A GIS-based assessment of the potential soil erosion and flood hazard zones in Ekiti State, Southwestern Nigeria using integrated RUSLE and HAND models

Highlights

• Flood and soil loss are major natural hazards in Ekiti State, Southwest Nigeria.

• Soil loss risk zones are highly related to the degree and inclination of slope.

• Soil erosion processes play a critical role in the occurrence of flood hazards.

• Land use and topography significantly affected soil loss and flood hazards.

Abstract

Soil loss estimation and flood hazard mapping cannot be overemphasized due to their environmental, economic and societal concern. Thus, the main objective of this study was to assess the potential soil erosion and flood hazards zones using Revised Universal Soil Loss Equation (RUSLE) and Hand Above Nearest Drainage (HAND) models, respectively for appropriate conservation and prevention measures in Ekiti State, Southwestern Nigeria. The result reveals annual soil erosion ranges from 0 to 889 t ha⁻¹ year⁻¹. The estimated total annual soil loss in the state was around 1.5 million tonnes, of this, 193223.3 tonnes which covers 7508 ha was lost at a rate much greater than the tolerable soil loss rate. Soil erosion vulnerability mapping was done using six (6) categories of soil loss severity from slight to very severe. Soil erosion rates varied from 0.21 t ha⁻¹ year⁻¹ in forests to 1.69 t ha⁻¹ year⁻¹ under bare soils. The very steep slope category had the highest soil erosion rate of 15.07 t ha⁻¹ year⁻¹ while the gentle slope regions had the least soil erosion rate of 0.39 t ha⁻¹ year⁻¹. About 23% of the area which covers 115847.58 ha was prone to high flood hazard whereas 15.69% (80932.30 ha) was moderately susceptible to flood risk. Approximately 15% of the area was susceptible to low flood hazard zone while 44.90% were vulnerable to very low flood occurrence. LULC has significant effect on the spatial pattern of soil loss in the study area. Potential soil loss risk zones were highly related to the degree and inclination of slope. The reported results can serve as preliminary information to determine erosion and flood hotspots in the study area and as input for policy decision for disasters prevention and conservation measures.

Introduction

Globally, soil loss is one of the most serious environmental threat in recent time (Devatha et al., 2015). Soil erosion by water has been acknowledged as a serious concern due to its impacts on the environment and the society at large (Angima et al., 2003, Singh and Panda, 2017). The severity of soil erosion is a product of various factors such as topography, soil features, severity of rainfall, runoff, cultivation mechanism and land cover (Nyssen et al., 2004, Stumpf et al., 2016, Kayet et al., 2018). Erosion is mainly caused by anthropogenic activities aided by geomorphologic processes. Land degradation from anthropogenic activities have been reported to have affected 1964.4 Mha of land globally, out which soil erosion by water accounted for degradation of 1903 Mha land (UNEP, 1997). Soil loss affects natural resources and agricultural production through the removal of fertile top soil with significant impacts on crop production (Baptista et al., 2015) and quality of the ecosystems (Fujaco et al., 2016). It has also been reported to cause dam and reservoir sedimentation (Balthazar et al., 2013), land wastage (Kayet et al., 2018), and economic loss (Zerihun et al., 2018).

Flood is an abnormal situation of water flow outside the original borders of a defined water body (Rincón et al., 2018). Floods results from complex hydrological, geological and geomorphological conditions with substantial social, economic and environmental damages (Mukherjee and Singh, 2019, Skilodimou et al., 2019). The occurrence of flood worldwide over the last forty (40) years has increased significantly (Vanolya and Jelokhani-Niaraki, 2019). About 31% of economic losses from natural disasters are caused by floods making it the most expensive natural hazards globally (Yahaya, 2008). According to Aderogba (2012), floods are the most common and serious natural hazards in Nigeria, with significant negative impacts on life and property. Since 1985, a total number of 1 803 deaths were reported from flood occurrences (EM-DAT-CRED). Similarly, significant floods have been experienced in Ekiti State in recent times. The magnitude of these flood events has not been seen in the past (Afolabi, 2019: The Guardian). Nearly every community in the state had been affected by flood and erosion because of the topography of the Ekiti region and had caused a lot of hardship to the people. The increase in the amount, seasonality and event-size distribution of rainfall (torrential rainfall) coupled with poor drainage networks and clogging of drains by locals have aggravated flood disasters in the State. Deforestation and urbanization are also important factors contributing to flood hazards. These have wreaked havoc on residential houses, highways, agricultural lands and severely damaged properties worth millions of naira. Loss of lives and loss of various ecosystem services have also been recorded in the state as consequence of the ravaging flood scenarios.

Ekiti State is experiencing significant changes in land use Land cover (LULC) and, as a result, an enormous rate of soil erosion and flood disasters due to rapid increase in population, unmanaged urbanization, agricultural expansion and energy poverty, which places demands on natural resources. Also prevalent in the State is overgrazing and deforestation due to expanding agricultural activities which appears to be the leading factor for land transformations in the state (Olorunfemi et al., 2018). Furthermore, the geomorphological attributes of the State, the rugged hilly terrain and undulating feature with a typical landscape that comprises of old plains, make it more susceptible to water erosion and flooding. The situation is now exacerbated by improper land use (such as cultivation of steep lands), lack of good management practices and non-adoption of soil conservation and flood control techniques, which are the major drivers of soil erosion and flood incidences. This has resulted in increased sedimentation, land and ecological degradation (Uddin et al., 2016) in the region. These changes if not checkmated, can lead to extensive and accelerated soil erosion.

