Processes and drivers of Prosopis invasions in Eastern Africa

Castillo, Maria Loreto (2019-12)

Thesis (PhD)--Stellenbosch University, 2019.

Thesis

ENGLISH ABSTRACT: Increased movement of humans and goods around the world has facilitated the transportation of many species into new geographic ranges. A significant number of these have become invasive, resulting in substantial ecological, social, and economic impacts. In order to develop effective management strategies, it is necessary to elucidate the drivers underlying invasion and to understand what determines species invasiveness. Progress in the understanding and management of biological invasions depends on proper taxonomic identification of invasive species. However, the taxonomy of many alien taxa remains problematic due to unresolved species relationships, geographic distributions and/or inter-specific hybridization, among others. Mequite trees from the genus Prosopis are problematic invasive species in many parts of the world. To resolve taxonomic uncertainty among Prosopis species globally, I used phylogenetic and population genetic approaches to examine evolutionary relationships and levels of genetic diversity and population genetic similarity among Prosopis species collected from four native regions (Argentina, Chile, Mexico and Peru) and six non-native regions (Australia, Hawaii, Kenya, Ethiopia, Tanzania and South Africa). The genetic analysis showed high phylogenetic similarity, low genetic differentiation between species from the native range and provided evidence for inter-specific hybridization between different Prosopis species in both native and non-native ranges. My findings suggest that hybridization between previoulsy allopatric species may occur frequently when they are co-introduced into new ranges. In addition, polyploid individuals were detected in both native and non-native areas, with tetraploid P. juliflora being highly differentiated from other diploid Prosopis species. Polyploidy is therefore proposed as an additional mechanism that facilitates reproductive isolation between some Prosopis species. Lastly, levels of genetic diversity suggest that invasive populations in Eastern Africa (Kenya and Ethiopia) probably resulted from multiple introductions of two species, P. juliflora and P. pallida. Prosopis invasion in Eastern Africa provided an excellent opportunity to examine how the ecological and genetic attributes of invasiveness, and drivers of invasion success, vary with context and taxon, because the founder trees of two species, tetraploid P. julifora and diploid P. pallida, are still present in the original plantations today. Here, I exploited these unique circumstances and examined the mechanisms – such as plasticity, rapid post-introduction evolution and hybridization – that contribute to the invasion success of these trees in Baringo County, Kenya and Afar Region, Ethiopia. I found that in Baringo County, despite the similar invasion history of P. pallida and P. juliflora and probable inter-species hybridization, only P. juliflora individuals became invasive in the region; indicating that the success of Prosopis invasion is not attributed to hybridization but potentially to the higher ploidy of P. juliflora. Similarly, in Ethiopia’s Afar Region, genotypes consisted exclusively of P. juliflora. In Kenya’s Baringo County, I performed common garden and reciprocal field transplant experiments that indicated that high levels of phenotypic plasticity and post-introduction evolution had contributed to the invasiveness of P. juliflora. Similar levels of plasticity were absent from introduced, but non-invasive, P. pallida. My results also showed that different demographic processes may be occurring in the Afar region (Ethiopia) and Baringo County (Kenya). In the latter, contemporary genetic change during the invasive spread, or founder effects during initial range expansion from plantations, may explain the genetic erosion I found along the range expansion of Prosopis. In Afar Region, successful spread may have been promoted by gene flow from “source” plantations to invasive genotypes, homogenizing standing genetic diversity across the invasion. Lastly, by using landscape resistance modelling in both areas, I showed that dispersal among Prosopis populations was not influenced by any of the attributes analysed: physical distance between populations, variables related to human and animal-mediated dispersal along roads and rivers, bioclimatic and altitudinal conditions. Therefore, the dispersal of Prosopis populations was not constrained by any landscape variable, and probably involved frequent human-assisted long-distance dispersal. Overal, this study showed that hybridization, polyploidy or both have contributed to the invasiveness of Prosopis. Finally, this study formed part of a larger international collaborative project entitled “Woody invasive alien species (IAS) in Eastern Africa: assessing and mitigating their negative impacts on ecosystem services and rural livelihoods” (hereafter referred to as Woody Weeds), with the overall objective of mitigating the impacts of woody IAS on biodiversity, ecosystem services and rural livelihoods in Eastern Africa. For this, basic knowledge about the invasion process and the impacts of woody IAS are being evaluated and diverse control and sustainable land management strategies are being proposed. In collaboration with PhD students involved in the Woody Weeds project, I propose key components of research projects addressing complex social-ecological topics that facilitate inter-disciplinary and, when interacting with stakeholders, trans-disciplinary research. Trans-disciplinary approaches should have a clear structure that transcends disciplines through amultidisciplinary team with common goals. To allow for integration and upscaling of findings, there should be a co-design of data collection using different methodologies in the same experimental units/scales. It is important to have the clear intention to identify management options with stakeholders, estimate their effects and test their implementation, as well as provide transdisciplinary training for all project participants. To facilitate the integration of the drivers of alien plant invasion into the development of effective management options, I concluded with a discussion of two principal questions: (i) How does an improved understanding of the eco-evolutionary drivers of invasiveness help us to better manage the problem? (ii) What implications does the better understanding of genetic and ecological drivers have for the use of particular control methods, especially biological control?

AFRIKAANSE OPSOMMING: Die toenemende beweging van mense en goedere oor die wêreld heen veroorsaak dat baie spesies na nuwe voorkomsgebiede vervoer word. ‘n Beduidende aantal van dié spesies word indringerspesies, met wesenlike ekologiese, sosiale en ekonomiese impakte. Ten einde effektiewe bestuurstrategieë te ontwikkel is dit nodig om die drywers van indringing en die indringingsvermoë van spesies beter te verstaan. Vooruitgang in die verstaan en bestuur van biologiese indringings berus op behoorlike taksonomiese identifikasie van indringerspesies. Die taksonomie van baie uitheemse taksa bly egter problematies weens, onder andere, onopgeloste spesieverwantskappe en geografiese verspreidings en/of kruisteling tussen spesies. Prosopis bome in die genus Prosopis is ernstige indringers in baie wêrelddele. Ten einde die taksonomiese onsekerheid van Prosopis spesies op te klaar, het ek filogenetika- en populasie genetika-benaderings gebruik om die evolusionêre verwantskappe, vlakke van genetiese diversiteit en populasie genetiese struktuur in vier inheemse streke (Argentinië, Meksiko, Chile en Peru) en ses nie-inheemse streke (Australië, Hawaii, Kenia, Etihopië, Tanzanië en Suid-Afrika) te ondersoek. Dié genetiese analise het hoë filogenetiese eendersheid aangedui, lae genetiese onderskeiding tussen spesies met soortgelyke inheemse verspreidings en dat kruisteling tussen verskillende Prosopis spesies in beide inheemse en nie-inheemse verspreidingsgebiede geskied. My bevindings dui daarop dat kruisteling dikwels kan plaasvind tussen voorheen-allopatriese spesies wanneer sulke spesies gesamentelik na nuwe areas ingevoer word. Poliploïed individue is gevind in beide inheemse en nie-inheemse areas, met tetraploïede P. juliflora wat hoogs gedifferensieer was van ander diploïede Prosopis spesies. Die voorkoms van poliploïd variasie word dus voorgestel as ‘n addisionele meganisme wat voortplantingsisolasie tussen sekere Prosopis spesies fasiliteer. Laastens dui die vlakke van genetiese diversiteit daarop dat die indringerbevolkings in Oos-Afrika (Kenia en Etihopië) waarskynlik die gevolg is van veelvoudige vrylatings van twee spesies, P. juliflora en P. pallida. Die indringing van Prosopis in Oos-Afrika bied ‘n uitstekende geleentheid aan om die ekologiese en genetiese eienskappe van indringing en die drywers van suksesvolle indringing verskil tussen kontekste en taksa te ondersoek, aangesien die stigterbome van twee spesies, die tetraploïede P. juliflora en die diploïede P. pallida, steeds teenwoordig is in hulle oorspronklike plantasies. In hierdie studie het ek dié unieke omstandighede benut om die meganismes – soos fenotipiese plastisiteit, kruisteling en evolusie – te ondersoek wat bygedra het tot die suksesvolle indringing van hierdie bome in Baringo-streek, Kenia en die Afar-streek in Etihopië. Ek het bevind dat slegs P. juliflora in die Baringo streek ‘n indringer geword het, ten spyte van die soortgelyke indringingsgeskiedenis van P. pallida en P. juliflora en waarskynlike kruisteling tussen die spesies. Dit dui daarop dat die suksesvolle indringing van Prosopis nie toegeskryf moet word aan kruisteling tussen P. pallida en P. juliflora nie, maar moontlik aan die hoër poliploïd vlakke van P. juliflora. In die Afar-streek van Etihopië is al die genotipes die van P. juliflora. In Kenia se Baringo -streek het ek ‘n tuineksperiment en ‘n wedersydse oorplantingseksperiment uitgevoer, wat getoon het dat hoë vlakke van