ENGLISH ABSTRACT: Current understanding of the hydraulic stability of a stacked geotextile tube structure under wave attack is limited. Failure mechanisms that lead to instability are complicated and there is, as yet, no generic approved design method. 2D physical modelling in the large wave/current flume of the Stellenbosch University was done to test various set-up and hydraulic conditions to determine the hydraulic stability of a stacked geotextile tube structure against wave attack. Sixty-five test runs of approximately 1,000 waves each were run. Modelling was done on two different scales that had good similitude, despite the fact that the same geotextile and fill material were used in both. The results provided by the physical modelling gave wave conditions larger than anticipated for hydraulic stability. It was found that the term ―failure‖ was too loosely defined in most cases and that, depending on the definition of structure failure the severity of the wave conditions at failure increased substantially. Sliding was found to be the key failure mechanism for a structure constructed from stacked, 80% sand filled, geotextile tubes. The crest tube receives the most severe loading and is the critical tube in the structure. Structures with double tube crests were found to be negligibly more stable than structures with single tube crests, but reduce energy transmission to the leeside of the structure. Impact loading of the structure combined with wave transmission over the structure explained the wave force on the crest tube of the structure. A modified Goda (1974) method incorporating a wave reduction factor for wave transmission and an angle descriptive of the crest tube position were used. The descriptive angle was derived from results obtained from the physical modelling. The use of this method provides results that correlate well with those found in the physical modelling and with results obtained in previous research. The method has the additional advantage that it is less constrained by limitations for application than those of previous studies / AFRIKAANSE OPSOMMING: Die begrip van die hidroliese stabiliteit van ‘n struktuur gebou uit gepakte geotekstielsandbuise teen golf aanval, is tans beperk. Faal meganismes wat lei tot die onstabiliteit van ‘n struktuur is ingewikkeld en daar is geen generiese aanvaarde ontwerp metode tans in gebruik nie. 2D fisiesemodellering is in die groot golfkanaal van die Universiteit Stellenbosch uitgevoer. ‘n Verskeidenheid van struktuur-uitlegte en hidroliese toestande is getoets om die hidroliese stabiliteit van die struktuur teen golf aanval te bepaal. ‘n Totaal van 65 toetse van ongeveer 1,000 golwe elk is voltooi. Modellering is op twee verskillende skale gedoen, wat goed vergelyk het ten spyte van die feit dat dieselfde geotekstiel en vul materiaal in albei gebruik is. Resultate verkry vanaf die fisiese modellering het groter as verwagte golftoestande vir hidroliese stabiliteit gegee. Dit is gevind dat die definisie van faal (mislukking) in die meeste gevalle swak beskryf is en dat, afhangende van wat as faal van die struktuur beskou word, die golftoestande aansienlik beïnvloed word. Die skuif van die buise is die hoof faal meganisme vir ‘n gepakte geotekstielbuis-struktuur met ‘n vulpersentasie van 80%. Die buis op die kruin van die struktuur word die swaarste belas en is die kritiese buis in die struktuur. Strukture met dubbel buis kruine is onbeduidend meer stabiel as dié met slegs ‘n enkele buis as kruin. Die energie wat na die lysy van die struktuur oorgedra word, is egter beduidend minder. Impak belasting van die struktuur gekombineer met golf-transmissie oor die struktuur, verduidelik die stabiliteit van die buis op die kruin van die struktuur. ‘n Gemodifiseerde Goda (1974) metode met ‘n golfverminderigs faktor word gebruik om golf-transmissie oor die struktuur te akkommodeer, saam met ‘n hoek wat beskrywend is van die posisie van die kruin buis. Die beskrywende hoek is afgelei uit resultate verkry uit die fisiesemodellering. Hierdie metode gee resultate wat goed korreleer met dié verskaf deur die fisiese modellering en die resultate van vorige navorsing oor geotekstielbuis stabiliteit. Die metode het ‗n bykomende voordeel deurdat dit minder begrens is deur beperkings m.b.t. toepassing as die van vorige studies.