Understanding the process of valley bottom gully formation and development to reduce reservoir sedimentation in the highlands of North-western Ethiopia

Summary

Problem

Gully erosion is one of the most severe land degradation processes in Ethiopia's highlands, posing a significant threat to people and the environment. It destroys the environment and the economy by disrupting various soil and land functions. Gully conservation efforts did better in semi-arid regions than in humid ones. Due to a lack of understanding of the erosion process, which results in a poor technical quality of soil and water conservation measures, and a lack of integrated approach, gully restoration efforts in sub-humid Ethiopia were unsuccessful. Understanding the soil and hydrological processes that control gully erosion and identifying gully hotspot areas is critical for successfully implementing gully reclamation measures.

Method

In the 2017 and 2018 Ethiopian rainy seasons, the geomorphic adjustment of two gullies, one eroding on Nitisol and the other on Vertisol, representing the dominant valley bottom soils in the Minzir catchment, was studied. Both gullies cut through communal grazing land in the valley bottom and are actively eroding. This research aims to understand the VBG erosion process using field data and modeling approaches. A literature review across different agroclimatic conditions supports the identification of parameters for further field investigation. Soil hydrological, meteorological, and gully morphological parameters were monitored to assess the impact of various interacting factors on gully erosion. Secondary data such as digital elevation model (DEM), soil, and land cover were used to locate gully erosion hotspots and rank factors based on their importance to gully erosion. Finally, we used a Python programming tool to develop a landscape gully erosion model called Landscape Evolution by Gully Erosion.

Findings

In Chapter 2, a literature review identified the causes and controlling factors of valley bottom gullies. It also found deficiencies in existing rehabilitation measures. Valley bottom gullies appear as incisions in the lower periodically saturated landscape where surface and subsurface flow concentrates. From the literature review, we found the following general trends: watershed characteristics determine the location of valley bottom gullies; an increase in water transported from the watershed initiates the formation of gullies; the rate of change of the valley bottom gullies, once formed, depends on the amount of rainfall and the soil and bedrock properties. In a humid climate, subsurface flow greatly enhances bank slippage and the advancement of gully heads. Valley bottom gully reclamation is generally effective in arid and semi-arid areas with a limited subsurface flow and deep groundwater tables. Similar remedial actions are ineffective in humid regions because they are not designed to account for the effect of subsurface flows.

In Chapter 3, we looked at the impact of soil hydrological and morphological characteristics on the formation of valley bottom gullies on Vertisol and Nitisol. Despite having a larger drainage area and steeper and taller gully walls, the Nitisol gully was more stable than the Vertisol, which had shallower and less steep sidewalls. In the Nitisol, soil pipes evacuate excess water from the gully bank to the gully channel, explaining the relatively deeper groundwater and lower soil moisture content in this soil than in the Vertisol. The drainage provided by soil pipes in the Nitisol prevented a sizeable saturated zone in the unconsolidated soil and contributed to the smaller gully head erosion rate. The impact of soil pipes and cracks in the Vertisol was the inverse of what was observed in the Nitisol (i.e., promoting water recharge into the soil). Furthermore, the discontinuous nature of the crack may promote pressure build-up, which can result in gully slumping. The Bank Stability and Toe Erosion Model (BSTEM) model was used to analyze gully bank stability, and it was found that groundwater and soil cracks affect gully head stabilities. Tension crack had a significant impact on gully head stability compared to GWT. For Vertisol, in addition to structurally stabilizing the gully head and sidewalls, rehabilitation measures should safely remove excess water from both the catchment and the gully bank to reduce bank collapses. Reduced upslope runoff and stabilization of the gully head and sidewalls may be sufficient for Nitisol.

In Chapter 3 gully erosion susceptibility map (GESM) was developed using frequency ratio (FR) and random forest (RF) models for the Minzir catchment. In addition to GESM, the RF model was used to rank gully predictor factors based on their importance to gully erosion. Among the selected fourteen gully predictor factors, soil type and groundwater table (GWT) are new variables introduced to the models that were not used in previous studies. Both models produced the best gully erosion prediction when the four most important gully erosion predictor factors were used (i.e., drainage density, elevation, land use, and groundwater table). The finding that the groundwater table is one of the most important gully predictor factors in Ethiopia is a novel and significant quantifiable finding and is critical to the design of effective watershed management plans. The findings of this study suggest that future planning and implementation of conservation measures in sub-humid regions of Ethiopia should focus on areas with higher drainage density, low-lying areas, grazing land, and shallower groundwater table, which are vulnerable to gully erosion.

In chapter 5, we developed a new numerical model called Gully Erosion by Headcut Migration (GEHM) that accounts for various environmental and climatic factors and models gully head erosion at a landscape scale. GEHM is a new python package that enables gully head erosion to be modeled along with other earth surface processes. The model uses terrainbento and landlab model components in combination with the gully head migration model. At user-specified time intervals, the model will save outputs such as topographic elevation, aquifer thickness, soil saturation, surface water discharge, sediment transport rate, and gully head location. The sensitivity analysis for selected gully erosion parameters showed that the model is sensitive to changes in groundwater level, hydraulic conductivity, drainable porosity, water erodibility, and head-cut height. Model outputs such as gully head retreat, cumulative soil erosion, aquifer thickness, and groundwater storage for a given input parameter change showed a similar trend with previous studies. Finally, the GHEM model can model changes in groundwater storage caused by soil losses from gully heads and hillslopes.