Elsevier

CATENA

Volume 46, Issues 2–3, 3 January 2002, Pages 189-205
CATENA

Incorporating soil surface crusting processes in an expert-based runoff model: Sealing and Transfer by Runoff and Erosion related to Agricultural Management

https://doi.org/10.1016/S0341-8162(01)00166-7Get rights and content

Abstract

In European loess belt soils, infiltration and erosion processes are strongly influenced by surface crusting. Modelling infiltration into these crusts has led to the development of equations of varying complexity, ranging from simple empirical equations to numerical solution of the Richard's equation. However, a number of issues important for modelling effective erosion at the catchment scale remain unsolved. The objective of this study was to contribute to the elaboration of an expert-based runoff prediction model able to simulate the influence of soil conservation practices in the context of loess soils susceptible to crusting. Experiments have been implemented both in the laboratory and in the field at various scales ranging from small plots up to catchments. The experimental results provided a set of reference infiltration and runoff data under a variety of different situations in terms of weather conditions, surface state, land use and agricultural practices. Infiltrability ranged from 2 mm h−1 for crusted surfaces up to more than 30 mm h−1 for undegraded surfaces. These references were used to develop decision rules in the forms of matching tables to characterise agricultural fields with an infiltration capacity for a given rainfall event. For the area of the Pays-de-Caux (Normandie, France), we defined five potential infiltrability classes of 2, 5, 10, 20 and 50 mm h−1. A runoff circulation network calculated from a Digital Elevation Model (DEM) combined with information on field operations allows the calculation of total runoff volume for a rainfall event at any point of the catchment. Calculated and measured runoff for a first series of events were in satisfactory accordance.

Introduction

Serious environmental and economical problems due to soil erosion are a global phenomenon. In the northwestern Paris Basin and in many cultivated areas of the loessian belt in Northern Europe, the erosion hazards are related to the soil susceptibility to sealing and crusting Auzet et al., 1990, King and Le Bissonnais, 1992. The consequences are of concern to the local authorities which have to face costs induced by property damage by soil-laden water, road clearance and watercourse pollution both by sediments and agricultural chemicals Papy and Douyer, 1991, Boardman et al., 1994. In order to characterise the dominant surface processes leading to overland flow, several research efforts have been carried out in the Haute Normandie region (France). Experiments have been implemented both in the laboratory and in the field at various scales ranging from small plots (Fox and Le Bissonnais, 1998) up to catchments Le Bissonnais et al., 1998a, Le Bissonnais et al., 1998b. Thereby, a set of reference infiltration and runoff data was collected under a variety of different situations in terms of weather conditions, surface state, land use and agricultural practices. The main purpose was to make a synthesis of these reference data in the elaboration of a simple event based runoff model able to simulate the effect of conservation measures in small cultivated catchments.

Numerous researches are focused on the development of runoff and erosion models, as for example, the Water Erosion Prediction Project (Nearing et al., 1989) or the EUROpean Soil Erosion Model (Morgan et al., 1994). Different approaches are developed to serve particular purposes which range from farm planning to conceptual tools used to enhance our understanding of complex hydrological processes DeCoursey, 1985, Foster, 1990, De Roo, 1993. Each objective corresponds to a level of precision for the outputs which, in turn, is associated to appropriate level of both process description and input resolution. In the present study, where the model is to be used for relative comparisons between alternative scenarios in the context of arable land on loess soils, the main requirements that have to be met are: spatial description of the catchment (location of conservation practices); consideration of the effect of sealing and crusting as driving factors for infiltration; consideration of agricultural induced features and availability of the input data.

Empirically based models such as the USLE have been widely used in the past (Foster, 1990). The use of these multiplication-of-factors type models is accepted for prediction of average soil losses at the field scale (Huang, 1995), but their ability to integrate different processes and connectivities, which emerge as dominant when moving from the plot scale to the catchment scale, is uncertain (Imeson and Kirkby, 1996). Another trend in modelling is represented by process-based models. They stand for current state of understanding of the physics of the process. Because of the complexity of the mechanisms involved in water erosion, even in these more conceptual approaches, empiricism often remains since most of the factors are either USLE-based (De Roo et al., 1996) or deduced from statistical analysis. There has been attempts to model infiltration in crusted soils, as reviewed by Le Bissonnais et al., 1998a, Le Bissonnais et al., 1998b, in process-based models. These efforts can be divided into three general categories: (1) Horton-type regression; (2) the Richards equation for layered soils; and (3) several adaptations of the Green–Ampt equation. Most of the equations presented in the literature have performed well within the limitations of their experimental conditions. These limitations become more restrictive when attempting to pass from the laboratory into the field. Another aspect with this type of model is the need for a large number of input data that are not always readily available. More generally, one of the main impediment of effective mathematical erosion modelling is due to the complexity of the processes involved: the more precisely one wants to reflect the reality, the more mechanisms have to be integrated often without having a thorough understanding of their interaction. Moreover, the variance due to spatial variability may be much larger than the variance related to the approximations incurred by a model, rendering the use of more accurate models irrelevant (Bergsma et al., 1996). De Roo (1998), after having tested the coupling of process-based model with GIS, arrived at the conclusion that there is a need for much simpler coupled GIS erosion models simulating only the dominant processes operating in the catchment. Accordingly, Harris and Boardman, (1990) considered an alternative approach to those mentioned above: the expert-system approach. It can handle databases containing qualitative information that are analysed on a basis of expert knowledge. Two preliminary conditions have to be considered: firstly, this approach operates most effectively within a local domain; secondly, there exists an important assumption that there is some relationship between the different parameters of the database, which is non-random. This approach offers the possibility to make database or expert knowledge directly applicable to field conditions.

In our modelling approach, to avoid over-parameterization and the associated uncertainties, we have focused on the more integrating parameters. To combine these parameters, we have defined an expert-based approach by developing decision rules in the forms of matching tables. Hence, experimental results are directly incorporated and can easily be updated although, unlike in most expert-based system, no procedure allows to implement new observations that do not describe the given context where the dominant parameters have been identified. Finally, to allow detailed spatial description of the catchment and better visualisation of the output, the model is integrated in a GIS. The objective of this paper is to describe the elaboration of the model called Sealing and Transfer by Runoff and Erosion related to Agricultural Management (STREAM).

Section snippets

Parametrizing at the local scale

The first step of the modelling consisted of a characterisation of the main parameters influencing runoff and infiltration in the studied context. This characterisation is based on the synthesis of laboratory and field experiments carried out in the Pays de Caux region:

  • microscale studies in the laboratory under simulated rainfall Le Bissonnais et al., 1995, Fox and Le Bissonnais, 1998 and in the field with the saturated patch method (Boiffin and Monnier, 1986);

  • plot scale (1 and 20 m2) studies

Upscaling

The second step of the modelling consists of incorporating the infiltration/runoff balance to the catchment scale (Fig. 3). First, it is necessary to build the runoff circulation network that takes into account the specific processes at this scale which have been handled as: (1) problems of relative positioning of identified homogeneous surfaces and their connectivity, and (2) problems of absolute positioning of identified homogeneous surfaces. Areas producing runoff would not have the same

Results and discussion

The model was implemented with the Blosseville catchment (89.7 ha) located in the northwestern part of the Paris Basin (Pays de Caux). The area is covered by silt loam soils developed on the loess quaternary deposit and containing at least 60% silt in the surface horizons. These soils are very sensitive to soil crusting because of low clay content (13–17%) and low organic matter content (1–2%). The topography is relatively smooth with slope gradients ranging between 1% and 4% on the plateau and

Conclusion

Water erosion is at the centre of a wide range of environmental, economical and social issues. The impacts of an erosional event cannot be fully evaluated until after the event has occurred, when it is too late for planning decisions. Preventive action proves to be more judicious but inevitably goes through an estimation of the risk, of the triggering or aggrieving circumstances. Therefore, one should ask to what extent an assumption of the reality through its mathematical or theoretical

Acknowledgements

This work was partially financed by the EU-CEO Research Project FLOODGEN (ENV. 4 CT96-0368). The authors are also grateful to Tom Wassenaar for his English language corrections of the manuscript.

References (43)

  • A Bergsma et al.

    Terminology for Soil Erosion and Conservation

    (1996)
  • J Boiffin

    Stages and time dependency of soil crusting in situ

  • J Boiffin et al.

    Infiltration rate as affected by soil surface crusting caused by rainfall

  • J Boiffin et al.

    Influence des systèmes de culture sur les risques d'érosion par ruissellement concentré: I. Analyse des conditions de déclenchement de l'érosion

    Agronomie

    (1988)
  • J.M Bradford et al.

    Interrill soil erosion processes: relationship of splash detachment to soil properties

    Soil Science Society of America Journal

    (1987)
  • L.M Bresson et al.

    Soil surface crust formation: contribution of micromorphology

  • A Casenave et al.

    Les états de surface de la zone sahélienne: influence sur l'infiltration

    (1989)
  • V Chaplot et al.

    Field measurements of interrill erosion under different slopes and plot sizes

    Earth Surface Processes Landforms

    (1999)
  • D.G DeCoursey

    Mathematical models for nonpoint water pollution control

    Journal of Soil and Water Conservation

    (1985)
  • De Roo, A.P.J., 1993. Modelling surface runoff and soil erosion in catchment using Geographical Information Systems:...
  • A.P.J De Roo

    Modelling runoff and sediment transport in catchment using GIS

    Hydrological Processes

    (1998)
  • Cited by (171)

    • Dynamic parameterization of soil surface characteristics for hydrological models in agricultural catchments

      2022, Catena
      Citation Excerpt :

      Crop operations are considered, as they may modify both crop cover (e.g. harvesting) and surface crusting and roughness through tillage operations (e.g. ploughing). Finally, conversion of soil surface states into hydrodynamic properties was performed using the procedure described in the STREAM model (Cerdan et al., 2002a). The Manning coefficient was derived from the experimental data proposed for various crop types by Gilley et al. (1991) and Morgan (2005).

    View all citing articles on Scopus
    View full text