Publications

Makanzu Imwangana, F.,  Moeyersons, J.,  Ozer, P.,  Ntombi, M. & Dewitte, O. 2017. ‘Gully erosion in Kinshasa: hydromorphogenic dynamics and development of prevention tool’. The Belgium Geographers Day. Book of abstracts. Lège. (PR)

Résumé de colloque

Soil erosion has multiple linkages with land planning, environment changes and urbanization in developing countries. Gully erosion has been recently reported as an emerging hazard connected with the rapid development of Sub Saharan towns in Africa. Since the start of urbanization at the end of the 60s’, the high town of Kinshasa (DR. Congo) suffers under an important gully erosion risk. In 2007, Kinshasa showed 308 mega-gullies having 94, 7 Km of cumulated length. The accumulated length evolution between 1957 and 2010 was exponential. The gully density varies from 0, 4 to 2 Km/Km2. The average depth and width are 7 and 21 m respectively. The following questions are important: “Why gullying has been initiated in the high town of Kinshasa and not in the surrounding rural area? What are the dynamics, factors and impacts? How to prevent it? So, the main objective of this work is to control the runoff process in order to help in the decisions making for the urban planning in Kinshasa. The work first focus on the identification of that runoff origin. The hypothesis that increased gullying would be due to a change in rainfall regime could be excluded. Indeed, none of the 17 verified rainfall indices show a statistically meaningful evolution able to cause the change of Kinshasa’s landscape. We notice stability in the evolution of the intensity of heavy rainfall and a slow decrease in the extreme rainfall intensity at the west side of the town where the occurrence of gullying is best expressed. On the other hand, the causal relationship between urbanization and gullying is firstly shown by the temporal spatial mapping of gullies in the urbanized zones in 1957, 1977, 2007 and 2010. It clearly appears that gullies develop with a delay of less than ten years after the built up of the sectors lacking an adequate drainage system. There are nearly not gullies outside of the urban zone. We notice on the SPOT pictures of 2006-2007 that runoff in the high town is concentrated in a network of anthropogenic origin constituted of roads, channels and tracks; 91% of mega-gully heads are linked to those infrastructures. 43.8 % of the 308 gullies, mapped, are qualified as axial mega-gullies and 51% as leak mega-gullies against 5.2% that don’t have any link with urban infrastructure. In order to find a way to fight against gullying, the research has first of all been directed to the identification of soil use/cover which generates critical runoff discharges. Based on a old mathematical simulation of the infiltration capacity of soil, which we fed with data from a rainfall simulator, we have developed a new method to calculate the site- and rainfall specific runoff coefficient. Calculations show that in the present rainfall regime, roads are by very far the biggest producers runoff. But the difference of the runoff coefficient between roads and other soil uses decreases with the increase of rainfall height and intensity. Investigations done among the population has provided data of 23 active rainfall events from 1975-2010, the pluviograms of which served to calculate runoff coefficients. These investigations show that the critical rainfall height for gullying in the high town of Kinshasa is 24.9 mm with an average intensity of 21.8 mm h-1. The topographic control of gully initiation is represented by the equation S=0.00008A-1.439l where A is the drainage area giving to the gully head, in hectares, and S is the natural slope in mm-1 at the gully head. In comparison with other regions of world, the sandy soil of Kinshasa figures among the most sensible soils for gullying in the world. Taking into account the importance of roads in the production of runoff, we replaced the critical (S-A) relation by the (S-Lc) relation, where Lc stands for the cumulative distance of roads in A. The (S-Lc) relation is called “road control”. Topographic and road controls allow a reorganization of runoff so that it becomes non- erosive. The knowledge of critical rainfall allows the introduction of early warnings in the weather forecast.