fenotipiese plastisiteit en evolusie bygedra het tot die indringing van P. juliflora. Soortgelyke vlakke van plastisiteit is nie gevind in P. pallida nie. My resultate wys ook dat daar moontlik verskillende demografiese prosesse plaasvind in die Baringo- en Afar-streke. In die Baringo -streek kan genetiese verandering gedurende die indringingsproses, of stigtereffekte tydens die aanvanklike verspreiding vanuit plantasies, moontlik die verlaging van genetiese variasie wat ek waargeneem het in die verspreiding van Prosopis verklaar. In die Afar-streek is die suksesvolle verspreiding moontlik bevorder deur deurlopende geenvloei tussen oorspronklike plantasies en indringerbevolkings, wat die genetiese diversiteit oor die indringerareas homogeniseer. Laastens het ek deur middel van landskapweerstandbiedendheids-modelle in beide areas getoon dat die beweging tussen Prosopis populasies nie beïnvloed word deur enige van die volgende eienskappe nie: fisiese afstand tussen bevolkings, veranderlikes verwant aan plant verspreiding deur mense en diere langs paaie en riviere, of bioklimaties en topografiese faktore. Dus word die verspreiding van Prosopis bevolkings nie beperk weens landskapveranderlikes nie en word waarskynlik bevorder deur gereelde langafstand-verspreiding deur mense. Om op te som, bewys hierdie studie dat kruisteling, poliploïdvorming, of beide dié faktore bygedra het tot die suksesvolle indringing van Prosopis in Oos-Afrika. Hierdie studie was deel van ‘n groter internasionale projek, naamlik “Woody invasive alien species (IAS) in Eastern Africa: assessing and mitigating their negative impacts on ecosystem services and rural livelihoods” (hierna verwys as “Woody Weeds”), met die oorhoofse doel om die impakte van indringerplante op biodiversiteit, ekosisteemdienste en landelike gemeenskappe in Oos-Afrika te verminder. Dít benodig kennis van die indringingsproses en die impakte van indringerplante, sowel as die ontwikkeling van verskeie beheermetodes en volhoubare grondbestuurstrategieë. In samewerking met ander PhD-studente van die Woody Weeds-projek, stel ek die sleutelkomponente voor van navorsingsprojekte wat komplekse sosio-ekologiese onderwerpe, soos indringerspesies, te ondersoek deur inter- en transdissiplinêre benaderings. Transdissiplinêre benaderings behoort duidelike projekstruktuur te hê wat verskeie dissiplines inkorporeer deur middel van ‘n multi-disiplinêre span met gemeenskaplike navorsingsdoelwitte. Om navorsingsbevindings te kan integreer en uit te bou, moet data-insameling gesamentelik deur alle projekdeelnemers ontwerp word, sodat verskillende metodes vir dieselfde skaal of eksperimentele eenhede gebruik kan word. Dit is belangrik om bestuursopsies saam met belanghebbers te identifiseer, om die effekte en implementering van die bestuursopsies te toets, sowel as om transdissiplinêre opleiding aan te bied aan alle projekdeelnemers. Ten einde die drywers van indringing deur uitheemse plante beter te integreer in die ontwikkeling van effektiewe bestuursopsies, sluit ek af met ‘n bespreking van twee kritiese vrae: (i) Hoe kan ‘n beter begrip van die eko-evolusionêre drywers van indringing ons help om die probleem beter te bestuur? (ii) Watter implikasies het insigte in die genetiese en ekologiese drywers van indringerspesies vir spesifieke beheermetodes, veral biologiese beheer?

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/107050
This item appears in the following collections: