Geography

Customary land-use practices create distinctive cultural landscapes, including landscapes where abandoned settlements host vegetation that attracts wild animals. Understanding how landscape patterns relate to land-use history can help clarify the ecological effects of particular land uses. This study examines relationships between chimpanzee habitat selection and Maninka settlement practice, to determine how settlement history has affected chimpanzee habitat in Mali’s Bafing Biosphere Reserve, where conservation practitioners assume that the characteristic settlement pattern reflects a process of settlement expansion into undisturbed habitat. Three types of data are reported: (1) ethnographic interviews on settlement history and practice; (2) systematic sampling of chimpanzee habitat use; and (3) ground-based mapping of settlement sites, surface water, and fruit-tree patches. These data show that the Maninka have a shifting settlement system, meaning that most sites are occupied for only relatively brief periods; and that some abandoned settlement sites host fruit-tree patches that are seasonally important chimpanzee habitat. Two main conclusions are: (1) settlement expansion has not occurred; instead, historic settlement has created habitat that is both attractive and available to chimpanzees. Anthropogenic habitat does not appear to be vital for chimpanzee survival, but it spatially and temporally supplements natural habitats. (2) Conservation policies meant to reduce the presumed threat of settlement expansion may have initiated an unintended, long-term threat of habitat loss for chimpanzees. While settlement practices may be a component of short-term threats to chimpanzees, such as hunting, when addressing these threats conservation practitioners should consider long-term settlement processes to avoid creating new threats.

Aim  Human settlement establishment and reproduction of the baobab tree (Adansonia digitata) appear spatially and temporally dependent because baobabs are abundant in many settlement sites in Africa. This paper tests the spatiotemporal relationship between baobab and settlement distribution.Location  South-western Mali.Methods  In an area of 183 km2, 1240 baobabs were located and mapped, their diameters measured, and habitat characteristics recorded for each individual. All occupied (n = 9) and abandoned (n = 84) settlements were located and mapped, and occupation dates were determined through interviews. Chi-squared analysis indicated baobab habitat preferences, and bivariate point-pattern analysis tested baobab–settlement spatiotemporal independence.Results  Baobabs and human settlements are positively spatially associated at most distances and for all baobab size-class–settlement age-class pairs. However, positive spatial association is significant only at distances < 500 m, and young settlements and large baobabs are not significantly associated. Positive association between small and large baobabs is marginally significant at <300 m, but observed significance is less than that for baobab–settlement positive association. Baobab abundance is not evenly distributed across the range of habitats it occupies; recruitment is strongest in settlements and fields, and on cliffs, while mortality is highest on cliffs. Ethnographic observations suggest that human settlement practices and fruit use are the main human factors contributing to baobab–settlement positive spatial association.Main conclusions  There are three main conclusions: (1) Human settlement and baobab recruitment are spatially dependant because settlement leads directly and indirectly to the development of baobab groves at settlement sites. (2) The lower than expected abundance of mature individuals in natural habitats, and the habitat preferences of the observed population, suggest that baobabs were introduced to south-western Mali, probably centuries ago. (3) Human mobility over decadal time-scales is necessary to maintain baobab population structure in landscapes dominated by shifting land use, where baobabs are not purposefully planted. Baobab population processes in such landscapes occur at the scale of human settlement.

Historical analysis of botanical literature concerning the trees Gilletiodendron glandulosum and Guibourtia copallifera in Mali’s Manding Plateau reveals that the dominant representation of these plants has helped to perpetuate colonial-era theories of vegetation history, African land management, and natural resource politics in West Africa. The French botanist Aubréville described these plants as proof of the theory of vegetation history that blamed poor land management by rural Africans for a steady and continuing destruction of vegetation from its presumed original forest climax. Although Aubréville’s representation of these trees was justified within the 1930s scientific context he worked, subsequent researchers uncritically maintained his conclusions even though the changed scientific context in which they worked did not justify such representation. Subsequent ecological research also failed to substantiate Aubréville’s representation of these trees, yet several influential modern botanical works have uncritically accepted colonial-era botanical literature founded on his ideas. Thus, modern botanical works have perpetuated a simplistic and inaccurate narrative of resource use under an appearance of objectivity. As a result, policy recommendations based on the modern botanical sources remain almost identical to colonial-era policies. Based on the similarity of colonial-era and modern portrayals of these trees, this paper argues that a regional discursive formation recently described by other authors may be expanded to include southern Mali, which carries negative implications for decentralization reform in Mali.

News media influence local to global interactions between people, societies, and governments by producing place images. Representations of Africa in Western news media are heavily imbued with colonialist notions of cultural geography. In particular, Western news media have represented conflicts in Africa as ‘tribal’, a trope that erases geographic and historical context, and discourages actions that could prevent or reduce violent conflict. To determine if ‘tribalism’ remains important in coverage of African conflict, we use framing analysis to evaluate news on Sudan’s Darfur region in The New York Times and The Washington Post during 2003–2009. We find that these newspapers predictably relied on stereotypes related to tribalism to simplify Darfur’s geography and make the conflict meaningful to intended readers. Tribal portrayal of African war is inherently political, and, problematically, neither newspaper recognized that their use of the tribal narrative was parallel to the views of both the Sudanese government and external observers that challenged the actions of the Sudanese government. However, we also found that stereotypical representations became less prominent over time, apparently because reporters found that the initial, simplistic framing of the conflict did not match their encounters with geographic reality. We emphasize the decline in stereotypical tropes, because this suggests behind-the-scenes negotiation about representations in these news organizations. Recognizing voices that challenge stereotypical portrayals is necessary to developing place images that are geographically more accurate.

This article argues that understanding how people classify physical geographic features is necessary for identifying fundamental, cross-cultural geographic concepts that are required for successful communication of geographic knowledge. Academic geographers have not given sufficient attention to systems of local geographic knowledge, even though promising theoretical frameworks exist, particularly in the field of ethnoecology. However, the research approach that has characterized ethnoecology is insufficient to develop ethnogeography as a field of inquiry, because ethnoecologists have overemphasized limited aspects of local knowledge systems, such as soils, which has often led researchers to incompletely sample local knowledge systems. Using ethnographic methods, this article analyses the content and structure of physical geographic knowledge in the Maninka language as spoken in southwestern Mali, and compares Maninka knowledge to that of other cultural groups. The results suggest that broad physical geographic concepts may be shared pan-environmentally, but that most physical geographic knowledge is contained in culturally specific classifications embedded within a broad cross-cultural framework. Academic geographers should expect only broad correspondence between their categories of physical geographic variation and those of people who classify biophysical features according to local knowledge systems. Finally, this article also shows that ethnoecological research will be advanced if geographic theories of place are given more prominence in ethnoecological studies.

This article considers how geospatial analyses can influence cartographic outputs in studies of the spatial structure of food environments. We make two contributions. First, we present a new approach to conceiving and visualizing urban food environments as ‘nutritional terrains’, in which the opportunities and costs of locating (healthful) food vary continuously across space. While other researchers have conceptualized and represented food environments as continuous phenomena, we use detailed data to produce maps of food accessibility that have high resolution both spatially and in terms of food availability. Second, we show that decisions made about measuring and modelling food accessibility can create artifactual patterns independently of actual variation in food-environment characteristics. Although the type of method-driven patterning we identify will not surprise cartographers, we argue that non-geographers using geographic information technologies to visualize food environments must give greater attention to the unintended consequences of choices made in geospatial analyses.

Analysis of botanical data is presented from the standpoint of chimpanzee natural history. The Sudano-Guinean gallery forest type dominated by the tree Gilletiodendron glandulosum appears to be important habitat for chimpanzees due to vegeta- tion structure, presence of permanent surface water, and, particularly, abundance of diverse food plants throughout the year. Based on fecal analysis, observation of feeding remains, observation of sympatric primates, ethnographic research, and literature review, sixty proba- ble chimpanzee food plants have been identified in the Gilletiodendron forest of Mali. Phytogeographical analysis indicates that chimpanzees in Mali’s Sudanian climate zone eat mainly Sudano-Guinean plant species. Heavy reliance on Sudano-Guinean vegetation may indicate that modern chimpanzee populations in savanna areas are relicts, and that the species was originally adapted to mesic Guinean forests. There appears to be niche separation based on topography between humans and chimpanzees which breaks down in times of human food shortage, and the potential for competition is high.

Geographers have generally neglected African influences in attempting to understand the historical development of cultural landscapes in the tropical and subtropical Americas. In this paper, I analyze the early post-Columbian history of live fencing in the Americas, focusing on Spanish-held areas in the 1500s and 1600s. Live fences characterize many modern landscapes in the tropical and subtropical Americas, but the historical geography of these landscape features is poorly known. I show that live fences bear Native American, European and African inheritances, but argue that the African contribution was particularly significant. Specifically, escaped slaves – or maroons – like many contemporaneous communities in Africa, experienced conditions of endemic warfare and labour shortage. Live fences were an effective and labour-efficient means of defence, and all descriptions of live fences in the tropical and subtropical Americas before about 1800 were observed in maroon settlements. As African communities integrated into the multicultural societies of tropical and subtropical America other benefits of live fencing came to be more widely valued and integral to land management throughout the region – though its African inheritance has been forgotten. To understand more completely the historical cultural ecology of the Americas, geographers must challenge the deeply rooted belief that Africans contributed only labour in the development of New World landscapes.

This paper examines evidence for the origin of the tree Spondias mombin in Africa, where it is generally described as introduced from the Neotropics. Based on biogeographical, ecological, and historical evidence, this paper argues that the tree reached Africa via natural, long-distance dispersal, and should be considered an African native plant. The idea that it is introduced appeared in 1868, when it was a scientifically justifi- able conclusion reflecting the limitations of nineteenth-century biogeographical knowledge. By the early twentieth century, representing Spondias mombin as a human introduction in Africa was no longer scien- tifically justifiable because of advances in knowledge of the African flora and plant dispersal ecology. None- theless, most authors continued to represent the tree as introduced, in part because such representation was consonant with dominant images of the African environment: (a) that the continent had suffered extensive deforestation in the recent past, and (b) that Africa’s flora was depauperate in fruit trees prior to the coming of Europeans. Recent authors continue to describe Spondias mombin as introduced because they have not sufficiently considered the historic contexts of primary and secondary sources on Neotropical plant intro- ductions in Africa. This paper concludes that the human role in creating the African portion of the trans- Atlantic tropical flora has been overemphasized. Natural plant dispersal across the Atlantic may be more frequent than generally accepted.

Most studies of vegetation in African savannas have adopted reductionist approaches to research, mostly by looking at only “natural” or “anthropogenic” spaces in focal landscapes. Few have considered how social and biophysical factors interact to create vegetation patterns across landscapes. More holistic understanding of pattern–process relationships across savanna landscapes as wholes is needed to manage the ecological connections of global and local change. To understand how process interactions create vegetation patterns in a savanna landscape, I sampled vegetation in 217 sites in southwestern Mali, where conditions are similar to many parts of West Africa. For each site, I collected site-specific data on several biophysical factors as well as histories of settlement or cultivation, based on ethnographic interviews. Hierarchical clustering, indicator species analysis, multiresponse permutation procedures, and nonmetric multidimensional scaling revealed patterns of floristic variation across the landscape that related to patterns of biophysical variation and disturbance history. The results show complexity in terms of both pattern and process. Patterns evident at different scales of organization are not necessarily congruent, and vegetation heterogeneity is greater than expected based on any subset of biophysical or social processes. Soil moisture, bedrock lithology, and disturbance history are particularly important in shaping vegetation characteristics. Complex patterns arise because vegetation shows three different responses to settlement and cultivation—turnover, homogenization, and resilience—with actual outcomes that depend on site-specific biophysical and social conditions. This research shows that research epistemology can shape physical geographic knowledge and that savanna landscapes must be seen as entities with characteristics that are maintained through human–environment interaction. La mayoría de los estudios sobre vegetación de las sabanas africanas ha adoptado enfoques reduccionistas a la investigación, principalmente por mirar solo los espacios “naturales” o “antropogénicos” en paisajes focalizados. Pocos son los que se han preocupado del modo como interactúan los factores sociales y biofísicos para crear patrones vegetacionales de un paisaje a otro. Se necesita un entendimiento más holístico de las relaciones patrón-proceso a través de los paisajes sabaneros, como un todo, para manejar las conexiones ecológicas de cambio global y local. Para entender cómo las interacciones de proceso crean patrones de vegetación en un paisaje sabanero, hice muestreos de vegetación en 217 sitios del sudoeste de Mali, donde las condiciones son similares a muchas partes del África Occidental. Para cada caso recolecté datos específicos para el sitio sobre varios factores biofísicos, lo mismo que historias de asentamiento o agricultura, a partir de entrevistas etnográficas. Agrupamiento jerárquico, análisis indicador de especies, procedimientos de permuta de respuesta múltiple y escalado multidimensional no métrico, revelaron patrones de variación florística a través del paisaje que se relacionaban con patrones de variación biofísica e historia de la perturbación. Los resultados muestran complejidad en términos tanto de patrón como de proceso. Los patrones que aparecen en diferentes escalas de organización no son necesariamente congruentes, y la heterogeneidad de la vegetación es mayor de lo esperado con base en cualquier subconjunto de procesos biofísicos o sociales. La humedad del suelo, litología del substrato rocoso e historia de la perturbación son particularmente importantes en la conformación de las características de la vegetación. Aparecen patrones complejos debido a que la vegetación exhibe tres respuestas diferentes al asentamiento y a la agricultura—rotación, homogeneización y resiliencia—con resultados reales que dependen de condiciones biofísicas y sociales específicas del sitio. Este estudio muestra que la epistemología de la investigación puede moldear el conocimiento geográfico físico y que los paisajes sabaneros deben considerarse como entidades con características que trascienden la interacción sociedad-medio ambiente.

This dissertation shows that both human activities and biophysical processes interact in complex ways to create an emergent ecosystem structure in southwestern Mali. This dissertation includes five body chapters. The first chapter is an analysis of settlement history in the research area, and situates the research in the context of current conservation practice in Mali’s Bafing Biosphere Reserve. This chapter shows that the indigenous Maninka people practice shifting settlement, and that frontier-style settlement expansion is not occurring in the area, as conservationists have assumed. The second chapter is an ethnographic study of Maninka physical geography terms, and shows that Maninka farmers perceive the landscape as highly heterogeneous, with few areas suitable for settlement or cultivation. The third chapter examines floristic patterns across the landscape, and shows that most floristic variation is due to edaphic features, especially the hydrogeology of a specific type of sandstone bedrock. Human activities have variable affects on vegetation, depending on various socioeconomic and biophysical factors. The fourth chapter shows that humans have affected the distribution of the baobab tree across the research area through activities that create suitable baobab habitat in settlement sites. The final body chapter shows that anthropogenic baobab groves represent important habitat for chimpanzees, and that conservation policies that affect settlement practice may reduce baobab regeneration and thus reduce chimpanzee habitat in the long term.

Full citation: Duvall, C.S. 2011. Ferricrete, forests, and temporal scale in the production of colonial science in Africa. In Knowing Nature: Conversations at the Interface of Political Ecology and Science Studies (M.J. Goldman, P. Nadasdy, & M.D. Turner, eds.): 113-127. University of Chicago Press: Chicago.

Fire regimes emerge partly from human activities that reflect cultural-ecological knowledge of the relationships among fire, vegetation, grazing, climate, and other variables, as well as social relations. More knowledge of such “fire cultures,” past and present, therefore remains necessary to better understand the causes and persistent con- sequences of landscape burning. In the neotropics, people have used fire for centuries to manage livestock pastures. Conventional wisdom has long posited that such practices derived solely from antecedent European and indigenous, Native American fire cultures. Analysis of accounts of rangeland burning from throughout the neotropics during colo- nial times, however, demonstrates that ranchers incorporated African fire cultures and that the timing of burning shifted from early during the dry season in the sixteenth cen- tury to late during the dry season by the nineteenth century.