Thesis (MSc)--University of Stellenbosch, 2005. / ENGLISH ABSTRACT: The primary aim of this study was to assess and compare the ecological integrity of the Lourens, Palmiet and
Hout Bay Rivers (South Western Cape, South Africa) by examining the macroinvertebrate community
structure at a series of representative sampling sites along the course of each river, using the South African
Scoring System - Version 5 (SASS-5) rapid bioassessment method. Secondary aims included an
examination of the effects of seasonal variability, biotope availability and site-specific environmental
variables on the macroinvertebrate community structure at sampling sites, as well as the preliminary testing
of the Integrated Habitat Assessment System (IHAS) for aquatic macroinvertebrates.
According to results obtained, the ecological integrity of sampling sites in the Mountain Stream Zone of the
three rivers was consistently good. The Hout Bay River in the upper portions of the Orange Kloof Reserve
was particularly near-pristine, with this area having been identified in this study as a potential biodiversity
‘hot-spot’ for aquatic macroinvertebrates. Downstream of the Mountain Stream Zone, there was a significant
deterioration in the ecological integrity of all three rivers due to a number of probable causes. Results based
on recorded SASS Scores and Average Score per Taxon (ASPT) values, using ‘biological bands’ generated
from reference sites in the South Western Cape, were generally similar to and supported by the
corresponding multivariate analyses undertaken. From the results of the various analyses undertaken in this
investigation and some of the problems encountered in interpreting the data, a number of recommendations
are made regarding future bioassessment studies based on the SASS within the national River Health
Programme (RHP).
To test the IHAS, secondary data were obtained from reference sites in the Mpumalanga and Western Cape
Provinces of South Africa. Assuming that SASS Scores at reference sites are the highest scores attainable,
one would expect to find a positive relationship between SASS Scores and IHAS scores at reference sites.
The assumption in this investigation was that this relationship should be linear. Non-parametric correlation
analyses were undertaken between SASS-4/5 Scores and IHAS scores, using Kendall’s Rank-correlation
Coefficient (τ), with separate analyses undertaken for different geomorphological zones and biotope groups.
Correlations between SASS Scores and IHAS scores were generally weak (τ-values mostly < 0.3) and
unsatisfactory, with no significant correlations (p < 0.05) for two-thirds of the data sets analysed and a wide
degree of scatter generally observed amongst data points in respective scatter plots. The performance of the
IHAS varied between geomorphological zones and biotope groups, with the Foothill: Gravel-bed Zone in
Mpumalanga showing the best results, particularly when the stones-in-current biotope group was analysed
separately. Further testing of the IHAS is required to confirm its relative performance in different
bioregions/ecoregions, geomorphological zones and biotope groups, which should be undertaken as a
priority research area within the RHP. Unsuccessful attempts to test the IHAS by means of multiple
regression analyses were undertaken, suggesting that such techniques should be avoided in further testing
of the IHAS. / AFRIKAANSE OPSOMMING: Die hoofdoel van hierdie studie was om die ekologiese toestand van die Lourens-, Palmiet- en
Houtbaairiviere (Suidwes Kaap, Suid Afrika) te bepaal en te vergelyk deur die bestudering van die
makroinvertebraatgemeenskapstruktuur by verteenwoordigende monsterpunte langs die riviere, met gebruik
van die “South African Scoring System” – Weergawe 5 (SASS-5) snelle biologiese bepalingsmetode.
Sekondêre doelwitte het die bepaling van die gevolge van seisoenele veranderlikheid, biotoop
beskikbaarheid en ligging-bepaalde omgewingsveranderlikes op die makroinvertebraatgemeenskapstruktuur
by monsterpunte ingesluit, asook die inleidende toetsing van die “Integrated Habitat Assessment System”
(IHAS) vir watermakroinvertebrate.
Volgens die resultate verkry, was die ekologiese toestand van monsterpunte in die Bergstroomsone van die
drie riviere konsekwent goed. Die Houtbaairivier in die boonste gedeelte van die Oranjekloofreservaat was
veral feitlik onversteurd en hierdie streek is in die studie as ‘n potensiaal biodiversiteit “hot-spot” vir
watermakroinvertebrate geidentifiseer. Stroomafwaarts van die Bergstroomsone was daar ‘n beduidende
verswakking in die ekologiese toestand van al drie riviere, as gevolg van ‘n aantal moontlike oorsake.
Resultate gebaseer op bepaalde “SASS Scores” en ‘Gemiddelde Waarde per Takson’ (“Average Score per
Taxon” - ASPT) waardes, met gebruik van ‘biologiese bande’ wat van verwysingsmonsterpunte in die
Suidwes Kaap afgelei is, was oor die algemeen soortgelyk aan en gestaaf deur die ooreenstemmende
multiveranderlikke (“multivariate”) statistiese analises wat gedoen is. Uit die resultate van die verskeie
analises wat in hierdie ondersoek gedoen is en sommige van die probleme wat in die dataverklaring gevind
is, is ‘n aantal aanbevelings gemaak met betrekking tot toekomende biologiese bepalingstudies vir die
nasionale Riviergesondheidsprogram (“River Health Programme” - RHP) wat op die SASS gebaseer is.
Om die IHAS te toets is sekondêre data van verwysingsmonsterpunte in die Mpumalanga en Wes Kaap
Provinsies van Suid Afrika verkry. As aangeneem word dat die “SASS Scores” by verwysingsmonsterpunte
die hoogste moontlike tellings is wat bereik kan word, sou ‘n positiewe verwantskap tussen “SASS Scores”
en IHAS tellings by verwysingsmonsterpunte verwag word. Die veronderstelling in hierdie studie was dat dié
verwantskap lineêr moet wees. Nie-parametriese korrelasieanalise tussen “SASS-4/5 Scores” en IHAS
tellings is gemaak, deur gebruik van Kendall se Rangkorrelasiekoëffisiënt (τ), met afsonderlike analises vir
verskillende geomorfologiese sones en biotoopgroepe verrig. Korrelasies tussen “SASS Scores” en IHAS
tellings was algemeen swak (τ-waardes < 0.3) en onbevredigend, met geen beduidende korrelasies (“p” <
0.05) vir twee-derdes van die datastelle wat geanaliseer is nie en ‘n wye verspreiding tussen datapunte in
die onderskeie “scatter plots” wat waargeneem is. Die funksionering van die IHAS was verskillend tussen
geomorfologiese sones en biotoopgroepe. Die beste resultate is vir die Voorheuwel: Gruisbeddingsone in
Mpumalanga verkry, veral indien die klippe-in-stroom biotoopgroep afsonderlik geanaliseer is. Verdere
toetsing van die IHAS is nodig om die relatiewe funksionering in verskillende biostreke/”ecoregions”,
geomorfologiese sones en biotoopgroepe te bevestig en dit behoort voorangs te geniet binne die RHP.
Pogings om die IHAS deur middel van veelvoudige regressie analise te toets het misluk, wat aandui dat
sulke tegnieke vermy moet word in verder toetsing van die IHAS.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/20937 |
| Date | 12 1900 |
| Creators | Ollis, Dean Justin |
| Contributors | Boucher, C., Esler, K.J., Stellenbosch University. Faculty of Agrisciences. Dept. of Conservation Ecology and Entomology. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | Unknown |
| Type | Thesis |
| Format | ix, 192 leaves : ill., maps |
| Rights | Stellenbosch University |
Page generated in 0.0032 seconds