Accelerated Erosion: Generation of Digital Elevation Models (DEMS) for Gullies in Irele Local Government Area of Ondo-State, Nigeria

EarthzineArticles, Earth Observation, Ecosystems Theme, Original, Sections, Themed Articles, Water

Cropped image of a map of Slovenia

*Adediji, A; Ibitoye, M.O. and Ekanade, O.

Department of Geography, Obafemi Awolowo University (OAU), Ile-Ife, Nigeria

*correspondent author e-mail: remiadediji2003@yahoo.co.uk

Keywords: DEM, Gully, GPS, Accelerated Erosion, Irele Local Government Area

Fig 1: Map of Irele Local Government Area showing the study settlements

Fig 1: Map of Irele Local Government Area showing the study settlements

Abstract

This study attempts to derive Digital Elevation Models (DEMs) for gullies at Ode-Irele, Lipanu, Akotogbo and Ajagba town in Irele Local Government Area (LGA) of Ondo-State, Nigeria. Gully morphometric attributes such as surface slope, catchments area, average depth and width as well as cross-sectional area and volume of material/ soil excavated were determined. The DEM for each of the study gully catchments was derived from the values of the spot heights and coordinates of each point obtained through the use of GPS and processed using Surfer 8.0 software. The values of slopes in the study gully catchments ranged from approximately 0058’27” at Lipanu to 4051’31” at Ajagba. However, the DEMs of gullies at Idogun, Ajagba and Ado Quarter displayed slopes that appeared steeper than the gully at Lipanu. Also, the slope shapes as revealed from the DEMs are dominantly convex which implied that overland flow will be generated from all sides of the slope. Therefore, the convexity of most of the gully catchments coupled with the termination of drainage channel constructed halfway as well as poor roads and drains maintenance by the community and government have led to the development of large deep gullies at Ode-Irele, Akotogbo and Ajagba in the study area. The development of gullies in the area could be minimized by encouraging planting of cover grasses in the building surroundings rather than cement paving of the surface which further makes the ground impervious.

Introduction

The mechanisms involved in soil erosion by water vary over space and time. Some of these mechanisms are rain drop splash, unconcentrated downslope wash (sheet erosion), concentrated downslope wash (rill and gully erosion), and a mixed process in which entrainment is by raindrop splash and downslope transport is by surface wash. It has also been observed that man can also influence the dynamics of each of these mechanisms and thus improper human land management can accelerate the rates of erosion which may result in the development of rills and gullies.

Gully resulting from accelerated soil erosion has been an issue of growing concern not only in the humid tropics but in many parts of the world. For instance, many published studies exist on the occurrence, assessment and monitoring of this phenomenon in an urban environment (e.g. Betts and DeRose, 1998 in North Island, New Zealand; Bocco et al, 1990 in Mexico; Ofomata, 1989, 2000 in southeastern Nigeria; Jeje, 1973, 1977, 1987 and 2005 in southwestern Nigeria). This issue is so important because the initiation of gullies, as well as the headward and lateral progression, releases large amounts of sediments and can enhance rates of overall landscape lowering and evolution (Hancock and Willgoose, 2001, 2002; Alonso et al, 2002). Also, this may result in increased sedimentation and water quality problems in many drainage basins. In addition, there has been considerable research into understanding of gully development and channelization, but much of this effort has been on evaluation of ephemeral gullies in disturbed or agricultural/urban settings, (e.g. Jeje, 1973, 2005; Patton and Schumm, 1975; Hancock and Evans, 2006).

Furthermore, Digital Elevation Models (DEMs) are increasingly used within a Geographic Information System (GIS) framework to model gully positions, features and development with GIS being also used in studies of landscape change particularly in the area of gully and terrain analysis (e.g. Prosser and Aberneathy, 1996; DeRose et al, 1998; Hancock et al, 2000; Torri and Borselli, 2003) in both small and large catchments. However, there are little or no known studies on the application of GIS to the study of gully position and development in an urban setting of southwestern Nigeria. In this regard, the present study attempts to generate DEMs for gullies in Irele Local Government Area (LGA) of Ondo State, Nigeria.

Study Area

Gullies at Ode-Irele, Lipanu, Akotogbo and Ajagba in Irele LGA constitute the study area. The Irele LGA is part of the lowland area of the southeastern part of Ondo-State. It is located between latitudes 06017’57’N and 06043’21’N and longitudes 04049’47″E and 05010’26″E (see Fig.1, Ibitoye et al. 2008).

Image of a table showing morphology of study gullies

The area falls within the Bitumen Belt of Ondo State and is predominantly populated by the Ikales of Yoruba extraction. According to the 2006 Population Census, the population of the Local Government Area was 145,166 (NPC, 2006). This population settled in agglomerations or clusters either in towns or villages. Also, as observed by Jeje (1988), the development of most of these settlements (towns and villages) occurred without any systematic planning; buildings sprang up without any recourse to physical planning particularly during evolution of traditional urbanization which took place in many of the Yoruba settlements in the late 18th century. This development has given rise to haphazard arrangements of buildings in the settlements (Ode-Irele, Lipanu, Ajagba and Akotogbo) where the gully studies are carried out.

The area falls within the tropical rainforest region and as such the climate is characterized by two broad seasons: (1) the rainy season (April-October) and (2) the dry season (November ‰ÛÒ March). The available meteorological data for Ode-Irele indicates that annual rainfall varies from 1,900mm to 2,700mm (see Agro-Climatological and Ecological Monitoring Unit, Akure).The annual average temperature ranges between 21.4° and 31.1°C and its mean annual relative humidity is about 77.1% (Iloeje, 1978).

The area has a general elevation of 45m above mean sea level, but the ground slope is imperceptibly in a north-south direction as evidenced by the flow direction of drainage systems such as the rivers Oluwa, Ufon, Mene and Salawa. Most of these rivers meander freely and are actively engaged in lateral erosion at meander bends thereby widening their channels.

According to Areola (1983), the area is underlain by terralsols that consist mainly of loams, sandy loam and in some places clay loams which are easy to cultivate but, however, suffered from excessive internal drainage and intensive leaching. Also, the original vegetation of the zone whichåÊåÊåÊ has been replaced by arable crop cultivation, rubber and palm plantations and exotic tree plantations established by forestry divisions and secondary forest regrowth (Areola, 1983).

Materials and Methods

Four settlements in Irele LGA were selected for the study of gully morphology and derivation of DEMs for the gully catchments. These settlements are Ode-Irele (the LGA headquarter), Akotogbo, Ajagba and Lipanu. Apart from Ode-Irele which was purposely selected due to its position as the administrative headquarter of the LGA, others were sampled using the table of random numbers of the list of prominent settlements in the Local Government. The erosion channels / gullies were identified in all the selected settlements, out of which five gullies were studied. Thus, for this study two gullies were selected at Ode-Irele, and one gully each at Akotogbo, Ajagba and Lipanu.

Image of Digital Elevation Map of gully catchment at Idogun Quarter, Ode ‰ÛÒ Irele
Fig 2a: Digital Elevation Map of gully catchment
at Idogun Quarter, Ode – Irele
Image of Digital Elevation Map of gully catchment at LA Primary School Area, Ode ‰ÛÒ Irele
Fig 2b: Digital Elevation Map of gully catchment
at LA Primary School Area, Ode – Irele
Image of Digital Elevation Map of gully catchment at Ajagba, Ode ‰ÛÒIrele LGA
Fig 2c: Digital Elevation Map of gully catchment
at Ajagba, Ode ‰ÛÒIrele LGA
Image of Digital Elevation Map of gully catchment at Ado Quarter, Akotogbo
Fig 2d: Digital Elevation Map of gully catchment
at Ado Quarter, Akotogbo
Image of Digital Elevation Map of gully catchment at Lipanu
Fig 2e: Digital Elevation Map of gully catchment
at Lipanu

The catchments of each gully were visually delineated based on the slope gradient and gully pattern. The points were selected along the catchment perimeter. A Global Positioning System (GPS) receiver (Garmin GPS 12) was used for determining the coordinates (northing and easting) of all turning points until the area enclosed by the gully catchment was covered. The area covered by each catchment was calculated using values of coordinates of turning points (see Ibitoye, 2006).

Also, for the generation of DEMs, the area of the study gully catchments was later divided into grids at regular interval of 50m. Along each traverse/gridline, at intervals of 50m, spot heights and their Cartesian coordinates were determined using GPS on an accuracy level of 10m. A total number of 126, 73, 69, 334 and 34 spot heights were determined for gully catchments at Ode-Irele, Ajagba, L.A. Primary School Area and Lipanu, respectively.

The data obtained from the field measurements were processed and used for the generation of contour maps and DEMs (Digital Elevation Models) of the selected gully catchments in Irele LGA. In this regard, the values of spot heights obtained from the gully sites were plotted against the coordinates of each point using the kriging analysis in Surfer 8.0 software.

Further, the gully morphometric attributes such as gully width, depth and cross-sectional area were measured using tape and leveling staff. The cross-sectional area of each of the selected gullies was determined using a formula adopted by Ofomata (2000) in southeastern Nigeria. The formula is given as:

A = wd ‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_‰Û_ (i)

Where A = cross-sectional area (m2)

w = mean width of the gully

d = mean depth of the gully

Also, the value of the cross-sectional area obtained was used to estimate the volume of soil removed by gully erosion from each of the study gully catchments. Thus, the Prismoidal formula was used in this study (see Bannister and Raymond, 1983). This method was used because it yields better accuracy than the End Area formula. The amount of sediment loss from the gully site was estimated by multiplying volume with the soil bulk density (see Jeje, 2005). The bulk density was determined from the core samples taken from the floor of each of the study gullies using a McCauley corer (core sampler) of approximately 5.5cm in diameter and 4cm in height. In all, 30 core samples were taken from the study gullies and analyzed in the laboratory for the determination of the bulk density by following the procedures outlined by Singh (1989). Sediment loss for each of the studied gullies was estimated as the product of the volume of soil loss and soil bulk density (see Jeje, 2005)

Results and Discussion

Morphometry of the Study Gullies

From the results obtained by physical assessment of the subject gully catchments, two types of gully shapes were identified. These are “V” shaped gully systems (compare “V” shaped gully systems in Holy, 1980 and Jeje, 2005). The gullies at the L.A. Primary School Area, Ajagba and Lipanu exhibited V-shaped for most of their lengths while those at Idogun, Ode-Irele and in Akotogbo displayed remarkably U-shaped for most of their lengths.

With respect to the slope gradient as shown in Table 1, the value of slopes in the study gully catchments ranged from approximately 0058’27” at Lipanu to 4051’31” at Ajagba. The slopes at Lipanu and L.A. Primary School Area are almost flat while those at Idogun, Ajagba and Ado Quarters, Akotogbo are not quite as steep.

Also, as shown by the DEMs of the study gullies coupled with the field observation, the slopes at Idogun, Ajagba and Ado Quarters in Akotogbo appeared steeper than others. Ordinarily under vegetal cover, these slope gradients should not enhance erosional processes but due to exposure to direct raindrop impact and human activities coupled with the poor soil aggregation; accelerated erosion has become pronounced particularly at Idogun Quarters in Ode-Irele, Ajagba and Akotogbo.

Further, the slope shapes as revealed by the DEMs of the study gullies are dominantly convex (see Figs. 2a, 2b, 2c, 2d and 2e). This implies that overland flow will be generated from all sides of the slope which invariably increase runoff into the gully channels. The DEM for Lipanu gully catchment exhibits relatively uniform flat terrains (see Fig. 2e) which ordinarily should enhance deposition and flooding but gullying occurred due to human activities.

Soil Characteristics and Accelerated Erosion

Soil physical characteristics and the estimated sediment loss from study gully sites are shown in Table 2. The soil textural properties of the studied gully catchments show a high proportion of sand with a mean value of 59%, while the mean values of clay and silt are 34% and 7% respectively. As observed in the field, the soil especially at the upper horizon is predominantly sandy clay and changed to sandy clay loam with depth. Also, most of the gully floors are characterized by sandy clayey regolith. With high proportion of sand at the upper layer, one would have expected these soils to be highly permeable and thus unsusceptible to soil erosion. However, because of compaction of the ground in the built-up (urban) environment, as in the case of the study gully catchments, the soils are relatively permeable and therefore susceptible to the action of sheetwash and gullying.

Table showing data on sediment from the study gullies

The values of soil loss from the study gully systems shown in Table 2 compared favorably with the total sediment loss obtained from Effon-Alaaye Gully Systems by Jeje (2005) in southwestern Nigeria and Ofomata’s (2000) findings in southeastern Nigeria. In fact, the volume of soil loss from the study gully catchments as shown in Table 2 further confirmed the findings of Harley et al (2003) in the North Island East Coast Region of New Zealand. For instance, 13,213.25m3 of soil removed from the 2nd order gully system at Ado Quarters, Akotogbo with the gully catchment area of 13.27ha (see Table 2) though higher but compared favorably with 6,900m3 of material eroded from the 7-8ha of gully system in the North Island East Region, New Zealand by Harley et al (2003).

As evident from Tables 1 and 2, the severity of gully erosion is not attributable to slope gradient. For instance, field observation showed that in the L.A. Primary School Area Gully system, the bulk of surface runoff that created gully erosion in the area was generated about 500-600m away from the gully site. In fact, the runoff originated from the main road, about 25m from the police station and routed through the built-up area via the Comprehensive Grammar School at Araromi/ Gbonye Street, down to Kanye Quarters where it developed into a pond-like feature before exit into Erioko Stream.

Image of a gully at Ado quarter, Akotogbo
Image of a gully at LA Primary school area, Ode-Irele
Image of a storey building collapsed into the gully floor at Akotogbo
Fig 3:
A) Gully at Ado quarter, Akotogbo
B) Gully at LA Primary school area, Ode-Irele
C) A storey building collapsed into the gully floor at Akotogbo

However, it is important to discuss how some human activities within the study area and specifically the study catchments normally promote the generation of surface runoff loaded with sediments. For instance, the construction of drainage channels along urban streets normally encourages the concentration of urban runoff but improper handling may enhance accelerated erosion. In this regard, as observed in the area, the termination half way of the construction work on erosion channels being constructed some years back by Ondo State Government in most of the study settlements triggers accelerated erosion (gully erosion). Runoff in the form of falls from the concretized portion of the channels onto the bare earth below gave rise to a gully from the points where the concrete gutters terminated (see Ibitoye, 2006). In fact, during the field survey, a gully head of 3.4m deep and 3.3m wide was measured where the concrete channel terminated at Idogun Quarters. However, the runoff that brought about the gully at Idogun Quarters was generated from the network of concrete drains that covers just about 8.3ha (32%) of the total gully catchment area (86.895ha).

Also, virtually all the study settlements were poorly provided with drainage systems. In fact, Akotogbo and Lipanu were hardly provided with any of these facilities. Thus, most of the surface runoff flows along the untarred roads and the unpaved drains. The drains and culverts provided were too few and inadequate to accommodate the volume of surface runoff generated during heavy rainstorms.

Another important factor responsible for the development of gullies in the area is the roofing materials. Buildings in the area are roofed with the galvanized iron sheets and without rainwater harvesting device which has severely increased the volume of runoff produced from any given rainfall. In fact, only a few houses in the area have their compounds paved with concrete, the surroundings of most of the houses are subject to severe compaction and subsequent sheet erosion (see Fig. 3a and 3b). This phenomenon was earlier observed by Okoye (1988) and Jeje (1988, 2005) in their studies on accelerated erosion in Aba and Effon-Alaaye in Nigeria, respectively. Also, it was observed that after gully has been initiated, the process of soil removal both vertically and laterally will continue at every rainy season. In most cases, lateral expansion is by basal undermining of gully sides by runoff and thus resulting to mass wasting of poorly consolidated sands and eventual collapse of mass into the gully floor (see Fig. 3c).

Conclusion

The gullies at Ode-Irele, Lipanu, Akotogbo and Ajagba in Irele LGA of Ondo-State, Nigeria were studied with a view of using the generated DEMs to explain the morphology and development of the studied gullies. The values of slopes in the study gully catchments ranged from approximately 0058’27” at Lipanu to 4051’31” at Ajagba which are generally low and gentle. Ordinarily, under vegetal cover these slope gradients cannot cause accelerated erosion but due to exposure to direct raindrop impact resulting from human activities coupled with the poor soil aggregation, gullies have become pronounced in Idogun Quarters in Ode-Irele, Ajagba and Akotogbo.

Also, the slope shapes displayed by the DEMs of the study gullies are dominantly convex which implies that overland flow will be generated from all sides of slopes. This invariably increases runoff into the gully channels. In other words, the convexity of most of the gully catchments allow runoff from all sides of slopes and the termination of drainage channels half way coupled with the poor roads and drains maintenance by the community and government have led to the development of large deep gullies at Ode-Irele, Akotogbo and Ajagba.

In the light of the above, since gullies result from the intensive scouring action of concentrated runoff, an effective measure is to reduce the velocity of the runoff and encourage infiltration. This can be achieved by encouraging the planting of cover grasses in the building surroundings and in the inter-spaces between buildings rather than cement paving and concretization of surface which make the ground more impervious.

Acknowledgements

I wish to acknowledge the comments of Prof. A. Gbadegesin and Dr. L. Ajibade on the draft paper.

REFERENCES

Alonso, C.V; Bennett, S.J. and Stein, O.R. 2002. Predicting head cut erosion and migration in concentrated flows typical of upland areas. Water Resources Research, 38 (12), 39-45.

Areola, O.O. 1978. Soil and Vegetal Resources. In Oguntoyinbo, J.S; Areola, O.O. and Filani, M. (Eds). A Geography of Nigerian Development‰Û. Heinemann Educational Book (Nig) Ltd; Ibadan, Nigeria, 105-126.

Bannister, A and Raymond, S. 1983 Surveying. Longman Scientific and Technical, England.

Betts, H.D. and DeRose, R.C. 1998 Digital elevation models as a tool for monitoring and measuring gully erosion. JAG, 1, 3, 91-100.

Bocco; G; Palucio, J. and Valenzuela, C.R. 1990. Gully erosion modeling using GlS and geomorphic knowledge” ITC Journal Netherlands, 253-261.

De Rose, R. C., L. J. Prosser and W. J. Young. 1998. “Regional patterns of erosion and sediment and nutrient transport in the Goulburn and Broken river catchments, Victoria, CSIRO Land and Water, Canberra, Technical Report 11/03

Hancock, G.R. and Evans, K.G. 2006. Gully position, characteristics and geomorphic thresholds on an undisturbed catchment in northern Australia”. Hydrological Processes, 20, 2935-2951.

Hancock, G.R. and Will goose, G.R. 2002. The use of a landscape simulator in the validation of the SIBERIA landscape evolution model: transient landforms. Earth Surface Processes and Landforms 27, 1321-1334.

Hancock, G.R and Willgrose, G.R. 2001 The use of a landscape simulator in the validation of the SIBERIA catchment evolution model declining equilibrium landforms. Water Resources Research, 37(7), 1981-1992.

Hancock, G.R; Willgoose, G.Rt. Evans, K.G; Moliere, and Saynor N.J. D.R. 2000 The Medium term erosion simulation of an abandoned mine site using the SIBERIA landscape evolution model. Australian Journal of Soil Research 38, 249-263.

Harley, D.B; Trustrum, N.A. and Ronald, C.D. 2003. Geomorphic changes in a complex gully system measured from sequential digital elevation models and implications for management. Earth Processes and Landforms, 18, 1043-1058.

Ibitoye, M.O. 2006. An assessment of accelerated soil erosion in Irele Local Government Area of Ondo-State, Nigeria, Unpublished M.Sc. Thesis, Institute of Environment and Ecology, Obafemi Awolowo University, Ile-Ife, Nigeria.

Ibitoye, M.O., Ekanade, O., Jeje, L.K., Awotoye, O.O., and Eludoyin, A.O. 2008. Characterisation of gully formed in built up area in southwest Nigeria. Journal of Geography and Regional Planning, 1(9), 164-171.

Iloeje, N 1978. Regional Geography of Nigeria. Heinemann.

Jeje, L.K. 1973. Erosion along Ife-Ondo Road in Western Nigeria. The Nigerian Geographical Journal, 16, 67-75.

Jeje, L.K. 1977. Some soil erosion factors and losses from experimental plots, in the University farm, University of Ife, Ile-Ife, Nigeria. The Nigerian Geographical Journal, 20, 59-65.

Jeje, L.K. 1987. Runoff and Soil loss from erosion plots in Ife area of southwestern Nigeria. In Gardiner, V. (Ed.) International Geomorphology. John Wiley, 459-472.

Jeje, L.K. 1988. Soil erosion characteristics, processes and extent in the lowland rainforest area of southwestern Nigeria. In Sagua, V.O; Enabor, E.E; Ofomata, G.E.K; Ologe, K.O. and Oyebande L. (Eds) Ecological Disaster in Nigeria: Soil Erosion. Federal Ministry of Science and Technology, Lagos, Nigeria, 69-83.

Jeje, L.K. 2005. Urbanization and Accelerated Erosion: The case of Effon-Alaaye in Southwestern Nigeria. Seminar Paper. Department of Geography, Obafemi Awolowo University, Ile-Ife, Nigeria.

National Population Commission (N.P.C). 2006. Nigeria Population Census, Abuja, Nigeria.

Ofomata, G.E.K. 1988. Soil erosion characteristics, processes and extent in the forest zone of Southeastern Nigeria. In Sagua, V.O; Enabor, E.E; Ofomata, G.E.K., Ologe, K.O and Oyebande L. (Eds) Ecological Disaster in Nigeria: Soil Erosion Federal Ministry of Science and Technology, Lagos, Nigeria, 50-68.

Ofomata, G.E.K. 2000. Classification of soil erosion with specific reference to Anambra State of Nigeria, Environmental Review 3, 2, 252-2551.

Okoye, T.O. 1988. Urbanization and erosion with particular reference to Aba. In Sagua, V.O; Enabor, E.E; Ofomata G.E.K; Ologe, K.O and Oyebande L. (Eds), Ecological Disaster in Nigeira: Soil Erosion. Federal Ministry of Science and Technology, Lagos, Nigeria, 265-275.

Ondo State Ministry of Environment. 2004. Agro ‰ÛÒ Climatological and Ecological Unit, Ministry of Agriculture and Water Resources, Akure.

Patton, P.C and Schumm, S.A. 1975. Gully erosion, northwestern Colorado: a threshold phenomenon.åÊ Geology 3:88-90.

Prosser, I.P. and Abernethy, B. 1996. Predicting the topographic limits to a gully network using a digital elevation and process thresholds. Water Resources Research 37(7): 2289-2298.

Prosser, I.P; Dietrich, W.E. and Stevenson, J. 1995. Flow resistance and sediment transport by concentrated overland flow in a grassland valley. Geomorphology 13, 71-86.

Prosser, L.J. and Young, W.J. 1998. Regional Patterns of Erosion and Sediment and Nutrient Transport in the Goulbhm and Broken River Catchments, Victoria”, CSIRO Land and Water, Carberra, Technical Report 11/03.

Torri, D. and Borselli, L. 2003. Equation for high rate of gully erosion, Catena 50: 449-467.

Singh, R.A. 1989. Soil physical analysis. Kalyanic Publishers, New Delhi-Ludhiana