately 50% of the catchment area is subject to irrigation.

The catchment is mainly rural, with the following land-use distribution: grain and corn (55%), grass (30%), forest (7%), heath (5%), urban (2%), and other (1%) (Fig. 5 ). During the past 50 yr, the catchment has been subject to significant anthropogenic interferences, including deforestation and engineering of the stream courses.

The overall objectives of the research are: (i) to carry out integrated and interdisciplinary measurements and experiments at multiple spatial and temporal scales; (ii) to establish a high-density, multiscale, high-quality, and long-time data set that can provide a platform for hydrological research with interdisciplinary focus; (iii) to improve the scientific basis for better water resources management decisions and for reducing the uncertainty in water balance closure at catchment scale; and (iv) to strengthen the graduate education and training program within hydrology.

The goals are achieved by integrating monitoring, measuring, modeling, and regionalization activities within the research catchment. Classic state-of-the-art measurement techniques in combination with novel sensor technologies are used to measure and analyze the multiscale spatial and temporal patterns of the land surface and subsurface systems, including system parameters, state variables, in- and outgoing water fluxes, and water fluxes between hydrological compartments. Fluxes of both water vapor and greenhouse gasses are measured above different vegetation types to assess the vegetative response to prevailing climatic conditions and the effects of land-use or climate changes on these fluxes.

The project takes advantage of the recent developments within ground-based, air-borne, and space-borne noninvasive geophysical, meteorological, and remote sensing sensors. The collected data form the basis for development of integrated and physically based models for different scales. The theoretical research will address the issue of scale and the pertinent scientific problem of how to transfer the knowledge obtained at small scale to larger scale given the spatial and temporal heterogeneity observed in nature and the associated highly nonlinear flow and flow exchange processes. Related to this, methods for deriving effective large-scale parameterizations will be developed that consider the effect of subscale heterogeneity. Validated integrated hydrological models will be applied for predicting the effect of climate change and land-use changes.

The measurement and monitoring activities are organized in a nested spatial scale approach encompassing the following investigation scales (Fig. 6 ): (i) total catchment area (∼2500 km2), (ii) large subcatchment defined by gauging station Alergaarde (∼1050 km2), (iii) small subcatchment defined by gauging station Holtum (∼80 km2), (iv) flight transects over the central part of the catchment (∼100 km2), (v) flight transects over Ringkøbing Fjord (catchment outlet) (∼10 km2), (6) three field sites representing the three land surfaces agriculture, forest and meadow (∼1 ha), and field sites equipped with instruments for measuring the exchange of water between groundwater and stream as well as groundwater and sea (several hectares).

Additionally, satellite data representing different spatial and temporal scales are collected from various sensors onboard various satellites (ENVISAT, Landsat, SPOT, MODIS, SMOS) for catchment-wide mapping of parameters and variables.

Research Results

This special issue of Vadose Zone Journal includes 11 papers that report research results obtained in the HOBE project over the first 3 yr since its inception in 2007.

The papers cover investigations of various hydrological processes at different spatial scales, including regional estimation of precipitation and the impact on the hydrological response at the catchment scale, measurement of evapotranspiration and greenhouse gases, unsaturated flow, groundwater–surface interaction, mapping of distribution of freshwater and sea water at brackish lagoon, and climate change impacts on hydrology.

Precipitation

Three papers deal with precipitation. He et al. (2011) present a method for generation of daily quantitative precipitation estimation (QPE) based on C-band Doppler radars. The method is a distance-dependent areal estimation method that combines radar data and ground observations. It is shown that the radar-based products have much more details on the spatial distribution of precipitation than when interpolating rain gauge data. The radar based precipitation product was used as input to a physically based, distributed hydrological model for the Skjern catchment and the hydrological model simulations were compared with the results obtained using interpolated rain gauge data as input. The study showed that radar QPE products are able to generate reliable simulations of stream discharge and that the significance of the spatial distribution of precipitation becomes particular important for subcatchments less than 400 km2. The study underlined the potential of radar QPE for improving the input of precipitation to distributed hydrological models and hereby improving the predictions of the hydrological responses both in short- and long-term perspectives.

Fu et al. (2011) analyzed the impact of the spatial resolution of precipitation and the method for bias correction of precipitation measurements on runoff, recharge, and groundwater head. Six different precipitation schemes were used as input to the same hydrological model as above. The results showed that stream discharge, groundwater recharge, and groundwater heads were affected by both the resolution of the precipitation input and the method used for correcting the under-catch by rainfall gauges. For the Skjern catchment the impact of resolution is reduced with increasing catchment size, and the effect on stream discharge is relatively low for subcatchment sizes above 250 km2 due to averaging effects.

Stisen et al. (2011) also examined the impact of rain gauge catch corrections as well as potential evapotranspiration input on stream discharge and groundwater levels using the same modeling framework as in the other two contributions. In contrast to the contribution by Fu et al. (2011) this study also included the use of dynamic corrections factor for precipitation. By this method the catch correction is estimated dynamically on a daily basis using data on wind speed, precipitation intensity, and temperature. In contrast to the more commonly used global correction factors, dynamic correction ensures that short-term variations and especially interannual variations are considered. The results on model performance using different catch correction methods showed that much better dynamics in the simulation of discharge can be achieved and that the current water balance problems can be minimized when using dynamic correction factors.

Evapotranspiration and Emission of Greenhouse Gases

The special section includes three contributions dealing with evapotranspiration and greenhouse gases. Ringgaard et al. (2011) present eddy-covariance measurements above three land surfaces located within the Skjern catchment representing agricultural land, spruce plantation, and wet grassland, respectively. The results demonstrate the existence of differences in seasonality in the radiation budget and the sensible heat flux for the different land surfaces. During the winter season evapotranspiration (ET) of grassland and forest is higher than incoming radiation due to incoming sensible heat caused by large-scale flow of relatively warm maritime air from the North Sea over the relatively cold land surface. The ET from the agricultural surface is very much controlled by crop development. Stomatal control is the limiting factor for transpiration in the forest, and the measurements documented that interception evaporation is a main contributor to total evapotranspiration.

At the same three sites Herbst et al. (2011) performed eddy covariance measurements of the exchange of greenhouse gases between the land surface and the atmosphere. Fluxes of CO2 were measured at all three sites while CH4 and N2O were measured above the wet grassland and the agricultural field, respectively. The measurements provided new insights into the behavior of different land-use types as sources and sinks for greenhouse gases. The agricultural site acted as a CO2 sink from April to June and a CO2 source during the rest of the year. On an annual basis, the forest plantation fixed twice as much CO2 as the agricultural site, and it remained a CO2 sink throughout the whole year. The annual CO2 fixation at the wet grassland site had an intermediate level, with the surface being a CO2 sink from March to October. However, the emissions of CH4 from the wet grassland and N2O from the agricultural land reduced the greenhouse gas sink for the respective land surfaces.

Schelde et al. (2011) compared water vapor fluxes during dry days using eddy covariance and root zone soil moisture depletion using time domain reflectometry (TDR) as two independent techniques for estimating ET. During some periods agreement between the two approaches was good, while during others the two estimates deviated due to different source areas contributing to the flux estimates. With certain limitations the TDR estimate constitutes a lower limit of ET during a 24-h cycle and consequently TDR estimates can be used for quality control and for confining eddy covariance measurements.

Unsaturated Flow

The paper by Haarder et al. (2011) describes a field experiment in which high-resolution ground penetrating radar (GPR) is used for nondestructive visualization of complex unsaturated flow patterns in a shallow subsurface arising from a forced infiltration experiment of dyed water. Dye-staining patterns, revealed by excavating a 2-m-deep trench through the infiltration area, were compared with changes in the GPR data. It was not possible to resolve the observed pattern within the dye-affected area by the GPR signals. However, the GPR data provided information about the unsaturated flow below both the dye-staining and the excavation. The study documented that GPR is useful as a nondestructive means of assessing unsaturated flow phenomena.

Groundwater–Surface Water Interaction

Two papers deal with groundwater–surface water interaction. Jensen and Engesgaard (2011) used temperature measurements to quantify groundwater seepage across a stream bed. The data documented a high degree of nonuniformity in the flux both in time and space. Measurements of heads in the meadow and below the stream and seepage meter measurements further substantiated the complex flow pattern and suggested convergence of flow near the streambed and hyporheic flow due to the presence of gravel layers.

In a related study Kidmose et al. (2011) investigated groundwater seepage and recharge for a lake in the Skjern catchment using a suite of measurements including seepage meter, hydraulic head, isotope fractionation δ18O, CFC ages, geophysical surveys, and sediment coring. The field data were used to constrain a groundwater model based on the MODFLOW code. As for the stream the results showed a complex and heterogeneous in- and outflow pattern, which is related to heterogeneity in the geological settings and in the hydraulic properties of the lake bed.

Hydrogeophysical Survey

Kirkegaard et al. (2011) studied the distribution of freshwater and sea water at the outlet of Skjern stream to the Ringkøbing lagoon. A geophysical campaign using an airborne SkyTEM instrument was used to map the freshwater body in the coastal lagoon. Existing hydrological data indicate possible outflow from the catchment underneath the survey area, a hypothesis that was supported by the geophysical results. The results also revealed the presence of a stratified geological setting incised by buried valleys beneath the lagoon. The data suggested that the valleys are saturated with seawater unlike the layers at same depth. The salinity distribution in the settings indicated the presence of a complex flow system beneath the lagoon.

Climate Change Impact on Hydrology

Van Roosmalen et al. (2011) present a study on the impact of climate change on the hydrology of the western part of Jutland including Skjern catchment. The study focuses on the impact of the choice of bias correction method on the projected hydrological change. Two methods were tested: (i) perturbation of observed data using climate change signals derived from a climate model and (ii) distribution-based scaling of the climate model output. The results suggested that for the particular catchment in question the choice of bias-correction method does not have a major influence on the projected changes of mean hydrological responses.

Outlook

This special section presents the results obtained during the first 3 yr of the hydrological observatory HOBE. We anticipate that the observatory will run for 10 yr or more and thus longer time series will become available. Collection of high-quality data representing different spatial and temporal scales and using classical state-of-the-art and new measurement techniques is a main emphasis of the observatory. The collected data provide the basis for cross-cutting research within spatial estimation of precipitation, vegetative control of evapotranspiration and emission of greenhouse gases, spatial estimation of soil moisture and recharge, and fluxes between groundwater and surface water bodies. Data and modeling go hand-in-hand to improve the understanding of processes and responses. The development and validation of integrated and distributed models for the entire catchment is the integrator of all research results.

In the coming years emphasis will be on measurements of stable isotopes to help improve the quantification of exchanges between hydrological compartments, on integrated hydrological modeling, and on the issue of scale. The latter is one of the most pertinent and challenging research questions in hydrology. The quantification of the hydrological processes at the catchment scale is fundamental for addressing present and future water resources problems affected by both anthropogenic impacts and climate change. The scale question needs to be addressed in a collaborative effort, and hopefully the data collected in the HOBE project will contribute to this endeavor.

The Villum Foundation has funded the hydrological observatory HOBE and the research reported in this special issue. The HOBE research group is grateful for the opportunities that this donation has provided.

Freely available online through the author-supported open access option.