APPLICATION OF EXISTING MODELS FOR THE PREDICTION OF GULLY GROWTH

Abstract

The erosion problem at Efon-Alaaye in Ekiti State, Nigeria, was investigated with a view to determining the type(s) of gullies and major factors responsible reconnaissance, for accelerated erosion in the area. During the 8 major gullies were identified, 4 of this, which were observed to be most active were selected for the study. The rainfall data of the area over the past 10 years was obtained. The drainage determining area for each of the gullies was surveyed with a view to the area, length of the gully head to water shed divide and the slope. Similarly, a total of 12 samples of soils with 3 samples each from the eroding layer of each of gully were collected and analyzed in order to determine the strength, physical and chemical properties of the soils. The width of each of-the gullies was measured at 0.5m vertically and 2.0m horizontally at the end-of years 2000, 2001, 2002 rainy seasons respectively in order to determine the magnitude of growth for each of the gullies. Using Beer & Johnson's advancement, and Thompson equations of growth and head the annual growth of each of the gullies was predicted. The results obtained show that Sheet wash, Rill and Gully types or erosion are plaguing the area with gully erosion being the most severe. It was also observed that the area experiences large mounts of rainfall with the average annual rainfall being as high as 1471 mm and intensities of 80mm/hr for 30 minutes duration for a return period of 2 years being recorded. The results also showed the slope of the area to be generally steep with slope varying between 9% and 22%. The soils were observed to be generally sandy and loose with clay and organic carbon contents varying between 1% and 6 % and 0.2 % and 0.6 % respectively. The stability of the gully sides show factors of safety varying between 0.53 and 1.67 with the lowest value being observed at Igbo-Egun. Simple prediction models were established for both annual surface growth and head advancement for the area. The models reveal an average value of the coefficient of surface growth Ka and head advancement Kb of 0.559 and 8x I0-5 respectively