Restoration potential of alien-invaded Lowland Fynbos

Journal Title
Journal ISSN
Volume Title
Stellenbosch : Stellenbosch University
ENGLISH ABSTRACT: Widespread degradation of ecosystems on a global scale has led to the need for ecological restoration in order to more effectively conserve and manage threatened ecosystems. While this is an emerging field, substantial challenges remain in achieving optimal success in restoration interventions, which need to be addressed in order to improve recovery potential of degraded sites. This is especially important in highly biodiverse regions in urgent need of restoration intervention, of which the critically endangered vegetation type Cape Flats Sand Fynbos, found within the Cape Floristic Region (CFR) in South Africa is a prime example. Of particular concern in this case is the impact of invasive alien plants, especially where invasive plant removal alone does not facilitate native vegetation recovery. Use of alternative passive clearing methods, or the incorporation of active seed sowing interventions, may improve restoration success in such cases. This dissertation investigated the following aspects of restoration: (1) seed ecology and the determination of effective germination cues of structurally important species in this vegetation type (chapter 2), (2) different passive and active restoration treatments in order to determine best alien clearing and management protocols (chapter 3), (3) predicting long-term vegetation recovery trajectories under different restoration treatments and pinpointing ecological thresholds through the use of a dynamic model incorporating vegetation recovery rates under passive and active restoration treatments (chapter 4). A seed ecology study investigated the effectiveness of smoke and heat exposure as a pre-treatment of seeds in improving germination success of species within Cape Flats Sand Fynbos. Fourteen species were exposed to a range of durations of exposure at different temperatures, in combination with exposure to smoke residue, after which seeds were germinated on agar to determine treatment success. Most species fell into one of two main groups: Seed germination in the first group (e.g. Anthospermum aethiopicum, Metalasia densa and Watsonia meriana) was greatest following either a lower temperature heat pulse, an extended period of mild temperature exposure, or no pre-treatment with heat. Seed germination in the second group (e.g. Pelargonium elongatum, Phylica cephalantha and Trichocephalus stipularis) was promoted after brief exposure to higher (100°C) temperatures. Species within the latter group mainly possessed physical dormancy. Passive Acacia saligna clearing treatments in the field involving either burning or stacking biomass after initial alien clearing, as well as active intervention involving sowing seeds of native species either directly after burning or a year later and either with or without seed pre-treatment, resulted in very different recovery trajectories over a two year survey period. No treatment resembled a reference uninvaded vegetation condition after two years. Clearing alien biomass without burning was the cheapest method and facilitated recovery in less degraded areas, but otherwise there was very limited or no native vegetation recovery and this facilitated secondary invasion by herbaceous weeds. Burning after clearing controlled some guilds of secondary invasive species, but stimulated mass acacia reestablishment and did not result in native vegetation recovery. The native seed bank was found to have been depleted during the period of invasion, hence the lack of autogenic recovery under passive treatments. However, while clearing without burning resulted in low acacia reestablishment, two years later the acacia seed bank was similarly reduced both with and without burning. Active seed sowing was the most expensive treatment but resulted in the highest recovery of native shrub cover and diversity. These findings suggest that a biotic threshold has been crossed where passive treatment does not result in recovery of vegetation components. Active sowing was able to partially reverse this threshold through improved recovery of total shrub cover, while pre-treatment of seeds before sowing improved establishment of some species. However, non-sprouting shrub cover was over-represented while resprouting shrubs and species of Restionaceae, which germinated successfully under greenhouse conditions, were under-represented relative to the reference condition. This suggests that barriers still exist in preventing establishment of some species from seeds sown in the field. A dynamic model was designed to analyze the effectiveness of the different restoration methods, enabling the extrapolation of recovery trajectories over a long time span which could not be determined from short-term field surveys alone. Data for rates of vegetation recovery under different passive and active restoration treatments over the course of two years of field surveys were fed into the model. The modelled simulations showed that different treatments in similar degraded states at the time of initial clearing resulted in vastly different recovery trajectories when extrapolated over an extended period of time. Active seed sowing was initially the most expensive treatment but resulted in the highest indigenous shrub recovery, which in turn decreased the costs of longer-term follow-up alien plant clearing, as well as providing competition against establishment of secondary invasive species. Clearing without burning was the cheapest method, but resulted in limited recovery of indigenous or acacia cover over the long-term, leaving barren ground prone to secondary invasion by herbaceous weeds. The model therefore supported the finding of the field restoration experiments that ecological thresholds have been crossed which prevented recovery of certain vegetation components. Active sowing was able to partially reverse these thresholds and resulted in sustained shrub cover, but even this treatment did not resemble the reference condition in terms of restoring a balance of structural components after simulating over an extended period of time. By synthesising the knowledge gained from the three aspects of this dissertation, and in light of what was previously known to work effectively, management recommendations were formulated to provide practitioners with more effective guidelines for best practise in future restoration of lowland fynbos vegetation. Firstly, it was concluded that biotic thresholds to restoration have been crossed within the study site, due to the lack of autogenic recovery of native cover and diversity under passive treatments as a result of a depleted native seed bank. Active interventions can partially reverse this threshold through the reintroduction of native seeds, and increase resilience of the vegetation to secondary invasive species. Secondly, the additional knowledge gained through germination tests using heat and smoke pre-treatment of seeds is highly valuable for incorporating into further restoration work. The use of species-specific pre-treatments for stimulating optimal germination can improve restoration success following active seed sowing intervention in the field. However, some species which respond well to seed pre-treatment in greenhouse conditions fail to establish from seed in the field, and these species are better restored through nursery propagation for planting into the field. Thirdly, the best treatment to use in the field was found to depend on the level of degradation at the time of initial alien clearing. A less degraded site with representative native vegetation structure i.e. more than 10% cover of both non-sprouting and resprouting native shrubs will recover after clearing without burning, while a more degraded site lacking native vegetation structure or cover will require active intervention in order to facilitate recovery of native cover and improve resilience to invasion by primary or secondary invasive species. Since burning after clearing resulted in mass recovery of Acacia saligna, both in passive and active restoration treatments, two alternative treatments to those tested in the field are proposed. The most appropriate treatment would depend on the size of the secondary invasive seed bank. If this seed bank is small, clearing without burning and sowing pre-treated native seeds should be more effective than burning. If the secondary invasive seed bank is large, then clearing and waiting two years to facilitate acacia seed bank reduction, after which a management burn is conducted, should deplete the secondary invasive seed bank. Sowing native seeds after burning would result in native seedlings experiencing less competition from invasive species before establishing. It may not be feasible to restore the entire diversity found in a reference habitat, but focusing on restoring vegetation structure will provide resilience against invasive species establishment, facilitate an appropriate fire regime and provide suitable habitat for reintroduction of rare or threatened species in future once follow-up control of invasive species is more manageable.
AFRIKAANSE OPSOMMING: Wydverspreide degradasie van ekosisteme op ‘n wêreldwye skaal het daartoe gelei dat ekologiese restourasie nodig is om bedreigde ekosisteme doeltreffend te bewaar en te bestuur. Terwyl dit 'n ontluikende gebied is bly daar groot uitdagings om optimale sukses in restourasie-intervensies te behaal. Die uitdagings moet aangespreek word om die restourasiepotensiaal van gedegradeerde terreine te te verbeter. Dit is veral belangrik in gebiede met hoë diversiteit wat dringend restourasie-intervensies benodig, waarvan die krities bedreigde vegetasietipe, Kaapse Vlakte Sandfynbos, wat binne die Kaapse Floristiese Streek (KFS) in Suid-Afrika voorkom, 'n uitstekende voorbeeld is. Van besondere belang in hierdie geval is die impak van indringerplante, veral waar indringerplantverwydering alleen nie inheemse plantegroei sal herstel nie. Die gebruik van alternatiewe passiewe skoonmaakmetodes, of die inkorporering van aktiewe saadsaai-intervensies, kan in sulke gevalle die sukses van die herstel verbeter. Hierdie proefskrif het die volgende aspekte van restourasie ondersoek: (1) saadekologie en die bepaling van effektiewe ontkiemingstoestande van struktureel belangrike spesies in hierdie veldtipe (hoofstuk 2), (2) verskillende passiewe en aktiewe restourasiebehandelings om die beste bestuursprotokolle vir uitheemse plantverwydering te bepaal (hoofstuk 3), (3) die voorspelling van langtermyn vegetasiehersteltrajekte onder verskillende restourasiebehandelings en die bepaling van ekologiese drempels deur die gebruik van 'n dinamiese model wat vegetasiehersteltempo bevat onder passiewe en aktiewe restourasiebehandelings (hoofstuk 4). 'n Saad-ekologie-studie het die effektiwiteit van rook- en hitteblootstelling ondersoek as 'n voorbehandeling van sade in die verbetering van ontkiemingsukses van spesies binne Kaapse Vlakte Sand Fynbos. Veertien spesies was aan 'n reeks tydperke van blootstellings by verskillende temperature blootgestel, in kombinasie met blootstelling aan rookresidu, waarna saad op agar ontkiem is om behandelingsukses te bepaal. Die meeste spesies het in een van die twee hoofgroepe geval: Saadkieming in die eerste groep (bv. Anthospermum aethiopicum, Metalasia densa en Watsonia meriana) was die grootste na 'n laer temperatuur hittepuls, 'n lang tydperk van ligte temperatuur blootstelling of geen voorafbehandeling met hitte. Saadkieming in die tweede groep (bv. Pelargonium elongatum, Phylica cephalantha en Trichocephalus stipularis) is na kort blootstelling aan hoër (100 ° C) temperature bevorder. Spesies in die laasgenoemde groep het hoofsaaklik fisiese dormansie gehad. Passiewe Acacia saligna-skoonmaakbehandelings in die veld wat die brand of stapeling van biomassa behels ná aanvanklike uitheemse plante verwydering sowel as aktiewe intervensie waarby die saai van inheemse spesies plaasvind word, hetsy direk na brand of 'n jaar later en óf met of sonder saadvoorbehandeling, gelei tot baie verskillende hersteltrajekte oor 'n tydperk van twee jaar. Geen behandeling het na twee jaar dieselfde gelyk as vegetasie sonder indringerplantegroei. Die verwydering van uitheemse plante sonder verbranding was die goedkoopste metode en het herstel binne minder-gedegradeerde gebiede die beste fasiliteer, maar andersins was daar baie beperkte of geen inheemse plantegroeiherstel nie en dit het sekondêre indringing deur onkruide gefasiliteer. Verbranding na die verwydering van uitheemse plante het sommige sekondêre indringerspesies beheer maar dit het ook akasia-hervestiging gestimuleer wat nie tot die herstel van inheemse plantegroei gelei het nie. Die inheemse saadbank was tydens die tydperk van indringing uitgeput, dus was daar nie outogene herstel onder passiewe behandelings nie. Alhoewel die skoonmaak sonder verbranding tot lae akasia-hervestiging gelei het, was die akasia-saadbank twee jaar later ook met en sonder verbranding verlaag. Aktiewe saadsaai was die duurste behandeling maar dit het die hoogste herstel van inheemse plantegroei en diversiteit tot gevolg gehad. Hierdie bevindinge dui daarop dat 'n biotiese drempel oorgesteek is waar passiewe behandeling nie tot die herstel van vegetasiekomponente lei nie. Aktiewe saai was in staat om hierdie drempel gedeeltelik om te keer deur verbeterde herstel van totale struikbedekking, terwyl voorbehandeling van sade voor saai, vestiging van sommige spesies verbeter. Nietemin, struikbedekking wat nie self uitloop was oorverteenwoordig, terwyl struikbedekking van self uitloop en Restionaceae-spesies, wat onder kweekhuis-toestande suksesvol ontkiem het, onderverteenwoordig was relatief tot die verwysingsvegetasie. Dit dui daarop dat daar nog struikelblokke bestaan wat oorkom moet word vir sekere spesies wat in die veld gesaai word van saad om te vestig. 'n Dinamiese model was ontwerp om die effektiwiteit van die verskillende restourasiemetodes te analiseer. Dit maak die ekstrapolasie van hersteltrajekte moontlik oor 'n lang tydperk wat nie alleen vanuit korttermyn veldopnames bepaal kon geword nie. Data vir die groeikoers van vegetasieherstel onder verskillende passiewe en aktiewe herstelbehandelings oor twee jaar van veldopnames was in die model gevoer. Die gemodelleerde simulasies het gewys dat verskillende behandelings in soortgelyke gedegradeerde state ten tye van die eerste skoonmaak tot baie verskillende hersteltrajekte gelei het wanneer dit oor 'n lang tydperk geëxtrapoleer was. Aktiewe saadsaai was aanvanklik die duurste behandeling, maar het tot die beste inheemse plantegroei gelei, wat die koste van langtermyn-opvolg uitheemse plantverwydering verminder het, asook die kompetisie vergroot het teen die vestiging van sekondêre indringerspesies. Verwydering sonder verbranding was die goedkoopste metode, maar het tot beperkte herstel van inheemse of akasia-bedekking oor die langtermyn gelei. Dit het die onbedekte grond blootgestel aan ʼn sekondêre inval deur onkruide. Die model ondersteun dus die bevinding van die veld-herstel eksperimente dat ekologiese drempels oorgesteek is wat die herstel van sekere vegetasiekomponente verhinder het. Aktiewe saadsaai kon hierdie drempels gedeeltelik omkeer en tot volhoubare struikbedekking lei, maar selfs hierdie behandeling het nie die verwysingsvegetasie weerspieël ten opsigte van die herstel van 'n balans van strukturele komponente nadat dit oor 'n lang tydperk gesimuleer is nie. Deur die kennis wat uit die drie aspekte van hierdie proefskrif verkry word, en in die lig van wat voorheen bekend was om doeltreffend te werk, is bestuursaanbevelings geformuleer om praktisyns te voorsien met meer effektiewe riglyne vir beste praktyke in die herstel van die land se fynbosplantegroei. In die eerste plek is daar tot die gevolgtrekking gekom dat biotiese drempels vir herstel in die studieterrein oorskry is weens die gebrek aan outogene herstel van natuurlike dekking en diversiteit onder passiewe behandelings as gevolg van 'n uitgeputte inheemse saadbank. Aktiewe intervensies kan hierdie drempel gedeeltelik keer deur natuurlike sade weer in te saai, en die veerkragtigheid van die plantegroei verbeter teen sekondêre indringerspesies. Tweedens, die bykomende kennis wat verkry is deur ontkiemingstoetse wat hitte- en rookvoorbehandeling van saad gebruik, is waardevol vir die inkorporering in verdere restourasiewerk. Die gebruik van spesiespesifieke voorbehandelings om optimale ontkieming te stimuleer kan die restourasie-sukses verbeter na aktiewe saadsaai in die veld. Sommige spesies wat goed op saadvoorbehandeling in kweekhuistoestande reageer, kan egter nie van saad in die veld vestig nie, en hierdie spesies word beter gevestig deur plante te kweek om in die veld te plant. Derdens is gevind dat die beste behandeling wat in die veld gebruik kan word, bepaal word deur die vlak van degradasie ten tye van aanvanklike uitheemse plantverwyder. 'n Minder gedegradeerde terrein met 'n verteenwoordigende inheemse plantegroeistruktuur, d.w.s. meer as 10% bedekking van inheemse struike sal ná verwydering sonder verbranding herstel, terwyl 'n meer gedegradeerde terrein sonder inheemse plantegroei of bedekking aktiewe ingryping benodig om die herstel van natuurlike plantbedekking, en veerkragtigheid teen primêre of sekondêre indringerspesies te fasiliteer. Aangesien verbranding na verwydering tot die herstel van Acacia saligna gelei het by beide passiewe- en aktiewe-restourasiebehandelings, word twee alternatiewe behandelings wat in die veld getoets word voorgestel. Die mees geskikte behandeling sal van die grootte van die sekondêre indringersaadbank afhang. As die saadbank klein is moet plantverwydering gedoen word sonder enige verbranding. In die geval sal die saai van voorafbehandelde inheemse sade doeltreffender wees as om te brand. As die sekondêre indringersaad bank groot is, moet daar twee jaar lank se verwydering gedoen word om die vermindering van akasia saad te vergemaklik, waarna 'n beheerde brand aangewend kan word. Die aksie sal die sekondêre indringersaadbank verwyder. Saai van inheemse sade na ‘n brand sal lei tot die vestiging van inheemse saailinge met minder mededinging van indringerspesies. Dit mag nie moontlik wees om die totale diversiteit in 'n verwysingshabitat te restoureer nie, maar om op die herstel van vegetasie te fokus sal veerkragtigheid teen indringerspesies gevestig word, 'n gepaste brandperiode fasiliteer en geskikte habitat verskaf vir die hervestiging van seldsame of bedreigde spesies in die toekoms sodra opvolgbeheer van indringerspesies meer beheerbaar is.
Thesis (PhD)--Stellenbosch University, 2018.
Ecological restoration -- Cape Floristic Region -- South Africa, Seed ecology -- Cape Flats Sand Fynbos, Ecological thresholds, UCTD