A river health assessment of selected Southwestern Cape rivers : riparian vegetation index (RVI) application, evaluation and alternatives

Withers, Melanie Jane (2003-12)

Thesis (MSc)--University of Stellenbosch, 2003.

Thesis

ENGLISH ABSTRACT: This study focuses on riparian vegetation condition at sites along three nvers within the southwestern Cape, South Africa namely the Lourens, Hout Bay and Palmiet Rivers. The Riparian Vegetation Index (RVI) is used as a tool to assess riparian vegetation condition. Data from this study will be used in State of Rivers reporting for the River Health Programme in South Africa. This study highlights sites of high conservation status with regards to vegetation condition and indicates sites of concern due to poor vegetation condition. Deviations from expected vegetation condition scores at certain sites along the rivers have highlighted certain flaws in the RVI method. Further, the repeatability and validity of the RVI method is tested. The method is tested for variability and validity across disturbance gradients at sites along the Eerste River. Results indicate that the RVI method produces repeatable results. The sub-index, Recruitment of Indigenous Riparian Species (RIRS), is the most variable in terms of assessor scores, while Structural Intactness (SI) is the most reproducible sub-index. This has important implications for the future development of the method, which should place more emphasis on reproducible sub-indices within the final RVI formula and should standardise those sub-indices that are highly variable. Statistical analysis indicates that the method does not reflect the perceived biological status of the vegetation. This may be due to the specified sampling duration, which should be increased in order to reflect more meaningful results. Criticisms of and difficulties with the RVI method as well as suggestions to improve the method are dealt with and an alternative method to the Riparian Vegetation Index is explored. The Floristic Quality Assessment (FQA) method, used to determine vegetation integrity in Illinois USA, is adapted to assess riparian vegetation condition at sites along the Lourens, Hout Bay and Palmiet Rivers. Coefficients of conservatism, determined for the flora observed in the assessment, attempt to categorise species according to their response to levels of habitat degradation. These values are incorporated into the Floristic Quality Index (FQI) formula, which yields the final floristic quality assessment results. Correlation analyses indicate that RVI and FQA results are strongly correlated although the RVI at this stage best represents the perceived biological status of vegetation at sites. The FQA method, or a statistically modified alternative, may be a suitable and more objective alternative to the RVI method.

AFRIKAANSE OPSOMMING: Hierdie studie fokus op die kondisie van die oewerplantegroei van verskeie studie gebiede langs drie riviere in die suidwes Kaap, naamlik die Lourens-, Houtbaai-en Palmietriviere. Die 'Riparian Vegetation Index' (RVI) is gebruik om die kondisie van die oewerplantegroei te ondersoek. Resultate van die studie sal gebruik word in 'State of Rivers' verslaggewing vir die Suid-Afrikaanse Rivier-Gesondheidsprogram (RHP). Hierdie studie op grond van oewerplantegroei, beklemtoon areas met 'n hoë bewaringsstatus en dui ook gebiede aan wat aandag vereis as gevolg van swak oewerplantegroei kondisie. Afwykings van verwagte plantegroei kondisiewaardes by sommige studie areas langs die riviere, beklemtoon sekere swakpunte in die RVI-metode. Verder, IS die herhaalbaarheid en geldigheid van die RVI-metode ondersoek op versteuringsgradiente langs die Eersterivier. Resultate toon dat die RVI-metode wel herhaalbare data produseer. Dit is gevind dat die sub-indeks 'Recruitment of Indigenous Riparian Species' (RIRS) die meeste varieer in terme van assessor-waardes terwyl die 'Structural Intactness' (SI) die mees herhaalbare sub-indeks is. Dit het belangrike implikasies vir die toekomstige ontwikkeling van die metode, wat herhaalbare sub-indekse in die finale formule van die RVI moet beklemtoon en varieerende sub-indekse moet standardiseer. Statistiese analises dui aan dat die RVI-metode nie die waargenome biologiese status van die plantegroei reflekteer nie. Dit mag die gevolg wees van die tyd wat gespesifiseer word vir dataversameling. Meer beduidende resultate mag egter verkry word as die gespesifiseerde tydsduur verleng word. Kritiek en struikelblokke met betrekking tot die RVI-metode, asook voorstelle om die metode te verbeter, word behandel. 'n Alternatiewe metode om die toestand van oewerplantegroei te bepaal is ook ondersoek naamlik die 'Floristic Quality Assessment' (FQA) metode, wat ontwikkel is om die integriteit van plantegroei in Illinois in die VSA te bepaal. Hierdie metode is aangepas om die toestand van oewerplantegroei langs die Lourens-, Houtbaai-en Palmietriviere te ondersoek. 'n Konserwatiwiteits-koeffient, wat bepaal is om die waargenome plantegroei te evalueer, poog om spesies te kategoriseer op grond van hul reaksie op verskillende vlakke van habitatdegredasie. Hierdie waardes word geinkorporeer in die 'Floristic Quality Index' (FQI) formule wat die finale kwaliteit van die plantegroei aandui. Analises toon 'n sterk korrelasie tussen resultate verkry deur beide die RVI- en FQA-metodes. Huidiglik koreleer die RVI egter sterk met die waargenome biologiese status van die plantegroei die beste. Die FQA-metode of andersins 'n statistiese gemodifiseerde alternatief daarvan mag egter meer gepas en objektief blyk te wees as die RVImetode.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/53291
This item appears in the following collections: