Catchment‐scale conceptual modelling of water and solute transport in the dual flow system of the karst critical zone

Hydrological and hydrochemical processes in the critical zone of karst environments are controlled by the fracture‐conduit network. Modelling hydrological and hydrochemical dynamics in such heterogeneous hydrogeological settings remains a research challenge. In this study, water and solute transport in the dual flow system of the karst critical zone were investigated in a 73.5‐km² catchment in southwest China. We developed a dual reservoir conceptual run‐off model combined with an autoregressive and moving average model with algorithms to assess dissolution rates in the “fast flow” and “slow flow” systems. This model was applied to 3 catchments with typical karst critical zone architectures, to show how flow exchange between fracture and conduit networks changes in relation to catchment storage dynamics. The flux of bidirectional water and solute exchange between the fissure and conduit system increases from the headwaters to the outfall due to the large area of the developed conduits and low hydraulic gradient in the lower catchment. Rainfall amounts have a significant influence on partitioning the relative proportions of flow and solutes derived from different sources reaching the underground outlet. The effect of rainfall on catchment function is modulated by the structure of the karst critical zone (e.g., epikarst and sinkholes). Thin epikarst and well‐developed sinkholes in the headwaters divert more surface water (younger water) into the underground channel network, leading to a higher fraction of rainfall recharge into the fast flow system and total outflow. Also, the contribution of carbonate weathering to mass export is also higher in the headwaters due to the infiltration of younger water with low solute concentrations through sinkholes.     

Snow model sensitivity analysis to understand spatial and temporal snow dynamics in a high‐elevation catchment

In this paper, we addressed a sensitivity analysis of the snow module of the GEOtop2.0 model at point and catchment scale in a small high‐elevation catchment in the Eastern Italian Alps (catchment size: 61 km²). Simulated snow depth and snow water equivalent at the point scale were compared with measured data at four locations from 2009 to 2013. At the catchment scale, simulated snow‐covered area (SCA) was compared with binary snow cover maps derived from moderate‐resolution imaging spectroradiometer (MODIS) and Landsat satellite imagery. Sensitivity analyses were used to assess the effect of different model parameterizations on model performance at both scales and the effect of different thresholds of simulated snow depth on the agreement with MODIS data. Our results at point scale indicated that modifying only the “snow correction factor” resulted in substantial improvements of the snow model and effectively compensated inaccurate winter precipitation by enhancing snow accumulation. SCA inaccuracies at catchment scale during accumulation and melt period were affected little by different snow depth thresholds when using calibrated winter precipitation from point scale. However, inaccuracies were strongly controlled by topographic characteristics and model parameterizations driving snow albedo (“snow ageing coefficient” and “extinction of snow albedo”) during accumulation and melt period. Although highest accuracies (overall accuracy = 1 in 86% of the catchment area) were observed during winter, lower accuracies (overall accuracy < 0.7) occurred during the early accumulation and melt period (in 29% and 23%, respectively), mostly present in areas with grassland and forest, slopes of 20–40°, areas exposed NW or areas with a topographic roughness index of ?0.25 to 0 m. These findings may give recommendations for defining more effective model parameterization strategies and guide future work, in which simulated and MODIS SCA may be combined to generate improved products for SCA monitoring in Alpine catchments.     

The Weierbach experimental catchment in Luxembourg: A decade of critical zone monitoring in a temperate forest - from hydrological investigations to ecohydrological perspectives