The estimation of soil erosion can be carried out using empirical or physically-based models (Roo et al., 1993, Uddin et al., 2016) which have come up in the last few decades. These includes Chemical Runoff and Erosion from Agricultural Management Systems (CREAMS), Agricultural Nonpoint Source model (AGNPS), Universal Soil Loss Equation (USLE), Revised Universal Soil Loss Equation (RUSLE), Modified Universal Soil Loss Equation (MUSLE) and Water Erosion Prediction Project (WEPP). Among these models, a large number of studies have utilized the USLE by Wischmeier and Smith (1978) and the RUSLE by Renard et al. (1997) in estimating soil loss (Onori et al., 2006, Chen et al., 2011, Zerihun et al., 2018). This is due to their simplicity, minimal data requirements, applicability in various topographical relief and regions (Perovic et al., 2013, Dutta et al., 2015) and the successful utilization of predicted results in effecting conservation and mitigation measures (Terranova et al., 2009, Vijith et al., 2017). For a long time, USLE model has been widely applied, however, the model has limitations in terms of its capability to properly predict spatially the distribution of soil erosion (Wang et al., 2009) and its applicability and effectiveness in estimating soil erosion under different conditions (Beskow et al., 2009, Demirci and Karaburun, 2012). To overcome the limitations, RUSLE model was developed for field estimation of soil loss in different regions (Renard et al., 1997, Zerihun et al., 2018). The RUSLE model is capable of identying the likelihood of soil erosion on a pixel cell-by-cell method (Shinde et al., 2010) and is therefore widely accepted in estimating the spatial distribution of soil loss at different spatial scales in different regions (Renard et al., 1997, Uddin et al., 2016). More so, the flexibility, time and cost effectiveness and practicality in data scarce areas enhanced its worldwide acceptance and application for estimating soil erosion (Zerihun et al., 2018). The applicability of the RUSLE in every region where loss of soil by water is a major challenge has already been established by Laflen and Moldenhauer (2003). Also, RUSLE can be applied in many circumstances including steep and undulating terrain (Gelagay and Minale, 2016). The integration of Geographical Information System (GIS) and Remote Sensing (RS) techniques with RUSLE has contributed to the derivation of the main factors/variables used in computing soil loss. In addition, spatiotemporal dynamic assessment and soil erosion modeling in large areas has been enhanced at relatively low cost and with greater accuracy compared to traditional methods (Wang et al., 2003, Demirci and Karaburun, 2012, Pradeep et al., 2015). This has been confirmed by a number of studies which combined RUSLE with RS/GIS technology for soil loss modelling (Pandey et al., 2007, Adediji et al., 2010, Perovic et al., 2013, Uddin et al., 2016, Duarte et al., 2016, Zerihun et al., 2018).

A number of basin-scale soil loss assessments have been conducted in the southwestern region of Nigeria. Akinmolayan et al. (2018) estimated potential annual soil loss within Asejire watershed using RUSLE model while Makinde and Oyebanji (2018) also estimated actual erosion risk using RUSLE model in Eti-Osa local government area of Lagos State. In other parts of the country, studies have equally been conducted to assess soil loss using RUSLE model (Adediji et al., 2010, Fagbohun et al., 2016, Njoku et al., 2017, Thlakma et al., 2018). Lal (1976) estimated soil erosion rate of about 5–120 Mg ha⁻¹ yr⁻¹ in his study on alfisols in Western Nigeria making the country an erosion hotspot region. Dike et al. (2018) estimated a higher rate of soil erosion in the Urualla watershed, Imo State, Nigeria, with annual soil loss ranging from 6 to 1200 t ha⁻¹ year⁻¹. Similar soil loss assessment conducted in the northern part of Nigeria showed a potential erosion rates ranging from 0.0 to 4185.12 t ha⁻¹ year⁻¹ (Adediji et al., 2010). Despite the increasing concerns about the impacts of soil erosion and flood on agricultural productivity and sustainable development of the State, few soil loss estimations have been conducted only at watershed scale while flood hazard assessment has been neglected. Olusa et al. (2019) used RUSLE and GIS to assess the impact of soil erosion on residential areas of Efon-Alaaye Ekiti, a part of Ekiti State. Other studies in the State only utilized questionnaires to assess the impact of soil erosion on the communities (Adegboyega, 2019). However, there has been no complete assessment of the spatial dynamics of soil erosion and flood in the State. Therefore, this study attempts to assess the current state of soil erosion, in particular the spatial distribution patterns and critical erosion hotspots, and delineate potential flood hazard zones in order to provide key information for soil loss and flood prevention and mitigation strategies by policy makers in the region.

Conversely, the Hand Above Nearest Drainage (HAND) terrain model is an effective model for flood potential distribution prediction (Nobre et al., 2015). The model computes the differences in elevation between each grid cell and the elevations of the flow path connected downslope grid cell where the flow enters the channel (Nobre et al., 2011) thereby providing an objective method for flood hazard assessment. Successful application of the model in mapping hydrological hazards in the São Paulo metropolitan zone, where relative vertical proximity to rivers functioned as a proxy for flood hazards has been reported (Nobre et al., 2010). In another study, Cuartas et al. (2012) similarly utilized the model to parameterize areas of soil water saturation in a physically based, distributed hydrological model using topography data acquired through RS techniques.

Thus, this study employed RS and GIS techniques to assess the spatial distribution of soil erosion and flood in Ekiti State, Southwestern Nigeria using RUSLE and HAND models and prioritize erosion and flood hazard zones in the area for appropriate conservation and control measures.