2165
Downstream Frontal Velocity Reduction Resulting from Baffles
Effets des déflecteurs dans la réduction des vitesses frontales dans un écoulement descendant.
Choi C.E., Ng C.W.W.
Department of Civil and Environmental Engineering, Hong Kong University of Science and Technology
Kwan J.S.H., Shiu H.Y.K., Ho K.K.S., Koo R.C.H.
Geotechnical Engineering Office, Civil Engineering and Development Department, HKSAR
ABSTRACT: Debris flows occur in mountainous regions during rainfall and can result in disastrous consequences to downstream
facilities if appropriate mitigation measures are not taken. Common mitigation measures include flow impeding structures within the
flow path. An array of baffles is a type of flow impeding structure used primarily to reduce the mobility of landslide debris.
However, they are usually designed on an empirical and prescriptive basis and the degree of impedance resulting from an array of
debris flow baffles is not well understood. This paper presents a series of systematic flume tests examining the influence of baffle
row number on reduction of debris frontal velocity. Photoconductive sensors installed at the base of the flume channel have been
used to measure the average frontal velocity of the debris flow downstream of the array of baffles. Results show that one row is
ineffective in reducing the debris frontal velocity. Two to three rows of baffles exhibit notable frontal velocity reduction.
RÉSUMÉ :
Les coulées de granulaires représentent un risque majeur en régions montagneuses en période de fortes pluies et peuvent
avoir des conséquences désastreuses pour les installations en aval si des mesures appropriées ne sont prises. De telles mesures
comprennent la construction de déflecteurs à matériaux sur le parcours des coulées avec pour objectif principal est de réduire la
mobilité de la masse granulaire en mouvement. Cependant, la construction de ces déflecteurs repose sur des hypothèses empiriques.
La contribution exacte des déflecteur est mal comprise à l’heure actuelle et elle est difficile à évaluer précisément. Cet article présente
une série de tests expérimentaux ayant pour objectif d’évaluer l'influence de l'augmentation du nombre de déflecteurs sur la réduction
de la vitesse frontale de la coulée granulaire. Des capteurs photoconducteurs placés dans le fond du canal de test ont été utilisés pour
mesurer la vitesse frontale moyenne des grains en aval des déflecteurs installées. Les résultats révèlent qu’au moins deux à trois
rangées de déflecteurs sont nécessaires pour avoir une réduction significative de la vitesse frontale de l’écoulement.
KEYWORDS: physical modelling; flume; debris flow; baffles; landslide mitigation measures
1 INTRODUCTION
Ground mass detached from landslide source travels down
hillside under gravitational actions. Landslide debris can have
high mobility and can result in serious consequences to
downstream facilities. Velocity of debris fronts can reach up to
30 m/s (Costa 1984; Rickenmann 1999) with peak discharges
several times greater than floods occurring in the same
catchment (Hungr et al. 1984). In order to mitigate this
hazardous phenomenon, debris-resisting structures are often
used as defence measures to retain landslide debris and impede
debris mobility (Mizuyama 2008). Defence measures may
include the rigid and flexible barriers (Wendeler et al. 2007),
levees, slit dams (Watanabe et al. 1980), and arrays of baffles.
The primary function of baffles is to impede the flow
pattern such that flow slows down after it passes through the
baffles (USFHA 2006). Baffles can be installed upstream of
barriers to reduce the impact resistance required by the barriers
and to promote lateral dispersion of flow in deposition basins
(Cosenza et al. 2006). Figure 1 shows rectangular gabion
baffles installed in front of a rigid barrier in Lantau Island,
Hong Kong. Baffles are usually designed by empirical and
prescriptive methods and their fundamental impedance capacity
is not well understood.
Similar studies on snow avalanche impeding obstacles
report that an individual row of obstacles can dissipate energy
by 20% and an additional row contributes to 10% extra energy
dissipation (Hakanordottir et al. 2001). Salm (1987) reports that
the degree of impedance can be estimated based on the
consideration of the cross-sectional blockage over the channel
area. However, the above criteria are applicable only for snow
avalanches because the flow regimes characterising avalanche
and debris flow are quite different (see Sect. 2.1 for further
details). This study aims to examine the reduction of
downstream debris frontal velocity resulting from an array of
baffles. The influence of the number of staggered rows of
baffles on downstream debris frontal velocity is investigated.
F
Figure 1. Array of baffles installed in front of a rigid barrier in Lantau
Island, Hong Kong
2 METHODOLOGY
2.1
Scaling
Three types of similitude are required for modelling debris
flow-baffle interaction; they are (i) geometric similarity, (ii)
kinematic similarity, and (iii) dynamic similarity. For
simplicity, geometric similarity is achieved by normalising
