Detection of sinkhole precursors through SAR interferometry
| dc.contributor.advisor | Kemp, J. | en_ZA |
| dc.contributor.author | Theron, Andre | en_ZA |
| dc.contributor.other | Stellenbosch University. Faculty of Arts and Social Sciences. Dept. of Geography and Environmental Studies. | en_ZA |
| dc.date.accessioned | 2017-03-09T06:44:22Z | |
| dc.date.available | 2017-03-09T06:44:22Z | |
| dc.date.issued | 2017-03 | |
| dc.description | Thesis (MSc)--Stellenbosch University, 2017. | en_ZA |
| dc.description.abstract | ENGLISH ABSTRACT: Differential Interferometric Synthetic Aperture Radar (DInSAR) is a mature ground deformation monitoring technique and research presented in this thesis supports this technology as a key tool for an operational sinkhole early warning system in South Africa. Sinkholes are an unpredictable geohazard that endangers life and property in dolomitic terrains. They are a significant threat in Gauteng, the most populated and urbanised province in South Africa. More than 3000 events have been recorded here in the last ~50 years that has led to the loss of 37 lives and more than 1.2 billion Rands in property damage. There is a need for risk mitigation measures in areas that are already developed. Such measures will also allow further development of sinkhole-prone land. Small-scale surface subsidence is frequently present prior to the collapse of a sinkhole yet not much is known about this phenomenon. This is mostly due to the unpredictable nature of sinkholes and the challenges in monitoring large areas for small-scale deformation. Nevertheless, it is hypothesised that the presence of precursory surface deformation can be exploited to develop early warning systems to mitigate further damages. Spaceborne DInSAR is able to monitor small-scale surface deformation over large areas and can be exploited to detect and measure precursors to sinkhole development. Recently, there have been successful case studies supporting this technique for sinkhole precursor detection. Yet the operational limitations have not been determined. The first results of DInSAR-based monitoring effort of dolomite areas associated with sinkhole development are presented here. TerraSAR-X was tasked to acquire data from January 2015 – January 2016 with short revisit times (generally 11 days but up to 77 days) resulting in 21 interferometrically compatible images. Sequential image pairs were processed through conventional DInSAR processing workflows. Three previously unknown deformation basins were detected, one of which could be confirmed in the field. This confirmed subsidence ultimately led to a burst high-pressure water supply pipeline. The detected deformation basins were between 40 m and 100 m in diameter. The maximum displacement was measured as 7 cm over 55 days. Deformation before the infrastructure damage could be detected up to 6 months in advance. The detection could have provided a viable early warning to landowners who were unaware of the subsidence. Another event that remains unconfirmed by in-situ observations was characterised by unexplained ground uplift. These results indicate that high-resolution X-band DInSAR is able to monitor dolomite-induced instability in an urban environment. Although some deformation features were observed, seven sinkhole events occurred in the observation period for which no DInSAR deformation could be detected. Two factors were identified to be major challenges to the detection of precursory deformation, 1) the minimum detectable scale of deformation and 2) coherence loss due to temporal decorrelation, mainly because of ground disturbance during large rainfall events. The physical presence of precursors is further discussed as a fundamental limitation to early warning systems. Future research should explore ultra-high resolution sensors and consider advanced DInSAR processing techniques to overcome coherence limitations. This research supports the statement that DInSAR could contribute to an operational early warning system and further recommends that local geohazard policy should consider its capabilities. Keywords: Remote sensing, SAR, Interferometry, TerraSAR-X, Sinkhole, Precursor, Dolomite, Gauteng | en_ZA |
| dc.description.abstract | AFRIKAANSE OPSOMMING: Differensiële Sintetiese Apertuur Radar Interferometrie (DInSAR) is 'n bevestigde grond vervorming monitering tegniek en navorsing in hierdie tesis ondersteun hierdie tegnologie as 'n belangrike instrument vir 'n operasionele sinkgat waarskuwingstelsel in Suid-Afrika. Sinkgate is 'n onvoorspelbare aard-gevaar wat lewe en eiendom in gevaar stel in dolomitiese terreine. Dit is 'n groot bedreiging in Gauteng, die mees bevolkte en verstedelikte provinsie in Suid-Afrika. Meer as 3000 gebeurtenisse is al hier aangeteken in die laaste 50 jaar en het gelei tot die verlies van 37 lewens en meer as 1,2 miljard Rand se skade aan eiendom. Daar is ‘n behoefte aan maatreëls om risiko te verminder in gebiede wat reeds ontwikkel is. Sulke maatreëls sal ook toelaat dat verdere ontwikkeling kan plaasvind in gebiede wat geneig is tot sinkgat vorming. Kleinskaalse oppervlak vervorming is dikwels teenwoordig voor die ineenstorting van 'n sinkgat. Daar is egter nog nie veel bekend oor hierdie verskynsel nie, deels as gevolg van die onvoorspelbare aard van sinkgate en die uitdagings om groot gebiede vir kleinskaalse (<10cm) vervorming te moniteer. Nogtans word dit veronderstel dat die teenwoordigheid van voorloper oppervlak vervorming benut kan word om vroeë waarskuwingstelsels te ontwikkel om verdere skade te beperk. Ruimte gebaseerde DInSAR is in staat om kleinskaalse oppervlak vervorming oor groot gebiede te monitor en kan benut word om voorlopers van sinkgat ontwikkeling op te spoor en te meet. Onlangse suksesvolle gevallestudies ondersteun hierdie tegniek vir sinkgat voorloper opsporing maar operasionele beperkings is nog nie bepaal nie. Die eerste resultate van DInSAR-gebaseerde monitering poging van dolomiet areas wat verband hou met sinkgat ontwikkeling word hier aangebied. TerraSAR-X is getaak om data te verkry vanaf Januarie 2015 - 2016 met kort herbesoek tye (oor die algemeen 11 dae, maar tot 77 dae) wat gelei het tot 21 interferometriese versoenbare beelde. Opeenvolgende beeld pare is verwerk deur middel van konvensionele DInSAR metodes. Drie voorheen onbekende vervorming areas is opgespoor, waarvan een in die veld bevestig is. Hierdie bevestigde grondvervorming het uiteindelik gelei tot die bars van ‘n hoë druk watervoorsiening pyplyn. Die vervormende areas wat ontdek is, was tussen 40 m en 100 m in deursnee. Die maksimum verplasing wat gemeet was, was 7 cm oor 55 dae. Vervorming kon opgespoor word tot 6 maande voor die infrastruktuur skade. Die opsporing kon 'n lewensvatbare vroeë waarskuwing gebied het aan grondeienaars wat nie bewus was van die versakking nie. Nog 'n gebeurtenis, wat nie bevestig is deur in-situ waarnemings nie, was gekenmerk deur onverklaarbare grond opheffing. Hierdie resultate dui daarop dat hoë-resolusie, X-band, DInSAR in staat is om onstabiliteit, as gevolg van dolomiet, te monitor in 'n stedelike omgewing. Maar, alhoewel sommige vervorming eienskappe wel waargeneem is, het daar sewe sinkgat gebeure plaasgevind gedurende die waarneming tydperk waarvoor geen vervorming opgespoor kon word nie. Twee faktore is geïdentifiseer as groot uitdagings vir die opsporing van voorlopers van vervorming, insluitend 1) die minimum waarneembare skaal van vervorming en 2) verlies van sein korrelasie as gevolg van oppervlak verandering - hoofsaaklik as gevolg van grond versteuring tydens swaar reënval tydperke. Die fisiese voorkoms van sinkgat voorlopers word verder bespreek as 'n fundamentele beperking vir vroeë waarskuwing stelsels. Toekomstige navorsing behoort ultra-hoë resolusie sensors te verken en moet oorweeg om meer gevorderde DInSAR verwerking tegnieke te gebruik om korrelasie beperkinge te bowe te kom. Hierdie navorsing ondersteun die stelling dat DInSAR kan bydra tot 'n operasionele vroeë waarskuwingstelsel en beveel verder aan dat plaaslike aard-gevaar beleid die nuwe tegnologie in ag moet neem. Trefwoorde: Afstandswaarneming, SAR, Interferometrie, TerraSAR-X, Sinkgat voorloper, Dolomiet | af_ZA |
| dc.embargo.terms | 2018-03-31 | |
| dc.format.extent | 139 pages : illustrations | en_ZA |
| dc.identifier.uri | http://hdl.handle.net/10019.1/100668 | |
| dc.language.iso | en_ZA | en_ZA |
| dc.publisher | Stellenbosch : Stellenbosch University | en_ZA |
| dc.rights.holder | Stellenbosch University | en_ZA |
| dc.subject | Remote sensing | en_ZA |
| dc.subject | Interferometry | en_ZA |
| dc.subject | Sinkhole | en_ZA |
| dc.subject | Dolomite | en_ZA |
| dc.subject | UCTD | |
| dc.title | Detection of sinkhole precursors through SAR interferometry | en_ZA |
| dc.type | Thesis | en_ZA |
Files
License bundle
1 - 1 of 1
Loading...
- Name:
- license.txt
- Size:
- 1.95 KB
- Format:
- Item-specific license agreed upon to submission
- Description: