A comparison between indigenous environmental knowledge and a conventional vegetation analysis in north central Namibia

https://doi.org/10.1016/j.jaridenv.2004.11.004Get rights and content

Abstract

Local communities use an indigenous classification of environmental land units for natural resource management in central north Namibia. These indigenous land units (ILUs) were compared with a conventional vegetation analysis to improve understanding by scientists. The indigenous classification is based on many criteria. Detrended correspondence analysis was carried out on 388 vegetation samples, collected in a participatory way. The ordination diagrams of species and samples were a good reflection of ecological variation in the area. The data were used to draw sample standard deviation ellipses around the average ILU score. Classes with highly ranked vegetation criteria had little overlap with each other, while classes with no vegetation criteria often had large overlaps with other land classes. Advantages and disadvantages of working with indigenous environmental knowledge are discussed.

Introduction

Indigenous knowledge can be defined in its broadest sense as accumulated knowledge, skill and technology of local people derived from their direct interaction with the environment (Altieri, 1990, pp. 551–564). Information passed on through generations is refined into systems of understanding natural resources and relevant ecological processes (Pawluk et al., 1992). Such information systems are often considered to be primarily concerned with soils (Barrera-Bassols and Zinck, 2003, pp. 1–12). Ethnopedology is the indigenous knowledge of soils and encompasses many aspects including indigenous perceptions and explanations of soil properties and soil processes, soil classifications, soil management and knowledge of soil–plant inter-relationships (Williams and Ortiz-Solorio, 1981; Hecht, 1990, pp. 151–160). However, indigenous environmental knowledge (IEK) also includes perceptions and explanations on geomorphology, landscape classifications, settlement strategies, soil–water–plant relationships and range management (Verlinden and Dayot, 2000, pp. 63–78). Local classifications of land units are based on IEK and not on ethnopedology and soils alone.

Studies of IEK have been increasing during recent decades and in the late 1990s, studies on local land classifications were undertaken in northern Namibia (Dayot and Verlinden, 1999, pp. 254–283; Rigourd and Sappe, 1999, pp. 34–42; Shitundeni and Marsh, 1999; Verlinden and Dayot, 2000; Hillyer, 2004). These studies all aim to understand the ways that local people view and classify the land with the ultimate aim of understanding and improving indigenous resource management. The studies emphasize descriptions of the units, identification criteria and any potentials and limitations of the land. Most of the studies interpret classes as pedological units but Dayot and Verlinden (1999) found a separate soil classification system alongside a land classification, referred to as indigenous land units (ILUs). The land classification in northern Namibia is considered to be based on an appreciation of inherent patterns of geomorphology and ecological processes, some of which are very complicated and hard to measure like soil moisture movement (Rigourd and Sappe, 1999; Verlinden and Dayot, 2000). In a review of 800 ethnopedological studies, Barrera-Bassols and Zinck (2003) found the number of local classes range between 4 and 20. In Namibia, nearly 40 classes have been recorded in the study area indicating the existence of a relatively complicated system.

In classifying land, people make use of criteria (Ettema, 1994) that refer to ‘physical’ and ‘perceptual’ dimensions of land classifications (Weinstock, 1984). The ‘physical’ dimension concerns the most readily observable criteria that farmers use to differentiate their land units, namely soil characteristics that can be discerned by sight, feel, taste or smell (Osunade, 1992b). In Namibia, these are mainly soil color and texture or landscape characteristics that are identified by species composition, elevation, vegetation structure or abundance of termitaria.

Perceptual criteria are not as concrete as those in the physical dimension nor are they always readily recognized through the senses. Examples in Namibia are soil–water movement, soil workability, suitability classes for certain crops, suitability classes for grazing, biological indicators for soil fertility (e.g. fertility increase by termites) and grazing, sensitivity classes to certain agricultural problems (e.g. light soil indicating low fertility), and non-agricultural classes based upon the use of soil as building and pottery material.

While the local land classification system is widely used in Namibia, there is a lack of understanding by scientists or extensionists because it cannot be understood or verified with a simple soil analysis or soil classification. Many local classes fall into the FAO classification arenosol (sandy soil) (Rigourd, 1998) and this has been found elsewhere (Osbahr and Allen, 2003). Soil chemical analyses indicate some significant differences between ILUs but definitely not all of the identified ILUs (Rigourd and Sappe, 1999; McDonagh and Hillyer, 2001; Hillyer, 2004). An improved understanding could be gained by increasing the number of soil parameters but soil moisture processes are notoriously difficult to measure and in this semi-arid environment, some very important processes happen only rarely and unpredictably causing difficulties for replicable sampling.

Vegetation is the result of many ecological processes and is relatively well studied in southern Africa. In tropical savannas, the relationship between vegetation, soil, geomorphology and geology is close (Cole, 1982, pp. 145–174). Therefore, a conventional vegetation analysis, using ordination techniques like a detrended correspondence analysis (DCA) or classification techniques like clustering, could assist in understanding a local environmental classification. A vegetation ordination or classification would provide a simple tool, if local criteria for classes were only based on plant indicators, structure and plant species relationships. In such an hypothetical case, the local land classification could be understood as an ordination of vegetation samples falling within clear groups of samples of the same land unit and separate from other classes where vegetation characteristics differ. However, since a wide range of criteria are used including non-vegetation criteria, it is not that simple.

