Using LiDAR derivatives to estimate sediment grain size on beaches in False Bay

Burns, James Joubert (2019-12)

Thesis (MA)--Stellenbosch University, 2019.

Thesis

ENGLISH ABSTRACT: The coastal zone is a dynamic environment that is regulated and influenced by natural ocean processes and anthropogenic phenomena. Beaches are important elements within the coastal zone and are highly valued for the ecosystem services that they provide, such as buffering against wave erosion and their recreational value. Characteristics of beaches, such as slope and grain size, need to be monitored for protection and management purposes. These continuously changing variables can indicate vulnerability to erosion, for example. South Africa’s False Bay includes a number of beaches that vary in terms of such vulnerability. Active remote sensing technologies are increasingly used as monitoring tools that allow for precise measurements of the status and changes in coastal environments. These technologies include LiDAR systems which have the capacity to record intensities which relate to surface characteristics. The aim of this study is to establish whether LiDAR derived intensity and slope of beaches can be correlated with, or serve as proxies for, sediment grain size. There are two supporting objectives. The first is to analyse airborne LiDAR data of False Bay to derive information on key physical characteristics of selected beaches and to relate these to the sand grain size of these beaches. The second objective is to conduct controlled laboratory-based analysis of sediment samples using a terrestrial LiDAR scanner, and to establish the correlations between different categories of grain size and measured LiDAR intensities. The results of the laboratory-based study reveal a strong correlation between LiDAR intensity and sediment grain size, for the individual size fractions. As grain size increases, the LiDAR intensity decreases linearly. For the natural mixed samples, a slight correlation was also detected between LiDAR intensity and average grain size; i.e. samples with a coarser average grain size generally presented lower measured intensities. Although the analysis of the airborne LiDAR data produced plausible beach slope information, the data does not reflect the expected correlation between recorded intensity and slope to describe grain size. Temporal decorrelation could be an explanation for this, since the airborne data was acquired approximately two years prior to sample collection. Interestingly, the correlations between the airborne LiDAR derived intensity and slope data contradict the widely acknowledged relationship that exists between these two variables: finer grained sediments are generally associated with flatter-sloped beaches, while coarser grained sediments are associated Stellenbosch University https://scholar.sun.ac.za iv with steeper-sloped beaches. Analysis of the airborne LiDAR data shows that intensity increases steadily with steeper slopes. This is unexpected as controlled laboratory-based measurements show that higher intensities are indicative of finer-grained sediments which are associated with flatter beaches. This study concludes that LiDAR has the potential for monitoring key physical characteristics of beaches. However, if airborne LiDAR data are to be used for such monitoring, further investigation is needed with the aim of establishing data capture and processing requirements in order to produce reliable intensity measurements. Although individual grain size fractions has been proven to strongly correlated with LiDAR intensity in a controlled environment, characterising natural beach sediments of mixed grain size fractions, using airborne LiDAR, is more difficult to accomplish .

AFRIKAANSE OPSOMMING: Die kussone is 'n dinamiese omgewing wat gereguleer en beïnvloed word deur natuurlike oseaanprosesse en menslike verskynsels. Strande is belangrike elemente binne die kussone en word hoog aangeskryf vir die ekosisteemdienste wat hulle bied, soos die beskerming teen erosie van die golwe asook die ontspanningswaarde wat strande bied. Die verskillende eienskappe van strande, soos helling en sedimentkorrelgrootte, moet gemonitor word vir beskerming en bestuursdoeleindes. Hierdie veranderlikes kan byvoorbeeld die kwesbaarheid van die strand vir erosie aandui. Suid Afrika se Valsbaai sluit 'n aantal strande in wat wissel van sulke kwesbaarheid. Aktiewe afstandswaarneming tegnologie word toenemend gebruik as moniterings gereedskap vir akkurate waarnemings van die status en veranderinge in kusomgewings. Hierdie tegnologie sluit LiDAR-stelsels in wat die vermoë het om intensiteite wat verband hou met die oppervlakkenmerke op te teken. Die doel van hierdie studie is om vas te stel of LiDAR-afgeleide intensiteit en helling van 'n strand gekorreleer kan word, of sedimentkorrelgrootte kan verteenwoordig. Daar is twee ondersteunende doelwitte. Die eerste is om LiDAR-data van Valsbaai wat vanuit die lug geneem is te analiseer om inligting oor belangrike fisiese eienskappe van sekere strande te ontleed en om dit dan met die sedimentkorrelgrootte van hierdie strande te vergelyk. Die tweede doelwit is om gekontroleerde laboratoriumgebaseerde analise van sedimentmonsters te maak deur 'n terrestriële LiDAR-skandeerder te gebruik om die korrelasies tussen verskillende kategorieë van sedimentkorrelgrootte met die ooreenkomstige LiDAR-intensiteite te bepaal. Die resultate van die laboratoriumgebaseerde studie toon 'n sterk verband tussen LiDAR intensiteit en sedimentkorrelgrootte vir die verskillende grootte klasse. Namate korrelgrootte toeneem, neem die LiDAR intensiteit lineêr af. Vir die natuurlik gemengde monsters is 'n ligte korrelasie ook tussen die LiDAR intensiteit en gemiddelde korrelgrootte opgespoor; byvoorbeeld, monsters met 'n growwe gemiddelde korrelgrootte het gewoonlik laer intensiteite getoon. Alhoewel die analise van die LiDAR-data vanuit die lug aanvaarbare helling informasie voorsien het, was daar nie 'n sterk verband tussen die aangeduide intensiteit en helling om sedimentkorrelgrootte te beskryf. Temporale verfraaiing kan wel 'n verduideliking hiervan wees aangesien daar ongeveer twee jaar tussen die versameling van die data vanuit die lug en die versameling van die monsters vir hierdie studie verloop het. Interessant egter is dat die korrelasies tussen die LiDAR-afgeleide intensiteit en hellingdata in teenstelling is met die algemene erkenning wat tussen hierdie twee veranderlikes bestaan: naamlik dat fynkorrelige sedimente in die algemeen geassosieer word met strande met ‘n kleiner, of platter, helling, terwyl grootkorrelige sedimente geassosieer word met strande met steiler hellings. Ontleding van die LiDAR-data uit die lug toon dat intensiteit geleidelik toeneem met steiler hellings. Dit is onverwags, aangesien beheerde laboratoriumgebaseerde metings toon dat hoër intensiteite op fynkorrelige sedimente dui, wat met platter strande geassosieer word. Hierdie studie het tot die gevolgtrekking gekom dat LiDAR wel die potensiaal het om die fisiese eienskappe van strande te monitor. As LiDAR-data vanuit die lug egter gebruik word vir sodanige monitering, is verdere ondersoek nodig om data-opname en verwerkingsvereistes vas te stel ten einde betroubare intensiteitsmetings te lewer. Alhoewel dit bewys is dat individuele korrelgrootte sterk gekorreleer is met LiDAR-intensiteit in 'n beheerde omgewing, is dit moeiliker om natuurlike strand-sedimente van gemengde korrelgroottes te karakteriseer.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/107062
This item appears in the following collections: