The feasibility of using landsat TM imagery for longterm vegetation monitoring in the fynbos biome

Kotze, Johann De Flaming (1999-07)

Thesis (MScNbew) -- University of Stellenbosch, 1999.

Thesis

ENGLISH SUMMARY: Those who would successfully manage the fynbos biome require management oriented information. Managers, strategic planners and researchers often lack information about fynbos areas and changes that occurs in these areas upon which they can, with reasonable confidence, base decisions. Landsat Thematic Mapper (TM) data provides a potentially strong base to work from. The objective of this project was to determine the feasibility of using TM imagery to quantify fynbos communities within a specific area that can be used for the long term vegetation monitoring in the fynbos biome. TM imagery have advantages over other data sources such as aerial photography, for large area land cover classifications, in part because of their frequent repeat cycles, large-area sample, wide spectral range, cost effectiveness and amenability to automated classification. Considering the very detailed scale the study was done at, another objective of the project was to determine the feasibility of using TM imagery to quantify fynbos communities at a far more detailed scale than the recommended operating scale for TM imagery. At this detailed scale fynbos vegetation can be managed effectively in the long term. A small scale vegetation and vegetation community study of the study area was done to obtain a detailed ground map. The vegetation communities as defined by image processing of the TM data was compared with the accurate ground map. By means of this comparison it was shown that TM imagery could be used effectively to obtain a detailed fynbos community description of a fynbos area at the desired scale. Another objective of the study was to determine the feasibility of Principal Component Analysis (PCA) to reduce the amount of natural variation and noise within TM imagery. It has been shown that PCA can be used effectively to achieve the latter and by doing so, baseline imagery can be obtained for a specific area and time period in terms of its vegetation communities, while portraying the maximum environmental variation within the imagery. The final part of the study entailed the determination of the feasibility of using TM imagery to determine vegetation change over time. It has been shown that TM imagery can be used effectively for this purpose. This resulted in some very interesting findings, especially with the recovery of burnt fynbos areas. The objectives of the study have been met and this scale of monitoring holds much promise. Further application is warranted in different areas as well as with additional thematic data.

AFRIKAANSE OPSOMMING: Diegene wat die fynbos bioom suksesvol sal bestuur vereis bestuursgerigde inligting. Bestuurders, strategiese beplanners en navorsers ondervind gereeld tekorte aan . informasie oor spesifieke fynbos gebiede asook veranderinge wat plaasvind in hierdie areas om hulle instaat te stel om grondige besluite te neem. Landsat "Thematic Mapper" (TM) data verskaf 'n baie sterk potensiele basis om vanaf te werk. Die doelwit van hierdie projek is die bepaling van die toepaslikheid van TM data om fynbos gemeenskappe binne 'n spesifieke area te kwantifiseer en dus sodoende hierdie data aan te wend vir langtermyn montering in die fynbos bioom. TM data het voordele bo ander data bronne soos lug fotos, vir die klassifikasie van grondbedekking oor groot areas, a.g.v. die gereelde beskikbaarheid van data, wye spektrale dekking, koste effektiwiteit en die eenvoudige en effektiewe gerekenariseerde verwerkbaarheid van data. Inaggenome die gedetaileerde skaal waarop die studie gedoen is, was 'n ander objektiefvan die studie die bepaling van die vermoee van TM data om fynbos gemeenskappe te kwantifiseer op 'n ver meer gedetaileerde skaal as die normaal aanbevole skaal. Op hierdie skaal kan fynbos plantegroei effektiefbestuur word oor die langtermyn. 'n Indiepte plantegroei en plantegroeigemeenskap studie van die area is gedoen om 'n baie akkurate plantegroeikaart van die area te verkry. Plantegroeigemeenskappe soos verkry met beeldverwerking is vergelyk met die akkurate plantegroeikaart. D.m.v. hierdie vergelyking is dit bewys dat TM data effektief gebruik kan word om 'n gedetaileerde fynbos gemeenskap beskrywing van 'n area te verkry teen die verlangde skaal. Nog 'n mikpunt van die studie was om die effektiwiteit van Hoofkomponent Analise (PCA) te bepaal om natuurlike variasie en geraas in TM data te verminder. Dit is gewys dat PCA effektief gebruik kan word om laasgenoemde doelwit te beryk en sodoende is basislyn data verkry vir 'n spesifieke area en periode in terme van plantegroeigemeenskappe, terwyl die maksimum omgewings variasie behou word. In die finale gedeelte van die studie is die effektiwiteit van TM data bepaal om plantegroei verandering oor tyd te monitor. Dit is bewys dat TM data effektief aangewend kan word vir laasgenoemde. Dit het gelei tot die verkryging van baie interessante resultate, veral in die geval van die herstel van gebrande fynbos areas. Die mikpunte wat daar gestel is vir die projek is bereik en die skaal van monitering lyk baie belowend. Verdere aanwending van die resultate is geregverdig in ander fynbos areas asook met adisionele satelliet data.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/51478
This item appears in the following collections: