The process basis of existing soil-erosion models is shown to be ill-founded. The existing literature builds directly or indirectly on Bennett's (1974) paper, which provided a blueprint for integrated catchment-scale erosion modelling. Whereas Bennett recognized the inherent assumptions of the approach suggested, subsequent readings of the paper have led to a less critical approach. Most notably, the assumption that sediment movement could be approximated by a continuity equation that related to transport in suspension has produced a series of submodels that assume that all movement occurs in suspension. For commonly occurring conditions on hillslopes, this case is demonstrably untrue both on theoretical grounds and from empirical observations. Elsewhere in the catchment system, it is only partially true, and the extent to which the assumption is reasonable varies both spatially and temporally. A second ground-breaking paper -that of Foster and Meyer (1972) -was responsible for subsequent uncritical application of a first-order approximation to deposition based on steady-state analysis and again a weak empirical basis. We describe in this paper an alternative model (MAHLERAN -Model for Assessing Hillslope-Landscape Erosion, Runoff And Nutrients) based upon particle-travel distance that overcomes existing limitations by incorporating parameterizations of the different detachment and transport mechanisms that occur in water erosion in hillslopes and small catchments. In the second paper in the series, we consider the sensitivity and general behaviour of MAHLERAN, and test it in relation to data from a large rainfall-simulation experiment. The third paper of the sequence evaluates the model using data from plots of different sizes in monitored rainfall events. From this evaluation, we consider the scaling characteristics of the current form of MAHLERAN and suggest that integrated modelling, laboratory and field approaches are required in order to advance the state of the art in soil-erosion modelling. Wainwright et al., 2001;Parsons et al., 2004Parsons et al., , 2006b. We have argued elsewhere (Parsons et al., 2004) that these problems arise from a fundamental misconception, and thus misrepresentation, of the component processes that make up soil erosion. The aim of this paper is, therefore, to review the conceptual basis of existing models and to propose an alternative model that addresses the problems inherent in existing approaches. Figure 2. Comparison of information on (a) travel distances and (b) virtual velocities from Hassan et al. (1992) for concentrated flows and Parsons et al. (1998) for unconcentrated flows. See the text for further discussion.822 J. Wainwright et al.Figure 4. Summary of the calculation algorithm used in MAHLERAN. The main components in terms of detachment and the use of travel distance and virtual velocity to estimate sediment discharge are highlighted. 824 J. Wainwright et al.appropriate erosion and deposition rates have been calculated. The sediment mass-balance Equation (6) is equiva...

Abstract. This study aims to compare impacts of climate change on streamflow in four large representative African river basins: the Niger, the Upper Blue Nile, the Oubangui and the Limpopo. We set up the eco-hydrological model SWIM (Soil and Water Integrated Model) for all four basins individually. The validation of the models for four basins shows results from adequate to very good, depending on the quality and availability of input and calibration data. For the climate impact assessment, we drive the model with outputs of five bias corrected Earth system models of Coupled Model Intercomparison Project Phase 5 (CMIP5) for the representative concentration pathways (RCPs) 2.6 and 8.5. This climate input is put into the context of climate trends of the whole African continent and compared to a CMIP5 ensemble of 19 models in order to test their representativeness. Subsequently, we compare the trends in mean discharges, seasonality and hydrological extremes in the 21st century. The uncertainty of results for all basins is high. Still, climate change impact is clearly visible for mean discharges but also for extremes in high and low flows. The uncertainty of the projections is the lowest in the Upper Blue Nile, where an increase in streamflow is most likely. In the Niger and the Limpopo basins, the magnitude of trends in both directions is high and has a wide range of uncertainty. In the Oubangui, impacts are the least significant. Our results confirm partly the findings of previous continental impact analyses for Africa. However, contradictory to these studies we find a tendency for increased streamflows in three of the four basins (not for the Oubangui). Guided by these results, we argue for attention to the possible risks of increasing high flows in the face of the dominant water scarcity in Africa. In conclusion, the study shows that impact intercomparisons have added value to the adaptation discussion and may be used for setting up adaptation plans in the context of a holistic approach.

This study intends to contribute to the ongoing discussion on whether land use and land cover changes (LULC) or climate trends have the major influence on the observed increase of flood magnitudes in the Sahel. A simulation-based approach is used for attributing the observed trends to the postulated drivers. For this purpose, the ecohydrological model SWIM (Soil and Water Integrated Model) with a new, dynamic LULC module was set up for the Sahelian part of the Niger River until Niamey, including the main tributaries Sirba and Goroul. The model was driven with observed, reanalyzed climate and LULC data for the years 1950-2009. In order to quantify the shares of influence, one simulation was carried out with constant land cover as of 1950, and one including LULC. As quantitative measure, the gradients of the simulated trends were compared to the observed trend. The modeling studies showed that for the Sirba River only the simulation which included LULC was able to reproduce the observed trend. The simulation without LULC showed a positive trend for flood magnitudes, but underestimated the trend significantly. For the Goroul River and the local flood of the Niger River at Niamey, the simulations were only partly able to reproduce the observed trend. In conclusion, the new LULC module OPEN ACCESSWater 2015, 7 2797 enabled some first quantitative insights into the relative influence of LULC and climatic changes. For the Sirba catchment, the results imply that LULC and climatic changes contribute in roughly equal shares to the observed increase in flooding. For the other parts of the subcatchment, the results are less clear but show, that climatic changes and LULC are drivers for the flood increase; however their shares cannot be quantified. Based on these modeling results, we argue for a two-pillar adaptation strategy to reduce current and future flood risk: Flood mitigation for reducing LULC-induced flood increase, and flood adaptation for a general reduction of flood vulnerability.

In the first paper in this series, we demonstrated that most process-based erosion models have a series of in-built assumptions that led us to question their true process basis. An alternative soil-erosion model (MAHLERAN -Model for Assessing Hillslope-Landscape Erosion, Runoff And Nutrients) based upon particle-travel distance has been presented in the first paper in this series and this paper presents the first of two evaluations of the model. Here, a sensitivity analysis shows that the numerical model is consistent with the analytical model of Parsons et al. (2004) and demonstrates that downslope patterns of sediment flux on hillslopes are a complex interaction of rainfall intensity, duration and pattern; hillslope gradient; surface roughness and sediment size. This result indicates that the spatial scaling of sediment transfers on hillslopes is a non-trivial problem and will vary from point to point and from event to event and thus from year to year. The model is evaluated against field data from a rainfall-simulation experiment on an 18 m × × × × × 35 m plot for which there are sub-plot-scale data on runoff hydraulics and sediment flux. The results show that the model is capable of reproducing the sedigraph with an overall normalized root-mean-square error of 18·4% and Nash-Sutcliffe efficiency of 0·90. Spatial and temporal patterns of particle-size distributions of the eroded sediment are also reproduced very well, once erosion parameters have been optimized for the specific soil conditions. Figure 2. Sensitivity analysis on a uniform, planar 100 m long × 30 m wide slope according to rainfall intensity considering both unconcentrated and concentrated erosion.Figure 6. Comparisons of spatial patterns of runoff and sediment production for the E2 event on the large shrubland plot at Walnut Gulch: (a) total runoff in litres at a point for the Equation (2) case; (b) total runoff in litres at a point for the Equation (3b) case; (c) total sediment movement in kilograms at a point for the Equation (2) case; (d) total sediment movement in kilograms at a point for the Equation (3b) case. This figure is available in colour online at www.interscience.wiley.com/journal/espl 974 J. Wainwright et al.Figure 7. Summary of discharge and erosion characteristics of the simulations of E2 on the large shrubland plot at Walnut Gulch. The x-axis represents the distance from the top of the plot (0 m) to the outlet at 35 m.Figure A1. Sensitivity analysis on a uniform, planar 100 m long × 30 m wide slope according to rainfall duration considering unconcentrated erosion only. The x-axis represents the distance from the top of the plot (0 m) to the outlet at 35 m. Each graph shows the sum of all the model cells at a given distance downslope for total flow discharge (m 3 ), sediment flux (kg m −1 ), sediment yield (kg m −2 ), detachment (kg), deposition (kg), net erosion (kg) and percentage of all detachment that is by raindrops (i.e. 100% = detachment entirely by raindrops, 0% = detachment entirely by concentrated flow erosion). 978 J...

In the two previous papers of this series, we demonstrated how a novel approach to erosion modelling (MAHLERAN -Model for Assessing Hillslope-Landscape Erosion, Runoff AndNutrients) provided distinct advantages in terms of process representation and explicit scaling characteristics when compared with existing models. A first evaluation furthermore demonstrated the ability of the model to reproduce spatial and temporal patterns of erosion and their particle-size characteristics on a large rainfall-simulation plot. In this paper, we carry out a more detailed evaluation of the model using monitored erosion events on plots of different size. The evaluation uses four plots of 21·01, 115·94, 56·84 and 302·19 m 2 , with lengths of 4·12, 14·48, 18·95 and 27·78 m, respectively, on similar soils to the rainfall-simulation plot, for which runoff and erosion were monitored under natural rainfall. Although the model produces the correct ranking of the magnitude of erosion events, it performs less well in reproducing the absolute values and particle-size distributions of the eroded sediment. The implications of these results are evaluated in terms of requirements for process understanding and data for parameterization of improved soil-erosion models. We suggest that there are major weaknesses in the current understanding and data underpinning existing models. Consequently, a more holistic re-evaluation is required that produces functional relationships for different processes that are mutually consistent, and that have appropriate parameterization data to support their use in a wide range of environmental conditions.To evaluate the performance of the model at a range of scales, detailed testing data are required that meet several criteria. First, they must relate to a variety of scales of measurement, with other environmental variability reduced to a minimum. Secondly, there must be detailed rainfall, runoff and sediment-production data for a range of storm events. These data must be of a high quality so that confidence can be placed in the test results. Thirdly, it is helpful if the focus of the test is on the sediment-transport component, and thus the data should relate to conditions where the hydrological conditions are reasonably well understood. The data presented by Brazier et al., 2007) meet all of these criteria. Four runoff plots (named Laurel, Abbott, Dud and Wise) were constructed at the Walnut Gulch Experimental Watershed, AZ, USA (31° 44′ 23″ N, 110° 3′ 53″ W) in areas with soils described as being on coarse-loamy, mixed, thermic, Ustochreptic Calciorthids and fine-loamy, mixed, thermic Ustalfic Haplargids (Breckenfield et al., 1995) and vegetation dominated by creosotebush (Larrea tridentata), albeit with a number of other shrub species including Acacia constricta, Dasylirion wheeleri, Rhus microphylla and Yucca elata, with a ground layer dominated by Dyssodia acerosa and Zinnia pumila. These plots had areas of 21·01, 115·94, 56·84 and 302·19 m 2 , respectively, and lengths of 4·12, 14·48, 18·95 and 27·78 m, resp...