Valentina Krysanova

Development and test of a spatially distributed hydrological/water quality model for mesoscale watersheds

Abstract The new watershed model SWIM was developed in order to provide a comprehensive GIS-based tool for hydrological and water quality modelling in mesoscale watersheds (from 100 to 10 000 km 2 ), which can be parametrized using regionally available information. SWIM is based on two previously developed tools—SWAT and MATSALU. The model integrates hydrology, vegetation, erosion and nitrogen dynamics at the watershed scale. SWIM has three-level disaggregation scheme and is coupled to the Geographic Information System GRASS. A robust approach is suggested for the nitrogen modelling in mesoscale watersheds. Model test and validation were performed sequentially for hydrology, crop growth, nitrogen and erosion in a number of mesoscale watersheds in the German part of the Elbe drainage basin. Firstly, the hydrological module was tested in five watersheds of different size with different topography, soil and spatial resolution of input data. After that the test was performed for the crop module in the state of Brandenburg, for the nitrogen module in the rural basin of Stepenitz, and for the erosion module in the mountainous Mulde basin. A comprehensive scheme of spatial disaggregation into subbasins and hydrotopes combined with reasonable restriction on the subbasin size allows the assessment of water resources and water quality to be performed with modest data requirements at the regional scale. The direct connection to land use and climate data provides a possibility to use the model for analysis of climate change and land use change impacts on hydrology and water quality.

Spatial Variation and Trends in PDSI and SPI Indices and Their Relation to Streamflow in 10 Large Regions of China

Time series of the average annual Palmer drought severity index (PDSI) and standardized precipitation index (SPI) were calculated for 483 meteorological stations in China using monthly data from 1961 to 2005. The time series were analyzed for 10 large regions covering the territory of China and represented by seven river basins and three areas in the southeast, southwest, and northwest. Results show that the frequencies of both dry and wet years for the whole period are lower for southern basins than for the northern ones when estimated by PDSI but very similar for all basins when calculated by SPI. The frequencies of dry and wet years calculated for 5- and 15-yr subperiods by both indices show the upward dry trends for three northeastern basins, Songhuajiang, Liaohe, and Haihe; a downward dry trend for the northwest region; a downward wet trend for the Yellow River basin; and an upward wet trend for the northwest region. Trend detection using PDSI indicates statistically significant negative trends for many stations in the northeastern basins (Songhuajiang, Liaohe, Haihe, and Yellow) and in the middle part of the Yangtze, whereas statistically significant positive trends were found in the mountainous part of the northwest region and for some stations in the upper and lower Yangtze. A moderately high and statistically significant correlation between the percentage of runoff anomaly (PRA) and the annual average PDSI and SPI was found for six large rivers. The results confirm that PDSI and SPI indices can be used to describe the tendency of dryness and wetness severity and for comparison in climate impact assessment.

Simulation modelling of the coastal waters pollution from agricultural watershed

Abstract The purpose of this research was to assess nonpoint nutrient pollution on an agricultural watershed and its influence on the eutrophication process in a sea-bay ecosystem. The method of simulation modelling was used for reliable determination of nitrogen- and phosphorus-loss dynamics in time and space on the watershed and evaluation of the influence of excess nutrient flow on the sea-bay ecosystem. Discrete simulation was chosen as the level of approach for basin submodels to provide point-scale simulation of 501 areas of pollution. The model of the bay ecosystem was realized as four systems of ordinary differential equations. An analysis of scenarios allows us to range different measures leading to more effective management practices for the basin.

