Remerciez-le!

Remerciez @Admin pour avoir partagé cet document gratuitement, de la manière la plus simple, en partageant sur les réseaux sociaux.

TROPICULTURA, 2013, 31, 1, 20-27 The Spatial Footprint of the Non-Ferrous Minin

TROPICULTURA, 2013, 31, 1, 20-27 The Spatial Footprint of the Non-Ferrous Mining Industry in Lubumbashi I. Vranken1,2*, Y.M. Amisi3, F.K. Munyemba4, I. Bamba2, F. Veroustraete5, M. Visser2 & J. Bogaert1 Keywords : Copper Belt- Katanga- Atmospheric deposits- Landscape metrics- Perception- Kevin Lynch Mots clés : Arc cuprifère- Katanga- Dépôts atmosphériques- Indices de structure spatiale- Perception- Kevin Lynch Summary In the south-eastern part of the Katanga Province (Democratic Republic of the Congo), high concentrations of copper and cobalt are found in the soils of the well-known “Copper Belt”. Due to dominant south-eastern winds, the metallurgic industry in Lubumbashi has been the source of spatially concentrated atmospheric deposits of non- ferrous metal particles and associated substances in a cone-shaped zone, situated north-west of the metal processing site. The existence of this zone has been evidenced using two different techniques: firstly, by means of landscape metric comparisons of the vegetation and bare soil patterns in two study areas, one inside the pollution cone and one outside; secondly, by means of the city perception theory developed by Kevin Lynch. Higher fragmentation and lower vegetation presence were observed inside the pollution cone, reflecting the negative impact of the atmospheric deposits. Those differences were higher for sites closer to the emission source. Lynch’s approach outlined the negative impact of diverse industrial plants on the perception of the local population. Six pollution districts and several contaminated paths, limits, nodes and polluting landmarks were identified. Citizens even recognize them as part of the collective image of the city. Résumé Empreinte spatiale de l'industrie minière non ferreuse à Lubumbashi Dans le sud-est du Katanga (République Démocratique du Congo), de hautes concentrations de cuivre et cobalt sont présentes dans les sols de l’Arc cuprifère. Suite à des vents dominants des secteurs sud-est, l’industrie métallurgique à Lubumbashi a été la source de dépôts atmosphériques de métaux non ferreux et substances associées, concentrés en une zone en forme de cône au nord-ouest de l’usine. L’existence de ce cône a été démontrée par deux techniques: premièrement, au moyen de comparaisons d’indi- ces de structure spatiale de la végétation et des sols nus; et ensuite avec la théorie de la perception urbaine développée par Kevin Lynch. Une fragmentation plus importante et une présence plus faible de végétation ont été observées dans le cône de pollution, ce qui reflète l’impact négatif des dépôts atmosphériques. Ces différences étaient plus élevées pour les sites plus proches de la source d’émission. L’approche de Lynch a mis en évidence l’impact négatif de plusieurs usines sur la perception de la population locale. Six quartiers de pollution et plusieurs voies, limites et nœuds contaminés ainsi que des points de repères polluants ont été identifiés. Les citoyens les reconnaissent même comme partie intégrante de l’image collective de la ville. Introduction Soil contamination by atmospheric deposits of non- ferrous metals has been described for ecosystems worldwide (5, 15, 34, 38). In the south-eastern part of the Katanga Province (Democratic Republic of the Congo), high concentrations of copper and cobalt are present in the soils of the well-known “Copper Belt” (14). This zone has been a place of intensive mining in the colonial period itself (until 1 Université de Liège, Gembloux Agro-Bio Tech, Unité Biodiversité et Paysage, Gembloux, Belgique. 2 Université Libre de Bruxelles, Ecole Interfacultaire de Bioingénieurs, Service d'Ecologie du Paysage et Systèmes de Production Végétale, Bruxelles, Belgique. 3 Université de Lubumbashi, Faculté des Sciences, Lubumbashi, R.D. Congo. 4 Université de Lubumbashi, Faculté des Sciences Agronomiques, Lubumbashi, R.D. Congo. 5 Flemish Institute for Technological Research, Atmospheric Processes and Remote Sensing Unit, Mol, Belgium. * Auteur correspondant : Email : ivranken@ulg.ac.be 20 TROPICULTURA, 2013, 31, 1, 20-27 1960) and afterwards, when mining activities and pyrometallurgic industry were continued by the Gécamines Company (4). The analysis of landscape patterns is justified by the pattern/process paradigm, a central hypothesis in landscape ecology (37) linking emerging patterns to underlying processes. Landscapes close to metal processing sites generally display a distinct pattern of scattered vegetation patches embedded in a matrix of bare soil (19, 26). Due to the dominant south-eastern winds, the metallurgic industry in Lubumbashi, capital of the Katangese Copper Belt, has been the source of spatially concentrated atmospheric deposits of non-ferrous metal particles and associated substances in a cone-shaped zone or “pollution cone”, situated