Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Afrique du Sud → Establishing indicators of biological integrity in Western Namibian rangelands

University of the Witwatersrand (1999)

Establishing indicators of biological integrity in Western Namibian rangelands

Zeidler J

Titre : Establishing indicators of biological integrity in Western Namibian rangelands

Auteur : Zeidler J

Université de soutenance : University of the Witwatersrand

Grade : Doctor of Philosophy PhD (Animal, Plant & Environmental Sciences) 1999

An Index of Biological Integrity (IBI) has been developed for rangeland condition assessment in arid northwestern Namibia. The usefulness of termites as bio-indicators has been tested and reliable sampling protocol for termite diversity in an arid environment has been developed. The study was conducted mainly at a high- and low-intensity site on each of three farms. Two of the farms were communally owned ; the other was commercially owned. No single reliable method for rapid termite diversity assessment seems to detect all termite taxa present. Three different methods for sampling termites generated different results. A standard belt- transect, a defined search protocol and a baiting experiment using three different types of bait, were compared. The search protocol required the lowest sampling effort, however produced the poorest results by detecting the lowest number of taxa. The standard belt-transect and baiting methods, both, produced similar results in numbers of taxa found, but often-sampled different termite taxa. Trinervitermes sp. were found mainly by applying the transect method, representatives of Macrotermes, however, through the baiting experiment. Microtermes and Psammotermes were detected with any of the used methods. They were the most frequently found termite taxa in the study area. From the results it becomes apparent that (1) several termite taxa are highly seasonal and (2) certain sampling methods are appropriate for certain termite taxa. Compared to other arid regions in Africa and elsewhere, the number of termite taxa collected from the study area (11 genera, 18 identified taxa) is close to numbers sampled from similar environments. The IBI includes measures of vegetation (forage quality, herbaceous, woody and litter biomass), soil parameters (nitrogen (N), organic carbon (OC), phosphorus (P), C:N ratio, light fraction (LF) and LF:C ratio), as well as termite diversity measures such as the community composition of functional groups. Application of the combined IBI at six study sites of varying land tenure and land use intensity reveals that the ecological resource base on the communally owned farms was more constrained than on the commercially owned farm. Organic Carbon (OC) levels were extremely low, ranging between 0.12 and 0.27 %. The OC level could be the factor limiting soil resilience in systems as arid as the study area. The OC levels were especially low under high land use intensity on the communal farms. The C:LF ratio was used as an indicator of soil resilience. The turnover of light fraction, the "active" OC pool, into soil organic matter, the "slow" and more stable OC pool, is indicative of transformation processes that are crucial to the maintenance of soil fertility. The C:LF ratio ranged between 5 and 9% in March 1998 and 6 and 17% in October 1998 at the six study sites. The higher levels measured in October, towards the end of the dry season, are explained by the work of termites and other soil related flora and fauna, which contribute to the replenishment of soil nutrients after the growing season (October-March). It was conspicuous that at one of the communal farms the C:LF ratio remained low over the seasons. Soil fertility was overall lowest at this site. The number and composition of termite taxa found at the various study sites differed, as did the composition of functional groups. This could affect the soil - soil biotic interface and interaction, since not all termite species will contribute to underlying ecosystem processes in a similar fashion. The term "land use intensity" was defined and explored in detail in this study. Livestock numbers obtained annually by the Ministry of Agriculture indicated that stocking numbers were highest at the commercial farm. This farm, however, obtained the best IBI scores in the habitat assessments. Stocking numbers at the commercial farm averaged 441 LSU over a 10-year period, whereas the two communally owned farms recorded averages of 372 and 356 LSU respectively. All farms were of similar size. It was therefore concluded that stocking numbers, per se, are a poor indicator of land use intensity. In-depth investigations at one of the communal farms aimed to establish, in close collaboration with the farmers : (1) reliable stocking numbers at their farms ; and (2) causes of land degradation. It is recommended that stocking numbers on communal farms ought to be established at a household level and not on a farm level. Tracking of animal numbers should be done in a participatory fashion and on a monthly basis to gain useful insights into de-stocking and restocking dynamics on communal farms. The farmers identified various causes of land degradation on their farm, naming especially lack of access to secure emergency grazing areas under prolonged drought conditions, and lack of decision-making powers on their own farms.

Présentation (NRF)

Page publiée le 24 septembre 2017, mise à jour le 3 janvier 2019