Assessment of gully erosion along major armoured roads in Mpumalanga, South Africa

Soil erosion is one of the most critical problems in the environment. Gullies, as a result of soil erosion are visible in large parts of South Africa. There are various elements that enhance gully formation like land use, climate, and anthropogenic elements. Armoured roads as an anthropogenic aspect change surface water flow because they are impermeable and thus lead to accelerated soil erosion and gully formation on their road shoulders.

Due to a shortage of studies addressing the linkage between aspects that enhance gully formation and armoured road drainage, the aim of this project is to determine how different factors impact gully development along major armoured roads. The objective is to determine if hillslope gradients and rainfall can be correlated to gully formation next to major armoured roads in the Mpumalanga province of South Africa.

This study will be conducted along national and regional roads in the Emakhazeni Local Municipality in South Africa’s Mpumalanga province. GPS locations will be physically collected of approximately 440 potential gullies next to the roads.

Possible gully locations next to the roads will be identified on Google Earth. These locations will be compared to the collected GPS locations in ArcGIS Pro. ILWIS software will be used to process the obtained 30 m ASTER GDEM data. Using spatial analyst tools, the locations of the gullies will be analysed in relation to elevation. Data on the change in amount of rainfall over the past 30 years will be obtained from rainfall stations closest to the measured gullies.

Dimensions of every 20^th gully will be taken with a surveyor’s tape to calculate gully volumes. These dimensions will include gully length, width, and depth. Measurements of the width and depth of each of these gullies will be averaged. The actual direction and direction of gullies in relation to the roads will also be noted.

The gully volumes, locations, rainfall data, and hillslope gradient will be stacked in ArcGIS Pro and extracted as a table, grouping the factors into gully extent categories. Simple linear regression and R² values will be used to assess the relationship between the factors. To determine if there was multicollinearity, the variance inflation factor will be calculated. The combined effect will be determined with multi-linear stepwise linear regression. ANOVA will be used to determine if significant differences are present. All the statistical analyses will be done with R in R studio.

By increasing the understanding of how hillslope gradient and amount of rainfall impact gully development, can add to previous knowledge about how gullies develop and extend. This will contribute to determining whether gullies that occur adjacent to armoured roads are created due to changes in these factors or due to the concentration of water at culverts. Furthermore, the use of GIS and remote sensing can also help to increasingly establish its use to assess gully development next to major armoured roads. Ultimately, it can aid in determining how suitable and sustainable land use and management practices are and where to effectively focus rehabilitation efforts.