An evaluation of southern Africa's elephant sub-populations as a metapopulation

Abstract

Elephant management traditionally centers on reducing ecological impact and human-elephant conflict by controlling numbers. However, such an approach only deals with symptoms, and ignores the causes of the problem. Planning for cases when a species is a nuisance in some areas, but threatened in others, could benefit from the application of metapopulation theory. The theory offers a framework that is elegant, and have ecological as well as political appeal. Applying classic metapopulation theory to long-lived species that are widely distributed in stochastic environments where they can resist extinctions is problematic. However, empirical evidence for metapopulation structure may exist when applying more lenient criteria. I examined the literature for empirical support of classic criteria set by Hanski (1999) and for a more lenient sub-set of criteria proposed by Elmhagen&Angerbjörn (2001) for specifically mammals. I propose that for small mammals (≤5kg) the full complement of classic criteria must be applied to yield perspectives on population regulation and conservation. However, for large (>100kg) and medium sized (>5≤100kg) mammals only habitat discreteness, potential of demographic asynchrony and the likelihood of dispersal among sub-populations must be evaluated. Metapopulation theory could then be useful when constructing conservation plans that ensure the persistence of a species and contribute to forces that stabilize populations regionally. I evaluated the applicability of metapopulation theory to southern Africa’s elephant sub-populations. I identified 51 discrete administrative sub-populations that occurred in six conservation clusters. Population growth rates varied across space and time within and among conservation clusters. Some sub-populations and conservation clusters increased or decreased while others remained stable. Therefore, elephant populations in southern Africa were in demographic asynchrony, both on a local and regional scale. I also suggest that dispersal may occur among sub-populations within clusters. Consequently, the regional population is stabilized by emigration to, or immigration from neighboring sub-populations as a result of demographic asynchrony across an ecological gradient. Elephant populations across southern Africa thus adhered to one and could possibly adhere to all metapopulation criteria. Observed changes in elephant numbers could also be the result of survey error. To gain an understanding of how survey error could affect estimates, I used dung counts and measurements to estimate population size and construct an age structure for the elephants living in the Maputo Elephant Reserve in Mozambique. I suggest that dung surveys can yield population estimates with known precision and can be used in monitoring programmes aimed at assessing population trends - despite the fact that it can be affected by observer bias. In this thesis I show that metapopulation theory provides the opportunity of applying a spatio-temporal approach to elephant conservation that is not obsessed with numbers. When implementing metapopulation theory, management no longer have to centre on elephants, but can focus on the landscape as a spatially and temporally dynamic area. Local fluctuations in elephant numbers could be construed within a regional context, rather than implementing management strategies on a local scale. Such an approach will focus on the causes rather than the symptoms of the elephant problem and may contribute to the persistence of elephants as well as other components of southern African biodiversity. AFRIKAANS : Die ekologiese impak van olifante en konflik tussen mense en olifante word tradisioneel hanteer deur olifant getalle te beheer. Ongelukkig los so ʼn benadering net die simptome en nie die oorsaak van die problem op nie. Gevalle waar ʼn spesie ʼn problem is in sommige areas maar bedreig is in ander, kan baat vind by die toepasing van die metabevolkings teorie. Die teorie bied ʼn elegante raamwerk wat op ekologiese en politieke gebiede aanklank vind. Die toepasing van die klassieke metabevolkings teorie op lang lewende spesies wat wyd versprei voorkom in stochastiese omgewings waar hulle weerstand kan bied teen uitsterwings skep egter probleme. Emperiese bewyse vir ʼn metabevolkings struktuur kan egter voorkom waneer meer gematigde kriteria ondersoek word. Ek het die literatuur ondersoek vir emperiese ondersteuning vir die klassieke kriteria wat Hanski (1999) voorgestel het, asook vir ʼn meer gematigte sub-groep van kriteria wat deur Elmhagen&Angerbjörn (2001) vir spesifiek soogdiere voorgestel is. Ek stel voor dat die klassieke kriteria aangewend kan word om bevolkings van klein soogdiere (5≤kg) te reguleer, beter te kan verstaan en dan te bewaar. Vir groot (>100kg) en medium groot (>5≤100kg) soogdiere kan die metabevolkings teorie net gebruik word as die bestaan van aparte habitate, demografiese asinkronie en die potensiaal van verstrooing tussen tussen subbevolkings bewys kan word. Die metabevolkings toerie kan dan gebruik word om bewarings inisiatiewe in te stel wat spesies in staat sal stel om voort te bestaan en oor die streek te stabiliseer. Ek het die toepasbaarheid van die metabevolkings teorie vir suidelike Afrika se olifant sub-bevolkings ondersoek. Ek het 51 aparte administratiewe sub-bevolkings geidentifiseer wat in ses ‘bewaringsklosse’ voorkom. Bevolkings groeitempos het binne in en tussen bewaringsklosse gewissel. Sommige het of toegeneem of afgeneem terwyl ander stabiel gebly het. Olifant sub-bevolkings in Suider Afrika was dus in demografiese asinkronie, op ʼn lokale sowel as op ʼn streeks vlak. Ek het ook voorgestel dat verstrooing kan voorkom tussen sub-bevolkings binne in bewarings klosse. Die olifant bevolking van die streek word dus deur emigrasie na, of immigrasie van naburige sub-bevolkings as gevolg van demografiese asinkronie oor ʼn ekologiese gradient gestabiliseer. Olifant bevolkings in Suider Afrika het dus voldoen aan een, en kan potensieel voldoen aan alle metabevolkings kriteria. Opmerklike veranderinge in olifant getalle kan ook wees as gevolg van foute wat tydens tellings gemaak word. Om beter te verstaan hoe sulke foute bevolking skattings affekteer, het ek olifant mis tellings en metings gebruik om ʼn bevolking skatting en ouderdomsstruktuur vir olifante in die Maputo Olifant Reservaat in Mosambiek saam te stel. Ek stel voor dat mis opnames bevolking skattings kan lewer wat bekende presiesie het en dat dit gebruik kan word in moniterings programme wat neigings in olifiant bevolkings ondersoek - alhoewel sulke skattings beinvloed kan word deur die vooroordeel van waarnemers. In hierdie tesis toon ek aan dat die metabevolkings teorie ʼn geleentheid skep vir ʼn ruimtelike-tydelike benadering in olifant bewaring wat nie net op getalle fokus nie. Wanneer die metabevolkings teorie toegepas word, kan bestuur op die landskap as ʼn ruimtelike en tydsgebonde dinamiese area fokus, in plaas van net op olifant getalle. Lokale wisselings in olifant getalle kan binne in ʼn streek konteks geinterpreteer word, eerder as om bestuurs inisiatiewe net op ʼn lokale vlak in te stel. So ʼn benadering sal fokus op die oorsprong in plaas van die simptome van die olifant problem en mag bydra tot die voorbestaan van nie net olifante nie, maar ook tot die biologiese diversiteit van Suider Afrika. Copyright