Advances in ecohydrological modelling with SWAT—a review

Ecohydrology is an integrative science studying the relationships between hydrological, biogeochemical and ecological processes in soils, rivers and lakes, and at the catchment scale. It proposes a “dual regulation” of a system by simultaneously studying ecological and hydrological processes to enhance the overall integrity of aquatic ecosystems in the face of human-driven alterations and Global Change (see definition in the UNESCO Ecohydrology Programme at http://typo38.unesco.org/en/ecohydrology.html). Ecohydrology deals with hydrological factors which determine the dynamics of natural and human-driven terrestrial ecosystems, and which, together with ecological factors, influence water dynamics and water quality. River basins have a hierarchical structure and natural boundaries, and can be considered as inherent integrators of the effects of many climatic and non-climatic factors. That is why river basins represent a suitable scale for integrated ecohydrological studies and modelling. An ecohydrological river basin model includes a hydrological submodel as a basic component. Other components describing biogeochemical cycles (carbon, nitrogen and phosphorus) and vegetation are coupled with the hydrological component, in order to include important interactions and feedbacks between the processes, such as water and nutrient drivers for plant growth, water transpiration by plants, and nutrient transport with water. Generally, vertical and lateral fluxes of water and nutrients in catchments are modelled separately, whereas climate and land-use related parameters are treated as external drivers. The spatial and temporal resolution of a model depends on data availability and the aim of the study. The scale of application, spatial resolution and objective of the study are connected: a fine spatial resolution may be needed for a small catchment in order to study water flow components and their pathways using tracers; a lumped model may be sufficient for the case where “precipitation–runoff” relationships are investigated in a homogeneous medium-scale catchment; whereas a coarser resolution could be applied for a mesoscale or large river basin for climate impact assessment. There are many different classifications of river basin models. The differentiating principle could be the modelling approach or the scale of model application. For example, one can distinguish physically-based, conceptual, or black-box models; lumped and distributed models; and deterministic and stochastic models. A physically-based hydrological or ecohydrological model describes the natural system using mainly basic mathematical representations of physical laws governing the transfer of mass, momentum and energy. As a rule, such a model has to be fully distributed by accounting for spatial variations in all variables and parameters. However, the inclusion of physical laws in a model does not by itself guarantee its high quality. Even if physical laws included in the model represent a rigorous mathematical description for a soil column under laboratory conditions where soil has been well mixed, this may not automatically be the case at the scale of the grid elements used in distributed hydrological models: hundreds of metres or even kilometres (Beven, 1996). Besides, the so-called physically-based models often include empirical and statistical equations, especially to represent non-hydrological processes. In contrast, the description of important physical processes is lacking in the simplified conceptual hydrological models (e.g. water movement through soil layers) and, therefore, it is

Implications of complexity and uncertainty for integrated modelling and impact assessment in river basins

The paper focuses on implications of complexity and uncertainty in climate change impact assessment at the river basin and regional scales. The study was performed using the process-based ecohydrological spatially semi distributed model SWIM (Soil and Water Integrated Model). The model integrates hydrological processes, vegetation/crop growth, erosion and nutrient dynamics in river basins. It was developed from the SWAT and MATSALU models for climate and land use change impact assessment. The study area is the German part of the Elbe River basin (about 100,000km^2). It is representative for semi-humid landscapes in Europe, where water availability during the summer season is the limiting factor for plant growth and crop yield. The validation method followed the multi-scale, multi-site and multi-criteria approach and enabled to reproduce (a) water discharge and nutrient load at the river outlet along with (b) local ecohydrological processes like water table dynamics in subbasins, nutrient fluxes and vegetation growth dynamics at multiple scales and sites. The uncertainty of climate impacts was evaluated using comprehensive Monte Carlo simulation experiments.

Regionalisation of the base flow index from dynamically simulated flow components — a case study in the Elbe River Basin

Abstract This study investigates possibilities for the regionalisation of flow components within large river basins. One main purpose is the derivation of an empirical relationship for the estimation of the average base flow index (i.e. the ratio of base flow to total flow, BFI), which can be included into simplified models in form of decision support systems aimed at rapid integrated water resources assessment. Based on results from the hydrological models ARC/EGMO and HBV in 25 mesoscale sub basins of the macroscale Elbe River Basin, the modelled BFI is regionalised for the whole German part of the Elbe Basin, divided into 114 subbasins. For regionalisation of the average BFI multiple regression (MREG) and two geostatistical approaches (ordinary kriging (OK) and external drift kriging (EDK)) are applied using different physical catchment attributes and observed climate data as independent variables. The average base flow index is strongly related to topographical, pedological, hydrogeological and precipitation characteristics and less influenced by land cover/land use properties of the catchments. As indicated by cross validations, both EDK and MREG are suitable for regionalising the BFI with coefficients of determination of 0.80 and 0.77, respectively. Furthermore, the plausibility of the estimated BFI values is shown by comparisons with parameters calculated from statistical and fractal analysis of daily flow time series, which represent a strong relation to the regionalised BFI values.

Modelling river discharge for large drainage basins: from lumped to distributed approach

The paper presents an upscaled application of the HBV model to the German part of the Elbe drainage basin, and intercomparison of lumped and distributed versions of the model. The objectives of the work were (a) to check the model performance for large-scale basins, and (b) to compare the lumped and distributed versions of the model. Three versions of the HBV model, one lumped and two distributed, were applied first to a number of sub-basins of the Elbe with different hydrological regimes (area > 1000 km2), and then to the whole German part of the basin (area 80 657 km2). The model performed well in all cases. The distributed model versions are more data intensive but enabled better results to be achieved. The perspectives for using the model for large-scale water quality assessment, for climate change impact studies and for coupled land-atmosphere modelling are discussed.

On Critiques of “Stationarity is Dead: Whither Water Management?”

We review and comment upon some themes in the recent stream of critical commentary on the assertion that “stationarity is dead,” attempting to clear up some misunderstandings; to note points of agreement; to elaborate on matters in dispute; and to share further relevant thoughts.

Parameter and input data uncertainty estimation for the assessment of long-term soil organic carbon dynamics

The use of integrated soil organic matter (SOM) models to assess SOM dynamics under climate change, land use change and different land management practices require a quantification of uncertainties and key sensitive factors related to the respective modelling framework. Most uncertainty studies hereby focus on model parameter uncertainty, neglecting other sources like input data derived uncertainties, and spatial and temporal properties of uncertainty. Sources of uncertainties assessed in this study stem from uncertainties in model parameterisation and from uncertainties in model input data (climate, soil data, and land management assumptions). Thereby, Monte Carlo based global sensitivity and uncertainty analysis using a latin hypercube stratified sampling technique was applied to derive plot scale (focusing on temporal propagation) and river basin scale propagation of uncertainty for long-term soil organic carbon (SOC) dynamics. The model used is the eco-hydrological river basin model SWIM (Soil and Water Integrated Model), which has been extended by a process-based multi-compartment model for SOM turnover. Results obtained by this study can be transferred and used in other simulation models of this kind. Uncertainties resulting from all input factors used (model parameters+model input data) show a coefficient of variation between 5.1 and 6.7% and accounted for+/-0.065 to+/-0.3% soil carbon content (0.06-0.15t Cha^-^1yr^-^1). Parameter derived uncertainty contributed most to overall uncertainty. Concerning input data contributions, uncertainties stemming from soil and climate input data variations are striking. At the river basin scale, cropland and forest ecosystems, loess and gleyic soils possess the highest degree of uncertainty. Quantified magnitudes of uncertainty stemming from the examined sources vary temporally and spatially due to specific natural settings (e.g. climate, land use and soil properties) and deliver useful information for interpreting simulation results on long-term soil organic carbon dynamics under environmental change. Derived from this analysis, key sensitive model parameters and interactions between them were identified: the mineralization rate coefficient, the carbon use efficiency parameter (synthesis coefficient) along with parameters determining the soil temperature influence on SOM turnover (mainly Q10 value) and the soil input related data (soil bulk density and initial soil C content) introduced the highest degree of model uncertainty. The here gained information can be transferred to other process-based SOM turnover models to consider stronger most crucial parameters introducing highest uncertainty contribution to soil C storage assessment under changing environmental conditions.

Integrating groundwater dynamics in regional hydrological modelling

Abstract The paper presents an integrated catchment model and a method with which it is possible to analyse local water table dynamics inside subbasins along with river flow on the regional scale. A simple but comprehensive mechanistic groundwater module coupled with the eco-hydrological model SWIM (Soil and Water Integrated Model), which integrates hydrological processes, vegetation, erosion and nutrient dynamics at the watershed scale, was used in the study. The reliability of the model results was tested under well defined boundary conditions by comparing the results with those from a two dimensional numeric groundwater model under steady-state and transient conditions as well as with observed data for two meso-scale basins, using contour maps of the long term mean water table, observed groundwater level data in wells and observed river discharge. Especially in lowland catchments, where the water table is relatively shallow, the dynamics of river discharge are mainly influenced by changes in groundwater contribution to river flow. However, a correct reproduction of river discharge by hydrological models does not guarantee the adequacy of simulated spatio-temporal dynamics of soil moisture, water fluxes and groundwater in the basin. But even though the primary purpose of distributed hydrological models is to reproduce river discharge and water fluxes in the entire catchment, they are often validated using only the observed river discharge at the basin outlet for comparisons. The additional use of groundwater observations for model validation can serve as a measure to overcome the problem. The study area is located in the lowland part of the Elbe river basin, which is representative for semi-humid landscapes in Europe, where water availability during the summer season is the main limiting factor for plant growth and crop yields. The importance of adequate reproduction of the groundwater dynamics is illustrated in an investigation of a decreasing trend in regional groundwater level.