north-west of the metal processing site and characterised by degraded vegetation (11, 20, 24). At the landscape level, the footprint of this long-term metal processing activity should therefore be detectable as a zone with higher vegetation fragmentation and higher bare soil presence. Unfortunately, this footprint, and hence the existence of the pollution cone, has not yet been evidenced using landscape metrics (8). Therefore, this paper compares the patterns of vegetation and bare soil of two oppositely placed study areas, one inside the pollution cone and one outside. According to Kevin Lynch (23), environmental perception or legibility is central for every living creature capable of motion because perception determines the way they exploit their environment to subsist, depending on their movements across the landscape. Features leading to stronger images for citizens (high “imageability”) then form the urban system, the analysis of which should be at the base of urban design (23). Industrial landscapes are an issue of concern when a perceptive approach is applied since industrial residues may spread far beyond the industrial infrastructure itself, e.g., bare soils resulting from vegetation degradation due to environmental contamination (3, 23, 29). The application of the Kevin Lynch theory (23) is considered complementary to the aforementioned analysis based on landscape metrics, and is hence the second objective of this paper. Evidencing the ecological footprint of the non-ferrous mining industry is considered crucial to confront local decision makers with the negative impact of metallurgic industry within an urban context. Material and Methods Landscape metrics A map of Lubumbashi containing four land cover classes (vegetation, bare soil, built-up, water) based on a Quickbird image of 2005 was used (27, 28); spatial resolution was set to 40 m to fit the recommended 30-100 m resolution range (32) for pattern analysis. Two regions of interest (ROI) of 79 km² were defined and compared, one inside the pollution cone, situated north-west of the emission source, i.e. inside the deposit range (18) and one outside the pollution cone, situated south-east of the emission source (Figure 1). Each ROI was divided in three subzones of 13 km², 26 km² and 52 km² (with the smallest subzone closest to the emission source) in order to enable the detection of the influence of spatial scale. Total area, number of patches, average patch size and area of the largest patch of the vegetation and bare soil classes were calculated and noted as av, nv, āv, amax,v and as, ns, ās, amax,s respectively. To avoid any influence of the absolute areas of vegetation and bare soil on pattern measurement and to exclude non-pollution related differences between the zones, two ratios were calculated to compare the relative presence of vegetation and bare soil. R1 (Equation I) expresses the dominance of vegetation over bare soil (R1>1) or the dominance of bare soil over vegetation (R1<1): R1=av a s (I) R1 is expected to be lower inside than outside the pollution cone because of the higher presence of bare soils due to toxic deposits. Secondly, R2 (Equation II) compares the average patch sizes of both classes: R2=av as (II) R2 is expected to be lower inside the pollution cone and higher outside because of the supposed higher vegetation fragmentation in the cone. The fragmentation degree of both land cover types was also measured by the index of the largest patch (7) expressing the dominance (%) of the largest patch [Dv for vegetation (Equation III), Ds for bare soil (Equation IV)] inside its class: Dv=amax , v av (III) D s=amax , s as (IV) 21 TROPICULTURA, 2013, 31, 1, 20-27 22 Figure 1: Map of the two study areas in Lubumbashi situated oppositely of the emission source (Gécamines smelter), indicated by a star symbol (A6). Each study area of 79 km² (rectangles A1/A5 respectively B1/B5) contains three subzones of 13 km² (rectangles A4/A5 respectively B4/B5), 26 km² (rectangles A3/A5 respectively B3/B5) and 52 km² (rectangles A2/A5 respectively B2/B5). As a consequence of the prevailing south-eastern winds, the pollution cone is expected to be situated north-west of the emission source. The grey zone indicates the central part of Lubumbashi, including the following municipalities: Katuba, Kampemba, Lubumbashi, Kamalondo, Kenya and Ruashi. TROPICULTURA, 2013, 31, 1, 20-27 Dv was expected to be higher outside the pollution cone and Ds was expected to be higher inside the pollution cone. Perception analysis In order to analyse the “imageability” and legibility of the city, cognitive cartography was realised. The study was carried out in 2007 and inscribed inside the agglomeration of Lubumbashi according to perceived limits empirically defined during field prospection (1). Two different methods were used, the first approach consisted of the identification of the structuring elements composing the city images: edges, paths, nodes, landmarks and quarters (23, 40). Point edges (1, 2) have also uploads/Ingenierie_Lourd/ 20-pdf.pdf