The Weierbach experimental catchment (0.45 km²) is the most instrumented and studied sub-catchment in the Alzette River basin in Luxembourg. Within the last decade, it has matured towards an interdisciplinary critical zone observatory focusing on a better understanding of hydrological and hydro-geochemical processes. The Weierbach catchment is embedded in an elevated sub-horizontal plateau, characterized by slate bedrock and representative of the Ardennes Massif. Its climate is semi-marine, with precipitation being rather evenly distributed throughout the year. Base flow is lowest from July to September, essentially due to higher losses through evapotranspiration in summer. The regolith is composed of Devonian slates, overlaid by Pleistocene slope deposits and entirely covered by forest with 70% deciduous and 30% coniferous trees. Since 2009, the Weierbach has been extensively equipped for continuously monitoring water fluxes and physico-chemical parameters within different compartments of the critical zone. Additionally, these compartments are sampled fortnightly at several locations to analyze δ¹⁸O and δ²H isotopic composition of water including rainfall, throughfall, soil water, groundwater and streamwater. This ongoing monitoring and sampling programme is used for answering pressing questions related to fundamental catchment functions of water infiltration, storage, mixing and release in forest ecosystems. A recently started research line aims at investigating interactions between forest eco-hydrosystems with the atmosphere and understanding how catchments will respond to a non-stationary climate.     

Investigating mechanisms of stormflow generation by natural tracers and hydrometric data: a small catchment study in the Black Forest,Germany

The importance and interaction of various hydrological pathways and identification of runoff source areas involved in solute transport are still under considerable debate in catchment hydrology. To reveal stormflow generating areas and flow paths, hydrometric behaviour of throughfall, soil water from various depths, runoff, and respective concentrations of the environmental tracers ¹⁸O, Si, K, SO₄ and dissolved organic carbon were monitored for a 14‐week period in a steep headwater catchment in the Black Forest Mountains, Germany. Two stormflow hydrographs were selected and, based on ¹⁸O and Si, chemically separated into three flow components. Their sources were defined using mixing diagrams. Additional information about stormflow generating mechanisms was derived from recession analyses of the basin's complete 5‐year hydrograph record. By providing insight into storage properties and residence times of outflowing reservoirs of the basin, recession analysis proved to be a valuable tool in runoff model conceptualization. Its results agreed well with hydrometric and hydrochemical data. Supported by evaluation of 30 hillslope soil profiles a coherent concept of stormflow generation could be derived: whereas in many steeply sloped basins in the temperate region soil water from hillslopes appears to have an immediate effect on the shape of the stormflow hydrograph, its role at this basin is basically restricted to the recharge of the groundwater reservoir in the near‐channel area. Storm hydrograph peaks appear to be derived from a small direct runoff component supplemented by a fast delivery of baseflow from the groundwater reservoir in the valley bottom. Copyright © 2001 John Wiley & Sons, Ltd.     

Contribution of glacier meltwater to streamflow in the Wind River Range,Wyoming, inferred via a Bayesian mixing model applied to isotopic measurements

The Wind River Range (WRR) of Wyoming has the largest concentration of alpine glaciers in the American Rockies and contributes to several major river systems in the western United States. Declines in the areal extent and volume of these glaciers are well documented, and eventual loss of alpine glaciers will reduce the amount of water available for agricultural and domestic use. The contribution of glacial melt to streamflow remains largely unquantified in Wyoming. We used isotope measurements and Bayesian modeling to estimate the fractional contribution of glacier meltwater to Dinwoody Creek (DC) in the WRR on bi‐weekly and seasonal (spring, summer, and fall) time scales over 2 years. In 2007 and 2008, we made temporally intensive measurements of the stable isotope composition of water from the DC watershed. Samples of the primary sources of streamflow (snowmelt, glacier melt, rain, and baseflow) were collected during field campaigns, and automated collection of stream samples occurred over the melt season. Isotope data (D and ¹⁸O) were analyzed within a hierarchical Bayesian framework that incorporated temporal and spatial correlations. Glacial melt contributed a significant proportion (～53–59%) to streamflow in a low‐flow year (2007) or when streamflow was low during a high‐flow year (2008). In 2008, a large and persistent snowpack contributed significantly (～0·42–51%) to streamflow in mid‐summer. The large contribution of glacial melt to streamflow suggests that the loss of glaciers may impact riparian ecosystems and human water supplies in the late summer and in years with low snowpack. Copyright © 2011 John Wiley & Sons, Ltd.     

