Thesis (MSc)--Stellenbosch University, 2001. / ENGLISH ABSTRACT: The burgeoning problem of alien plant invasions in South Africa necessitates effective
decision-making based on an understanding of the complex processes that govern these
invasions. Due to the spatial nature of the problem, this study explored the use of Geographic
Information Systems and spatial models for predicting the spread of alien vegetation and
assessing the effectiveness of clearing strategies. The Spatially Explicit Individual Based
Simulation (SEIBS) model was identified as a potentially useful tool for alien plant
management. This thesis documents the further investigation and development that was
necessary before recommendations could be made regarding the future use of the model.
The landscape version of the SEIBS model was adapted to allow for the convenient input and
output of spatial data, making it possible to simulate invasions in different areas. An ArcView
extension was developed in order to facilitate the pre- and post-processing of the spatial data
required and created by the model. Changes were also made to the fire routine of the model.
The new version of the model was called Clear.
A series of model tests for Pinus pinaster were conducted to assess the sensitivity of the Clear
model to spatial resolution, initial spatial fragmentation and heterogeneity. These tests
revealed that the model was sensitive to changes in resolution and needed to be
reparameterised when using different resolutions. The initial level of fragmentation was
shown to have a major influence on the invasion rate. Although greater levels of spatial
heterogeneity with respect to vegetation age did not significantly affect the spread rate, it did
increase the effectiveness of clearing strategies based on clearing juvenile or sparse
vegetation.
Based on these tests, it is concluded that the model can be readily applied to different areas,
provided the influence of spatial characteristics is understood and accommodated. The Clear
model was shown to be a useful tool for evaluating clearing strategies and for investigating
invasion rates. It is recommended that the model be introduced to a wider audience, in order
to obtain user feedback and further improve the accessibility of the model. / AFRIKAANSE OPSOMMING: Die toenemende probleem van uitheemse indringerplante in Suid Afrika, noodsaak effektiewe
besluitneming wat gebaseer is op 'n begrip van die ingewikkelde prosesse wat indringing
beheer. As gevolg van die ruimtelike geaardheid van die probleem, is die gebruik van
Geografiese Inligtingstelsels en ruimtelike modelle vir die voorspelling van die verspreiding
van indringerplante en die evaluasie van die effektiwiteit van opruimingstrategieë in hierdie
studie ondersoek. Die Spatially Explicit Individual Based Simulation (SEIBS) model is as 'n
moontlike geskikte hulpmiddel vir die bestuur van uitheemse indringerplante geïdentifiseer,
alhoewel verdere ondersoeke en ontwikkeling nodig was voordat aanbevelings vir die gebruik
van die model gemaak kon word.
Vir hierdie studie is die landskapweergawe van die SEIBS model aangepas om die maklike
toevoer en afvoer van ruimtelike data te fasiliteer. 'n ArcView uitbreiding is ontwikkel om
met die voor- en naprosessering van ruimtelike data, wat deur die model gebruik en geskep is,
te fasiliteer. Veranderinge is ook aan die vuur sub-roetine van die module gemaak. Die nuwe
weergawe van die model word Clear genoem.
'n Reeks toetse is vir Pinus pinaster gedoen om die sensitiwiteit van die Clear model te toets
teenoor ruimtelike resolusie, aanvanklike vlak van versnippering en vlak van heterogeniteit.
Vanuit die toetse het dit geblyk dat die model sensitief was ten opsigte van verandering in
resolusie en dat die model se parameters verstel moes word wanneer verskillende resolusies
gebruik word. Daar is ook gewys dat die die vlak van aanvanklike versnippering 'n groot
impak op die verspreidingstempo het. Alhoewel hoër vlakke van ruimtelike heterogeniteit
teenoor plantegroei nie 'n merkbare impak op die verspreidingstempo gehad het nie, het dit
wel die effektiwiteit van opruiming-strategieë, gebaseer op die opruiming van jong of yl
verspreide plante, verbeter.
Die gevolgtrekking wat uit die toetse gemaak kan word is dat die model geredelik op
verskillende areas toegepas kan word, op die voorwaarde dat die invloed van ruimtelike
eienskappe in ag geneem word en in berekening gebring word. Dit word aanbeveel dat die
model wyer bekendgestel word om sodoende gebruikersterugvoer te bekom.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/52341 |
| Date | 12 1900 |
| Creators | Schonegevel, Lucille |
| Contributors | Van Niekerk, A., Stellenbosch University. Faculty of Arts and Social Sciences. Dept. of Geography & Environmental Studies. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | Unknown |
| Type | Thesis |
| Format | 61 pages: illustrations |
| Rights | Stellenbosch University |
Page generated in 0.0022 seconds