Elsevier

CATENA

Volume 75, Issue 1, 15 September 2008, Pages 38-44
CATENA

Magnitude of soil erosion on the northern slope of the Uluguru Mountains, Tanzania: Interrill and rill erosion

https://doi.org/10.1016/j.catena.2008.04.007Get rights and content

Abstract

The magnitude of interrill and rill erosion was determined on the northern slopes of the Uluguru Mountains, Tanzania which is representative for larger areas of East African Arch Mountains, where population pressure is high and land degradation is severe. The aim of the study was to develop a database to support soil conservation in the area. The study was done on two distinct geomorphic units with respect to altitude and hence rainfall distribution pattern: mountain ridges with an altitude ranging from 1000 to 1500 masl and mean annual rainfall of 2300 mm and mountain foothills whose altitude and mean annual rainfall are 550 to 900 masl and 900 mm, respectively. Total soil loss was measured on 36 individual bounded plots measuring 1.2 m × 20 m using Gerlarch troughs on each day with rain from July 2000 to June 2001. The plots were located on six different geopedologic units, nine on mountain ridges and the rest on the mountain foothills. The slope gradient on the terrain ranged from 30% to 70%. The plots were put under maize cultivation as the main crop. Soil loss through rill erosion was estimated by volumetric measurements of rills on each soil erosion plot. The soil loss due to interrill erosion was obtained by subtracting soil loss through rill erosion from the total soil loss measured in the Gerlarch troughs. The results indicate that soil loss due to both interrill and rill erosion was very high with mean soil loss of 69 and 163 t/ha/year, respectively. Rill erosion accounted for about 58% of the total soil loss while interrill erosion contributed to the remaining 42%. Both interrill and rill erosion were higher in the mountain ridges with mean soil loss of 88 t/ha/year and 210 t/ha/year compared to 49 and 116 t/ha/year in the mountain foothills, respectively. Rill erosion was significantly higher (P  0.001) in all geopedologic units with slope gradient above 40% (mean soil loss ranged between 91 and 258 t/ha/year) compared to interrill erosion with mean soil loss varying from 41 to 115 t/ha/year. In geopedologic units with slope gradient above 60% both interrill and rill erosion were highly active while in geopedologic units with slope gradient below 40% the two processes were less active. The results demonstrate that rill erosion is more important than interrill erosion in the study area particularly where the slope gradient exceeds 40%. The results further show that the major part of the studied area has moderate interrill erosion (10–50 t/ha/year) and severe to very severe (> 100 t/ha/year) rill erosion. This study clarifies the magnitude of interrill and rill erosion which is important for designing soil conservation on agricultural fields.

Introduction

Highlands of East Africa suffer severely from soil degradation since the deforestation of the natural mountain forests and the cultivation of large areas. High amounts of soil losses due to severe soil erosion are reported from arable lands in the mountainous areas of East Africa (Oldeman et al., 1990). In these areas, past studies showed that rates of soil loss by combined processes of interrill and rill erosion are very high, i.e. in excess of 50 t/ha/year which exceeds the tolerable values generally recommended to be 10 to 12 t/ha/year (Milliman and Meade, 1983). Although interrill and rill erosion are recognised to be the most important forms of soil erosion in the mountainous areas of East Africa, their relative contribution to overall soil loss is not known due to lack of such detailed quantitative information.

In the mountainous areas of Tanzania, soil degradation caused by water erosion is rampant and the magnitude of soil loss is increasing at an alarming rate (Lundgren and Rapp, 1974, Mboya et al., 1998, Westerberg and Christiansson, 1999). Soil loss rates recorded in different mountainous areas of Tanzania (Temple and Murray-Rust, 1972, Rapp et al., 1973) exceed the tolerable values generally assumed to be 0.02 to 1.0 mm/year compared to the mean rate of soil formation of 0.01 to 0.02 mm/year (Morgan, 1995). In the Usambara Mountains soil erosion is estimated to vary from 72 t/ha/year to 120 t/ha/year (Lundgren, 1980, Pfeifer, 1990) while a soil loss of 28 to 72 t/ha/year was observed in the arable lands on the slopes of Mount Kilimanjaro (Temple, 1972). Based on studies by Rapp et al. (1973) very high amounts of soil loss were reported from the arable lands of Morogoro catchment in the Uluguru Mountains. The calculated average sediment yield in the year 1966 to 1970 was estimated to be 312 t/ha/year. Sheet wash (interrill erosion) from the cultivated land constituting about 10% of the catchment is thought to supply the main flow of the sediments. Sheet wash (interrill erosion) measurements on erosion plots from cultivated land at Mfumbwe in the Uluguru Mountains showed a soil loss of 336 t/ha/year (Temple and Murray-Rust, 1972).

In the Uluguru Mountains, farmers tend to pay more attention to landslides neglecting other forms of soil erosion (Temple and Rapp, 1972, Kilasara and Rutatora, 1993, Lulandala et al., 1993, Westerberg and Christiansson, 1999, Verbesselt and Mertens, 2000). However the erosion process in these areas is complex and the importance of other forms of soil erosion leading to denudation has widely been reported. Accordingly, interrill erosion (Temple and Murray-Rust, 1972), rill and gully erosion (Lundgren, 1978, Sorensen and Kaaya, 1998) and episodic landslides (Westerberg, 1999, Westerberg and Christiansson, 1999) are extremely important forms of soil erosion in the Uluguru Mountains. The alarming rate of soil erosion in the Uluguru Mountains and other mountainous areas in East Africa has raised a lot of ecological, environmental and economic concerns. However, efforts to arrest soil erosion in these areas have progressed very slowly due to lack of adequate data and link between specific soil erosion processes and their control measures. The way forward to combat soil erosion in the Highlands of East Africa has to be supported with information on the contribution of the various processes of soil erosion while taking into account both their spatial distribution and the degree of severity. This kind of information will assist in developing appropriate soil conservation strategies in the Uluguru Mountains, Tanzania and in similar areas in East Africa.

The objective of this study was to determine the magnitude and severity of interrill and rill erosion in different geomorphic units on the northern slopes of the Uluguru Mountains, Tanzania with the aim to develop a database to support soil conservation in the area and in similar environment of the East African Highlands.

Section snippets

Study area

The study area is located on the northern slopes of the Uluguru Mountains between 350,295E and 354,368E and 9,237,500N and 92,436,97N UTM coordinates (Fig. 1). The climate of the area is classified as sub-humid tropical savannah of the low latitude environment (Sharma, 1987). The mean annual rainfall varies with altitude, from 900 mm at around 550 masl to 2300 mm at 1500 masl. It is distributed into 2 distinct periods, a long rainy season (masika) which lasts from March to May and short rains (

Materials and methods

The rate of interrill and rill erosion was determined on two distinct geomorphic units: mountain ridges and mountain foothills. The selection of these units was mainly based on geopedologic characteristics presented in Table 1.

Total soil loss was measured on 36 individual bounded plots each measuring approximately 1.2 m × 20 m (Fig. 3) and equipped with Gerlarch troughs (Sutherland and Bryan, 1989, Prasuhn, 1992) each with a capacity of 120 l. The erosion plots were located on six different

Magnitude of interrill and rill erosion in the study area

The mean soil loss due to interrill and rill erosion in the study area is presented in Table 2. Interrill and rill erosion accounted for 69 and 163 t/ha/year of soil loss, respectively. Rill erosion was therefore more aggressive, accounting for an average of 70% of the total soil loss while interrill erosion contributed to the remaining 30%. These results agree with many studies (Zachar, 1982, Govers and Poesen, 1988, Herweg, 1992, Smolska, 1999) which have reported that on fields where

Conclusions

Interrill and rill erosion rates in the northern slopes of the Uluguru Mountains vary spatially along the landscape both in terms of the type of process and the degree of severity. The rates of interrill and rill erosion observed in this study are much higher when compared to total soil loss by interrill and rill erosion measurements seen from previously reported rates. The results of this study demonstrate that interrill and rill erosion are among the soil erosion processes highly active in

Acknowledgements

The authors wish to express their sincere appreciation to the Soil and Water Conservation on the Uluguru Mountains Research Project under SUA–Flemish Inter University Council (SUA–VLIR) Collaboration Programme for the financial assistance towards this work.

References (44)

  • KimaroD.N. et al.

    Characteristics and management of soils located on specific landform units in the northern slopes of Uluguru Mountains, Tanzania

  • LalR.

    Soil erosion in the tropics

  • LulandalaL.L. et al.

    Socio-economic Survey of Falkland and Magadu Villages for an Appropriate Soil Conservation System: a Preliminary Report of the UMISCP

    (1993)
  • LundgrenL.

    Studies of soil and vegetation development on fresh landslide scars in the Mgeta Valley, Western Uluguru Mountains, Tanzania

    Geografiska Annaler

    (1978)
  • LundgrenL.

    Comparison of surface runoff and soil loss from runoff plots in forest and small-scale agriculture in the west Usambara Mountains, Tanzania

    Geografisfiska Annalar

    (1980)
  • LundgrenL. et al.

    A complex landslide with destructive effects on the water supply of Morogoro Town, Tanzania

    Geografiska Annaler

    (1974)
  • MagundaM.K. et al.

    Soil loss and runoff from agricultural land use systems in the Sango Bay micro-catchment of the Lake Victoria

  • MboyaT.O. et al.

    Assessment of soil erosion using remotely sensed and ancillary data

  • McCoolD.K. et al.

    Adapting the Universal Soil Loss Equation to the Pacific Northwest

    Transactions of the American Society of Agricultural Engineers

    (1982)
  • McCoolD.K.J. et al.

    Revised slope steepness factor for the Universal Soil Loss Equation

    Transactions of the American Society of Agricultural Engineers

    (1987)
  • MeyerL. et al.

    Experimental approaches for quantifying the effect of soil erosion on productivity

  • MillimanJ.D. et al.

    World-wide delivery of river sediment to the oceans

    Journal of Geology

    (1983)
  • Cited by (0)

    View full text