Testing a spatially distributed tracer‐aided runoff model in a snow‐influenced catchment: Effects of multicriteria calibration on streamwater ages

Integrating stable isotope tracers into rainfall‐runoff models allows investigation of water partitioning and direct estimation of travel times and water ages. Tracer data have valuable information content that can be used to constrain models and, in integration with hydrometric observations, test the conceptualization of catchment processes in model structure and parameterization. There is great potential in using tracer‐aided modelling in snow‐influenced catchments to improve understanding of these catchments' dynamics and sensitivity to environmental change. We used the spatially distributed tracer‐aided rainfall‐runoff (STARR) model to simulate the interactions between water storage, flux, and isotope dynamics in a snow‐influenced, long‐term monitored catchment in Ontario, Canada. Multiple realizations of the model were achieved using a combination of single and multiple objectives as calibration targets. Although good simulations of hydrometric targets such as discharge and snow water equivalent could be achieved by local calibration alone, adequate capture of the stream isotope dynamics was predicated on the inclusion of isotope data in the calibration. Parameter sensitivity was highest, and most local, for single calibration targets. With multiple calibration targets, key sensitive parameters were still identifiable in snow and runoff generation routines. Water ages derived from flux tracking subroutines in the model indicated a catchment where runoff is dominated by younger waters, particularly during spring snowmelt. However, resulting water ages were most sensitive to the partitioning of runoff sources from soil and groundwater sources, which was most realistically achieved when isotopes were included in the calibration. Given the paucity of studies where hydrological models explicitly incorporate tracers in snow‐influenced regions, this study using STARR is an important contribution to satisfactorily simulating snowpack dynamics and runoff generation processes, while simultaneously capturing stable isotope variability in snow‐influenced catchments.     

Run‐off processes from mountains to foothills: The role of soil stratigraphy and structure in influencing run‐off characteristics across high to low relief landscapes

The critical zone features that control run‐off generation, specifically at the regional watershed scale, are not well understood. Here, we addressed this knowledge gap by quantitatively and conceptually linking regional watershed‐scale run‐off regimes with critical zone structure and climate gradients across two physiographic provinces in the Southeastern United States. We characterized long‐term (~20 years) discharge and precipitation regimes for 73 watersheds with United States Geological Survey in‐stream gaging stations across the Appalachian Mountain and Piedmont physiographic provinces of North Carolina. Watersheds included in this analysis had <10% developed land and ranged in size from 14.1–4,390 km². Thirty‐four watersheds were located in the Piedmont physiographic province, which is typically classified as a low relief landscape with deep, highly weathered soils and regolith. Thirty‐nine watersheds were located in the Appalachian Mountain physiographic province, which is typically classified as a steeper landscape with highly weathered, but shallower soils and regolith. From the United States Geological Survey daily mean run‐off time series, we calculated annual and seasonal baseflow indices (BFI), minimum, mean, and maximum daily run‐off, and Pearson's correlation coefficients between precipitation and baseflow. Our results showed that Appalachian Mountain watersheds systematically had higher minimum daily flows and BFI values. Piedmont watersheds displayed much larger deviations from mean annual BFI in response to year‐to‐year variability in precipitation. A series of linear regression models between 21 landscape metrics and annual BFIs showed non‐linear and complex terrestrial–hydrological relationships across the two provinces. From these results, we discuss how distinct features of critical zone architecture, with specific focus on soil depth and stratigraphy, may be dominating the regulation of hydrological processes and run‐off regimes across these provinces.     

Interannual variability in glacier contribution to runoff from a high‐elevation Andean catchment: understanding the role of debris cover in glacier hydrology

We present a field‐data rich modelling analysis to reconstruct the climatic forcing, glacier response, and runoff generation from a high‐elevation catchment in central Chile over the period 2000–2015 to provide insights into the differing contributions of debris‐covered and debris‐free glaciers under current and future changing climatic conditions. Model simulations with the physically based glacio‐hydrological model TOPKAPI‐ETH reveal a period of neutral or slightly positive mass balance between 2000 and 2010, followed by a transition to increasingly large annual mass losses, associated with a recent mega drought. Mass losses commence earlier, and are more severe, for a heavily debris‐covered glacier, most likely due to its strong dependence on snow avalanche accumulation, which has declined in recent years. Catchment runoff shows a marked decreasing trend over the study period, but with high interannual variability directly linked to winter snow accumulation, and high contribution from ice melt in dry periods and drought conditions. The study demonstrates the importance of incorporating local‐scale processes such as snow avalanche accumulation and spatially variable debris thickness, in understanding the responses of different glacier types to climate change. We highlight the increased dependency of runoff from high Andean catchments on the diminishing resource of glacier ice during dry years.     

Identifying large‐scale erosion and deposition processes from airborne gamma radiometrics and digital elevation models in a weathered landscape

Patterns of erosion and deposition are difficult to identify and measure at catchment and regional scale but it may be possible to infer their distribution from remote sensing using easily measured surrogate variables. Airborne geophysical surveys provide data on gamma ray emissions from surface and near‐surface material and allow estimation of K, Th and U content. Gamma ray signatures are largely determined by lithology but also change with weathering and with erosion and deposition, and may be used as a partial surrogate for those processes. Comparison of gamma ray signatures with topographic characteristics closely related to sediment transport capacity, downstream sorting of sediments, and the extent of erosion and deposition shows strong linkages. Studies in four small catchments in southeastern Australia indicate that K consistently increases as slopes become steeper while Th and U may either increase or decrease. This suggests the presence of fresh rock rather than weathered material and implies removal of material by erosion, although some patterns may result from systematic changes in lithology across catchments. Analysis by lithology confirms the increase in K with slope in granites, metamorphosed sediments and basalt, and also shows a tendency for U to decrease with slope in the granites and basalt. Gamma ray emissions vary only slightly with catchment area (a surrogate measure of water discharge) suggesting that water erosion is limited or that discharge is not closely related to area. Gamma radiometric profiles down hillslopes, averaged across the full range of gradients, show that, in most cases, radioelements initially decrease, probably reflecting increased weathering, but then increase close to valley floors because of accumulation of fine sediments. Analysis by lithology confirms the increase in radioelements close to valley floors in granites and metamorphosed sediments but the trend is less clear in basalt. Gamma ray profiles down floodplains are variable and indicate the amount of deposition and accumulation of weathered material in valley openings and exposure of bedrock in valley constrictions. Simple erosion and deposition models, based on the conservation of mass equation, and applied to the four catchments, show that all radioelements increase as the potential for deposition increases. They reach a low point in zones of no net erosion or deposition and subsequently increase as erosion becomes more intense and weathered material is lost from slopes. Analysis by lithology largely confirms this pattern. The usefulness of airborne geophysical survey data is limited by flight line spacing with most data being flown at a 200–400 m spacing in Australia. However, general trends in erosion and deposition can still be distinguished and there is capacity for calibrating long‐term erosion and deposition models once better approaches to interpretation of gamma ray data have been developed. Copyright © 2000 John Wiley & Sons, Ltd.     

Sediment production in large gullies of the Mediterranean area (NE Spain) from high‐resolution digital elevation models and geographical information systems analysis

Recent studies in the Mediterranean area have shown gully erosion to have a very significant contribution to total soil loss. In the Penedès vineyard region (NE Spain), between 15 and 27% of the land is affected by large gullies and gully‐wall retreat seems to be an ongoing process. Multi‐date digital elevation model (DEM) analysis has allowed computation of sediment production by gully erosion, showing that the sediment production rates are very high by the, up‐to‐date, usual global standards. Here, we present a study carried out using large‐scale multi‐date (1975 and 1995) aerial photographs (1 : 5000 and 1 : 7000) to monitor sediment yield caused by large gullies in the Penedès region (NE Spain). High‐resolution DEMs (1 m grid) were derived and analysed by means of geographical information systems techniques to determine the gully erosion rates. Rainfall characteristics within the same study period were also analysed in order to correlate with the soil loss produced. Mass movement was the main process contributing to total sediment production. This process could have been favoured by rainfalls recorded during the period: 58% of the events were of an erosive character and showed high kinetic energy and erosivity. A sediment production rate of 846 ± 40 Mg ha^?1 year^?1, a sediment deposition rate of 270 ± 18 Mg ha^?1 year^?1 and a sediment delivery ratio of 68·1% were computed for a gully area of 0·10 km². The average net erosion within the study period (1975–95) was 576 ± 58 Mg ha^?1 year^?1. In comparison with other methods, the proposed method also includes sediment produced by processes other than only overland flow, i.e. downcutting, headcutting, and mass movements and bank erosion. Copyright © 2003 John Wiley & Sons, Ltd.     

Significance of processes in the near‐stream zone on stream water acidity in a small acidified forested catchment

The near‐stream zone has received increasing attention owing to its influence on stream water chemistry in general and acidity in particular. Possible processes in this zone include cation exchange, leaching of organic matter and redox reactions of sulphur compounds. In this study the influences of processes in the near‐stream zone on the acidity in runoff from a small, acidified catchment in central southern Sweden were investigated. The study included sampling of groundwater, soil water and stream water along with hydrological measurements. An input–output budget for the catchment was established based on data from the International Co‐operative Programme on Integrated Monitoring at this site. The catchment was heavily acidified by deposition of anthropogenic sulphur, with pH in stream water between 4·4 and 4·6. There was also no relationship between stream flow and pH, which is indicative of chronic acidification. Indications of microbial reduction of sulphate were found in some places near the stream, but the near‐stream zone did not have a general impact on the sulphate concentration in discharging groundwater. The near‐stream zone was a source of dissolved organic carbon (DOC) in the stream, which had a median DOC of 6·8 mg L¹. The influence on stream acidity from organic anions was overshadowed by the effect of sulphate, however, except during a spring flow episode, when additional organic matter was flushed out and the sulphate‐rich ground water was mixed with more diluted event water. Ion exchange was not an important process in the near‐stream zone of the Kindla catchment. Different functions of the near‐stream zone relating to discharge acidity are reported in the literature. In this study there was even a variation within the site. There is therefore a need for more case studies to provide a more detailed understanding of the net effects that the near‐stream zone can have on stream chemistry under different circumstances. Copyright © 2001 John Wiley & Sons, Ltd.     

Prediction of regional-scale groundwater recharge and nitrate storage in the vadose zone: A comparison between a global model and a regional model

Extensive nitrogen loads at the soil surface exceed plant uptake and soil biochemical capacity, and therefore lead to nitrogen accumulation in the deep vadose zone. Studies have shown that stored nitrogen in the vadose zone can eventually reach the water table and affect the quality of groundwater resources. Recently, global scale models have been implemented to quantify nitrate storage and nitrate travel time in the vadose zone. These global models are simplistic and relatively easy to implement and therefore facilitate analysis of the considered transport processes at a regional scale with no further requirements. However, the suitability of applying these models at a regional scale has not been tested. Here, we evaluate, for the first time, the performance and utility of global scale models at the regional scale. Applied to the Loess Plateau of China, we compare estimates of groundwater recharge and nitrate storage derived from global scale models with results from a regional scale approach utilizing the Richards and advection-dispersion equations. The estimated nitrate storage was compared to nitrate observations collected in the deep vadose zone (>50 m) at five sites across the Loess Plateau. Although both models predict similar spatial patterns of nitrate storage, the recharge fluxes were three times smaller and the nitrate storage was two times higher compared with the regional model. The results suggest that global scale models are a potentially useful screening tool, but require refinement for local scale applications.     

Empirical and conceptual modelling of the suspended sediment dynamics in a large tropical African river: the Upper Niger river basin

A 7-year sediment transport monitoring on the Upper Niger rivers was used to study the relationship between suspended sediment concentration and river discharge. During annual floods, these relationships show positive hysteresis. This paper presents the results of two models that estimate the time evolution of suspended sediment concentration using water discharge data only. The first model is based on a statistical approach using two relationships, one for the rising stage period of the flood and one for the recession period of the annual flood; the second model is a lumped conceptual one; it supposes that the sediment flux observed in the river comes from two different sources of sediment and that these two sources may be regarded as two different reservoirs. The erosion of the first reservoir represents hillslope erosion observed during the runoff season. Sediment supply from this ‘reservoir’ is limited in time because depletion occurs during the runoff season. The second reservoir is unlimited in time and quantity and its erosion represents contributions coming from bank erosion and mobilisation of deposits in the channel network.

Both of the models are compared with a simple rating curve based model. The model results show that the conceptual model has the highest efficiency to reproduce from weekly discharge only the time evolution of weekly suspended sediment concentrations, the time evolution of weekly sediment fluxes, and the global annual sediment yields.     

Meta‐analysis of flow modeling performances—to build a matching system between catchment complexity and model types

Hydrological models play a significant role in modelling river flow for decision making support in water resource management. In the past decades, many researchers have made a great deal of efforts in calibrating and validating various models, with each study being focused on one or two models. As a result, there is a lack of comparative analysis on the performance of those models to guide hydrologists to choose appropriate models for the individual climate and physical conditions. This paper describes a two‐level meta‐analysis to develop a matching system between catchment complexity (based on catchment significant features (CSFs)) and model types. The intention is to use the available CSF information for choosing the most suitable model type for a given catchment. In this study, the CSFs include the elements of climate, soil type, land cover and catchment scale. Specific choices of model types in small and medium catchments are further explored with all CSF information obtained. In particular, it is interesting to find that semi‐distributed models are the most suitable model type for catchments with the area over 3000 km², regardless of other CSFs. The potential methodology for expanding the matching system between catchment complexity and model complexity is discussed. Copyright © 2014 John Wiley & Sons, Ltd.     

Combined use of isotopic and hydrometric data to conceptualize ecohydrological processes in a high‐elevation tropical ecosystem

Few high‐elevation tropical catchments worldwide are gauged, and even fewer are studied using combined hydrometric and isotopic data. Consequently, we lack information needed to understand processes governing rainfall–runoff dynamics and to predict their influence on downstream ecosystem functioning. To address this need, we present a combination of hydrometric and water stable isotopic observations in the wet Andean páramo ecosystem of the Zhurucay Ecohydrological Observatory (7.53 km²). The catchment is located in the Andes of south Ecuador between 3400 and 3900 m a.s.l. Water samples for stable isotopic analysis were collected during 2 years (May 2011–May 2013), while rainfall and runoff measurements were continuously recorded since late 2010. The isotopic data reveal that andosol soils predominantly situated on hillslopes drain laterally to histosols (Andean páramo wetlands) mainly located at the valley bottom. Histosols, in turn, feed water to creeks and small rivers throughout the year, establishing hydrologic connectivity between wetlands and the drainage network. Runoff is primarily composed of pre‐event water stored in the histosols, which is replenished by rainfall that infiltrates through the andosols. Contributions from the mineral horizon and the top of the fractured bedrock are small and only seem to influence discharge in small catchments during low flow generation (non‐exceedance flows < Q₃₅). Variations in source contributions are controlled by antecedent soil moisture, rainfall intensity, and duration of rainy periods. Saturated hydraulic conductivity of the soils, higher than the year‐round low precipitation intensity, indicates that Hortonian overland flow rarely occurs during high‐intensity precipitation events. Deep groundwater contributions to discharge seem to be minimal. These results suggest that, in this high‐elevation tropical ecosystem, (1) subsurface flow is a dominant hydrological process and (2) (histosols) wetlands are the major source of stream runoff. Our study highlights that detailed isotopic characterization during short time periods provides valuable information about ecohydrological processes in regions where very few basins are gauged. Copyright © 2016 John Wiley & Sons, Ltd.     

A comparison of high‐resolution specific conductance‐based end‐member mixing analysis and a graphical method for baseflow separation of four streams in hydrologically challenging agricultural watersheds

Quantifying the relative contributions of different sources of water to a stream hydrograph is important for understanding the hydrology and water quality dynamics of a given watershed. To compare the performance of two methods of hydrograph separation, a graphical program [baseflow index (BFI)] and an end‐member mixing analysis that used high‐resolution specific conductance measurements (SC‐EMMA) were used to estimate daily and average long‐term slowflow additions of water to four small, primarily agricultural streams with different dominant sources of water (natural groundwater, overland flow, subsurface drain outflow, and groundwater from irrigation). Because the result of hydrograph separation by SC‐EMMA is strongly related to the choice of slowflow and fastflow end‐member values, a sensitivity analysis was conducted based on the various approaches reported in the literature to inform the selection of end‐members. There were substantial discrepancies among the BFI and SC‐EMMA, and neither method produced reasonable results for all four streams. Streams that had a small difference in the SC of slowflow compared with fastflow or did not have a monotonic relationship between streamflow and stream SC posed a challenge to the SC‐EMMA method. The utility of the graphical BFI program was limited in the stream that had only gradual changes in streamflow. The results of this comparison suggest that the two methods may be quantifying different sources of water. Even though both methods are easy to apply, they should be applied with consideration of the streamflow and/or SC characteristics of a stream, especially where anthropogenic water sources (irrigation and subsurface drainage) are present. Copyright © 2014 John Wiley & Sons, Ltd.     

Investigating the use of spatial discretization of hydrological processes in conceptual rainfall runoff modelling: a case study for the meso‐scale

In this study a simple modelling approach was applied to identify the need for spatial complexity in representing hydrological processes and their variability over different scales. A data set of 18 basins was used, ranging between 8 and 4011 km² in area, located in the Nahe basin (Germany), with daily discharge values for over 30 years. Two different parsimoniously structured models were applied in lumped as well as in spatially distributed according to two distribution classifications: (1) a simple classification based on the lithology expressed in three permeability types and (2) a more complex classification based on seven dominating runoff production processes. The objective of the study was to compare the performances of the models on a local and on a regional scale as well as between the models with a view to identifying the accuracy in capturing the spatial variability of the rainfall‐runoff relationships. It was shown that the presence of a specific basin characteristic or process of the distribution classification was not related with higher model performance; only a larger basin size promoted higher model performance. The results of this study also indicated that the permeability generally contained more useful information on the spatial heterogeneity of the hydrological behaviour of the natural system than did a more detailed classification on dominating runoff generation processes. Although model performance was slightly lower for the model that used permeability as a distribution classification, consistency in its parameter values was found, which was lacking with the more complex distribution classification. The latter distribution classification had a higher flexibility to optimize towards the variability of the runoff, which resulted in higher performance, however, process representation was applied inconsistently. Copyright © 2007 John Wiley & Sons, Ltd.     

Projection of climate change for daily precipitation: a case study in Shih‐Men reservoir catchment in Taiwan

A statistical downscaling model, based on the outputs of general circulation models (GCMs) as predictors, was proposed to simulate the daily precipitations in the Shih‐Men reservoir catchment in Taiwan. The structure of the proposed downscaling model is composed of two parts: classification and regression. Predictors of classification and regression models were selected from the large‐scale weather variables in the National Centers for Environmental Prediction (NECP) reanalysis data based on statistical tests. Discriminant analysis and support vector regression (SVR) were applied to build the classification and regression models. The outputs of five atmosphere‐ocean GCMs, which are reported to have properly considered tropical cyclone information and East Asian Monsoon modelling, were used for projecting future precipitations. Data from four grids covering Taiwan were used for developing the downscaling model. The potential of the downscaling models in simulating local precipitations was evaluated, and downscaling results reveal that the proposed downscaling model can reproduce local daily precipitations from large‐scale weather variables. Projected local precipitations under two emission scenarios show that the precipitations in the study area tend to decrease. Copyright © 2010 John Wiley & Sons, Ltd.