For a small catchment in the north-western part of Iran, the Silsoe model (Morgan et al., 1982) is used to calculate the sediment yield from slopes. In the Silsoe model, the erosion process is separated in a detachment and transport component. Erosion from the fields can be either transport- or detachment-limited. A detachment map of the area, depending on e.g. rainfall, soil and vegetation cover has to be compared to the transport map which is a function of land cover, overland flow and slope. In the case study data layers have to be prepared for several input parameters in the model. The Terrain Mapping Unit approach (TMU approach) has been used to map various physiographic units.

Description of the catchment area

The catchment is relatively small and located south of Zanjan city in the north-western part of Iran. The altitude ranges from 1800 to 2300 m. The average annual rainfall of 211 mm falls mainly in the months April and May and in the months October and November. In the winter there is a snow cover. The geomorphology consists essentially of denudational and structural hills with glacis which are now under dissection in the upper part of the catchment valley. In the downstream part of the catchment the glacis are small or absent because of the new cycle of erosion, which has led to the formation of active gullies. The incision causes a high sediment delivery ratio in most of the subcatchments. Fresh deposition was observed only at few places along the contact of the steep slopes and the glacis, south and southwest of the village. The soils are related to the lithology and generally show a texture varying from sandy loam to sandy clay or clay. Depth and topsoil conditions are highly variable, while on some of the steeper slopes little erosion takes place either because of high infiltration or stony pavement.

The following soil data was obtained from field and laboratory measurements:

The land cover consists mainly of strongly grazed range land vegetation consisting of herbs, low shrub and grassland with different canopy densities. On the moderately to fairly steeply sloping lands the old agricultural fields have been abandoned. On the glacis the main cover is wheat.

Theoretical background; the SILSOE model

In this study, the detachment erosion is compared with the transport capacity of the overland flow (Morgan et al., 1982). The attractiveness of the method besides its simplicity lies in the recognition that erosion can be either transport- or detachment-limited.

Detachment

Rainfall detachment is a function of the annual kinetic energy, the soil detachability index and the percentage rainfall interception. In this case study Rainfall detachment (D) is calculated as:

D = Kd*{KE*e ^(-a*INT)}^b

Annual kinetic energy can be calculated from autographic rain gauge charts using equations or can alternatively be estimated from the rainfall data using locally derived
empirical equations.

KE = R*(11.9+8.7*log₁₀I)

The soil detachability index needs local adjustments but in this excercise the values based on texture classes as given in the catchment description are used. For the interception by low vegetation, estimates from literature may be used.

Transport

The Transport capacity in the Silsoe model is a function of the cover factor, the volume of overland flow and slope. Transport capacity can be computed as:

T = C*Q²*sin(S)

The volume of overland flow is determined from the annual rainfall and from the ratio of soil moisture storage and the amount of rainfall per rainy day.

Q = R*e^-Rc/Ro

The soil moisture storage is given by the moisture content at field capacity, the bulk density of the top soil, the top soil rooting depth, and the ratio of actual and potential evapotranspiration.

Rc = 1000*Rd*Ms*Bd*(Ea/Ep)^0.5

The crop factor (C) has been determined from the subfactors canopy cover, basal cover and stoniness and by considering the cropping calendar and the dynamics of the vegetation of the pasture lands. Also stoniness and sealing of the soil may be included in the parametrization.

Procedure and analysis

Most data needed in the Silsoe model is related to physiographic units and land use. For each location in the catchment (represented by pixels), the detachment map and the transport map have to be calculated separately. After that, their output values have to be compared to come to the final erosion rate map.

First, you need to create a soil table with the soil data as given in the theoretical part of this exercise.

To calculate the detachment map, this soil table has to be linked to the map that contains the terrain mapping units. This can be done by joining the soil table via the detachment factor column with the TMU table. After this, an attribute map based on the detachment factor can be created. Finally, the detachment map can be generated by some map calculations.

To calculate the transport map with mapcalc, several input maps must be prepared:

• a slope map.
  The slope map is calculated by creating a DEM of the catchment area with contour interpolation, filtering this DEM in x and y direction and performing the mapcalc statement Sloperad= ATAN((HYP(dx,dy))/pixsize(DEM)).
• a crop factor map.
  The crop factor map is calculated by creating an attribute map using the crop factor column of the TMU table.
• a soil moisture storage map.
  Before the soil moisture storage map can be computed, information about moisture content at field capacity and bulk density should be added to the TMU table. This is done by joining the information of the soil table with the TMU table. With a tabcalc expression a new column with soil moisture storage values is calculated. From this column an attribute map is created.
• a overland flow map.
  The soil moisture storage map together with several other values derived from field survey, hydrological data records or values taken from literature serve as input for the overland flow map. The overland flow map is calculated with a mapcalc statement.

When both the detachment map and the transport map are correct, the erosion map can be calculated. The erosion is either detachment-limited or transport-limited.

The detachment map is compared with the transport map and for each pixel in the catchment the smallest value is selected. This minimum value is taken as the modelled erosion rate. Five erosion classes are defined using histogram equalization, and by slicing, the map is reclassified into the erosion classes.

Discussion of results and conclusions

Actual erosion rates have been measured at seven places in the form of accumulation behind some water retention structures. They vary from 8 to 400 tons/km²/year. On the whole, the results using Morgan's approach work reasonably well, except for one observation. The attributes of the terrain mapping units have also been used for the determination of the spatial model parameters, and this is reflected in the patterns of the resulting maps. Hence they are not independent.

The results of the approach by Morgan et al. (1982) has a good similarity with the field survey map, except in the units north-west of the village (in the center of the catchment).

The advantage of this approach is that erosion assessment uses functions to obtain quantitative estimates, but the functions for erosion by overland flow on fields should be tested for their use in small catchments. The flow component in the model is not very realistic since there is no transfer of flow downslope to other pixels. Incorporation of flow accumulation downslope will probably result in a more realistic approach in terms of the effects of depth and velocity of flow on the erosion.

References

• Meijerink, A.M.J., de Brouwer, H.A.M., Mannaerts, C.M and Valenzuela, C. (1994). Introduction to the use of geographic information systems for practical hydrology. UNESCO, Div. of Water Sciences. ITC Publ. no.23, 243 pp.
• Meijerink, A.M.J. (1988a). Data acquisition and data capture through terrain mapping units. ITC Journal 1988-1, 23-43.
• Meijerink, A.M.J. (1988b). Modelling in the land and water domain with a versatile GIS, ILWIS; experiences from a large tropical catchment. In: J. Bouma and A.K. Bregt (eds.), Land qualities in space and time. Proc. Symp. Int. Soc. of Soil Science, Wangeningen. The Netherlands, pp. 73-87.
• Morgan, R.P.C., Hatch, T and Wan Suleiman, Wan Harun. (1982). A simple procedure for assessing soil erosion risk; a case study for Malaysia. Zeitschrift fur Geomorph. N.F. Suppl-Bd. 44: 69-89.

For more information on this case study, contact:

A.M. van Lieshout
Department of Water Resources,
International Institute for Geo-Information Science and Earth Observation (ITC),
P.O. Box 6, 7500 AA Enschede, The Netherlands.
Tel: +31 53 4 874 306, Fax: +31 53 4 874 336, e-mail: lieshout@itc.nl