In the Namibian case, it appears that criteria are ranked according to their importance for each ILU. Some ILUs have vegetation criteria that rank high while some ILUs score higher with respect to geomorphology or other soil features. This suggests that the ordination diagram of ILUs with important vegetation criteria should show distinct groups in a sample ordination and other classes less distinctive. Also, the upper hierarchical groups in the local taxonomy should show more scattered samples, occupying more space in the ordination diagrams and overlapping with sub-classes. Nevertheless, there should be a consistent pattern; for example, ILUs with criteria for elevation should have types of vegetation indicating drier circumstances and ILUs with criteria for depressions should have a species composition indicating relatively more moist environments. The resulting ordination diagram should show whether or not the local criteria have ecological meaning. This study explores the use of vegetation analysis, including a conventional vegetation ordination, using DCA (Jongman et al., 1995) to improve the understanding of the ILU classification in north central Namibia.

Section snippets

Study area

Fig. 1 locates the study area in Africa and Namibia.

Approximate rainfall isohyets indicate a range of annual precipitation of between 350 mm in the south-west and 550 mm in the northeast (Hutchinson, 1995, pp. 17–37). There are three seasons: cold dry, May–August; hot dry, September–December; hot wet, January–April. There is great variation in temperature between day and night: in winter, the night temperatures drop to 7 °C with day temperatures rising to 27 °C or higher. During the hot season, the

Indigenous land units and their criteria

Table 1 lists 39 ILUs described so far with the main criteria classes and their ranking. Different dialects for the same unit are separated with the symbol ‘/’ while ‘-’ is used to identify a sub-unit (in the case of a local tree name) or a transition unit between two different ILUs, but recognized as a separate ILU. Results show that criteria for soil aspects, vegetation characteristics and landform are jointly used in most classes. Elevation, soil and soil–water characteristics are the only

Discussion

Results of this study confirm the relevance of comparing the ordination result with the indigenous environmental classifications. The ordination of species and samples showed very high eigenvalues on the first two axes (Jongman et al., 1995), indicating that most of the variation in the vegetation ordination is explained by the first two axes. Furthermore, the diagrams of the species and samples could be clearly interpreted with the main ecological aspects of the environment in the study area.

Acknowledgements

The participation by so many individuals in the field work and their willingness to share their knowledge is highly appreciated. M.K. Seely and A.E.M. Hillyer are gratefully thanked for commenting thoroughly on drafts of the paper. The Department for International Development (DfID), UK through the Northern Namibia Environmental Project, the French Mission for Co-operation and Cultural Affairs through the Northern Livestock Development Project in Windhoek and the Department for International

References (48)

  • B.E. Dialla

    The Mossi indigenous soil classification in Burkina Faso

    Indigenous Knowledge & Development Monitor

    (1993)
  • Du Plessis, W.P., 1991. Preliminary vegetation map of Owambo. Report on Preliminary Biological Survey of the Owambo...
  • Northern Namibia Environmental Project, Namibia. General and Technical Proposal to ODA

    (1996)
  • A. Erkkilä

    Living on the landchange in forest cover in north central Namibia 1943–1996

    Silva Carelica

    (2001)
  • A. Erkkilä et al.

    Forestry in Namibia 1850–1990

    Silva Carelica

    (1992)
  • C.H. Ettema

    Indigenous Soil Classifications. What is Their Structure and Function, and How Do They Compare to Scientific Soil Classifications?

    (1994)
  • G.E. Gibbs-Russell et al.

    Grasses of Southern Africa

    Memoirs of the Botanical Survey of South Africa

    (1990)
  • W. Giess

    A preliminary vegetation map of South West Africa

    Dinteria

    (1971)
  • M. Githinji et al.

    Social and ecological sustainability in the use of biotic resources in sub-Saharan Africa

    Ambio

    (1993)
  • S.B. Hecht

    Indigenous soil management in the Latin American Tropicsneglected knowledge of native peoples

  • M.O. Hill

    DECORANA—a FORTRAN Program for Detrended Correspondence Analysis and Reciprocal Averaging

    (1979)
  • Hillyer, A.E.M., 2004. Room to grow: local knowledge, land use and legumes in the farming systems of northern Namibia....
  • Hutchinson, P., 1995. The climatology of Namibia and its relevance to the drought situation. In: Moorsom, R. (Ed.),...
  • R.H.G. Jongman et al.

    Data Analysis in Community and Landscape Ecology

    (1995)
  • Cited by (44)

    • Distribution and dynamics of the Cynodon dactylon invasion to the cultivated fields of pearl millet in north-central Namibia

      2022, Journal of Arid Environments
      Citation Excerpt :

      Studies on indigenous land units in north-central Namibia have highlighted the unique characteristics of each type. For instance, the Omutunda type consists of elevated areas with a hardpan, with depths ranging from 0.5 m to 1 m, and is often used for cultivation (Verlinden and Dayot 2005). Prudat et al. (2018) also pointed out that Omutunda has high agricultural value and is preferred for cultivation.

    • Knowing, farming and climate change adaptation in North-Central Namibia

      2011, Global Environmental Change
      Citation Excerpt :

      Given the widespread use by farmers of the land unit system, it is surprising that it does not appear to find its way more systematically into agricultural extension, or indeed into agricultural policy more broadly. Verlinden and Dayot (2005:166) attribute this to “scepticism” on the part of scientists and extensionists. This plausible explanation nonetheless leaves room for further enquiry.

    View all citing articles on Scopus
    View full text