Allin et al. Eustatic sea-level controls on the flushing of a shelf-incising submarine canyon Joshua R. Allin1,2,†, James E. Hunt1, Michael A. Clare1, and Peter J. Talling3 1National Oceanography Centre, University of Southampton Waterfront Campus, European Way, Southampton, SO14 3ZH, UK 2School of Ocean and Earth Sciences, University of Southampton Waterfront Campus, European Way, Southampton, SO14 3ZH, UK 3Departments of Earth Sciences and Geography, University of Durham, Durham DH1 3LE, UK ABSTRACT in other basin turbidite records. The log- change is regarded as a dominant control on normal distribution of turbidite recurrence submarine fan and canyon development by alter- Turbidity currents are the principal pro- intervals seen in the Iberian Abyssal Plain is ing the location of sediment deposition relative cesses responsible for carving submarine demonstrated to result from the variable run- to the shelf edge, thereby limiting its delivery to canyons and maintaining them over geologi- out distance of turbidity currents, such that the deep ocean by mass transport processes cal time scales. The turbidity currents that distal records are less complete, with possible (“lowstand model”; Vail et al., 1977; Shan- maintain or “flush” submarine canyons are influence from diverse sources or triggering mugam and Moiola, 1982; Posamentier and some of the most voluminous sediment trans- mechanisms. The changing form of turbidite Vail, 1988; Piper and Savoye, 1993; Ducassou port events on Earth. Long-term controls on recurrence intervals at different locations et al., 2009; Lebreiro et al., 2009; Covault and the frequency and triggers of canyon-flushing down the depositional system is important be- Graham, 2010). However, whether or not factors events are poorly understood in most canyon cause it ultimately determines the probability like eustatic sea-level change are a control on the systems due to a paucity of long sedimentary of turbidity current–related geohazards. recurrence rates of large-volume landslides and records. Here, we analyzed a 160-m-long turbidity currents worldwide has little empirical Ocean Drilling Program (ODP) core to de- INTRODUCTION support, based upon few well-dated examples termine the recurrence intervals of canyon- that do not provide the required statistical power flushing events in the Nazaré Canyon over Turbidity currents are among the most volu- for testing (Urlaub et al., 2013; Pope et al., 2015). the last 1.8 m.y. We then investigated the role metrically important sediment transport mecha- Furthermore, it is also unclear whether or not sig- of global eustatic sea level in controlling the nisms operating on Earth’s surface. An individual nals of environmental change that propagate into frequency and magnitude of these canyon- turbidity current can be capable of transporting deep water (>4000 m) are recorded and preserved flushing events. Canyon-flushing turbidity as much sediment as all the world’s rivers in in distal marine sedimentary archives (Covault currents that reach the Iberian Abyssal Plain one year combined (Talling et al., 2007; Korup, and Graham, 2010; Romans and Graham, 2013; had an average recurrence interval of 2770 yr 2012). The most voluminous turbidity cur- Allin et al., 2016; Romans et al., 2016). over the last 1.8 m.y. Previous research has rents are triggered by large (>1 km3) submarine documented no effect of global eustatic sea landslides originating from continental slopes FLUSHING OF SUBMARINE CANYONS level on the recurrence rate of canyon flush- and volcanic islands. These large landslides, ing. However, we find that sharp changes and their often associated turbidity currents, Turbidity currents in submarine canyons are in global eustatic sea level during the mid- pose considerable geohazard risk and have the proposed to be one of two broad end-member Pleistocene transition (1.2–0.9 Ma) were asso- poten tial to generate tsunamis that can damage types: those that are restricted to, and fill or ciated with more frequent canyon-flushing coastal settlements and cause considerable loss recharge the canyon with sediment, and those events. The change into high-amplitude, of life (Bondevik et al., 1997; Tappin et al., 2008; that flush sediment from the canyon and con- long-periodicity sea-level variability dur- Harbitz et al., 2006). Landslides and turbidity tinue into deeper water (Parker, 1982; Piper and ing the mid-Pleistocene transition may have currents may also damage expensive submarine Savoye, 1993; Canals et al., 2006; Piper and remobilized large volumes of shelf sediment infrastructure, such as pipelines and telecom- Normark, 2009; Talling et al., 2012; Allin et al., via subaerial weathering, and temporarily munication cables (Bruschi et al., 2006; Carter 2016). “Filling” turbidity currents triggered by increased the frequency and magnitude of et al., 2012, 2014; Pope et al., 2016). For these localized failures, hyperpycnal river discharge, canyon-flushing turbidity currents. Turbidite reasons, understanding the triggering mecha- or storms accumulate sediment within can- recurrence intervals in the Iberian Abyssal nisms and long-term frequencies of volumetri- yons over hundreds or even thousands of years Plain have a lognormal distribution, which is cally large submarine landslides and turbidity (Paull et al., 2005; Canals et al., 2006; Arzola fundamentally different from the exponential currents is important for geohazard assessment. et al., 2008; Khripounoff et al., 2009; Masson distribution of recurrence intervals observed Nonrandom processes like climate-driven et al., 2011; Talling et al., 2012; Talling, 2014). sea-level change are proposed to be an important “Flushing” turbidity currents can be defined as †Present address: School of Geography and Envi- control on the recurrence rates of large-volume the infrequent (>100 yr to >1000 yr) and large- ronment, Oxford University Centre for the Environ- landslides and turbidity currents (Maslin et al., scale (partial) erosion of unconsolidated sedi- ment (OUCE), University of Oxford, South Parks Road, Oxford OX1 3QY, UK; joshua .allin@ouce .ox 2004; Owen et al., 2007; Brothers et al., 2013; ments within a submarine channel by turbulent .ac .uk. Smith et al., 2013). Similarly, eustatic sea-level or cohesive sediment gravity flows, which then GSA Bulletin; January/February 2018; v. 130; no. 1/2; p. 222–237; https://doi.org/10.1130/B31658.1; 13 figures; 1 table; published online 29 August 2017. 222 Geological Society of America Bulletin, v. 130, no. 1/2 © 2017 The Authors. Gold Open Access: This paper is published under the terms of the CC-BY license. Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/130/1-2/222/4013594/222.pdf by guest on 01 October 2021 How do multiple sea-level cycles affect the flushing of a submarine canyon? deposit the sediment on slope fan lobes and dominant during transgressions and highstands (3) identify the distribution form, or shape, of distal basin plains (Piper and Normark, 2009). have been documented (Piper and Savoye, turbidite recurrence interval data. The distribu- Criteria for identifying canyon-flushing events 1993; Covault and Graham, 2010; Covault and tion form of turbidite recurrence data may yield in the depositional record include volumes of Fildani, 2014). In addition, several authors have information about possible triggering mecha- >0.2 km3, the presence of erosional hiatuses noted that many deep-marine successions dis- nisms of canyon-flushing events in Nazaré within canyons and channels, and lateral con- play little or no statistically detectable order, and Canyon, as well as the ways in which different tinuity of turbidites within a canyon-fed basin they have argued for more rigorous statistical distribution forms arise in depositional systems. (Paull et al., 2005; Talling et al., 2007; Piper approaches to complement qualitative assess- and Normark, 2009; Masson et al., 2011; ments of stratal organization (Wilkinson et al., GEOLOGICAL SETTING Talling, 2014). 2003; Sylvester, 2007; Chen and Hiscott, 1999). Turbidity currents that fill, or recharge, Furthermore, uncertainties in stratigraphic age The Iberian Abyssal Plain is located 200 km Nazaré Canyon have previously been analyzed control can make distinguishing order from off the western coast of Portugal between 40°N using sediment cores obtained from the canyon randomness in event recurrence particularly and 43°N. It extends ~700 km to the northwest levees. These filling turbidity currents are pre- challenging (Urlaub et al., 2013; Pope et al., at an average water depth of 5300 m (Fig. 1). dominantly active during sea-level lowstand, 2016). The paucity of robust statistical analy- The basin is bounded by the Galicia Bank to the and their recurrences conform to a normal distri- ses of stratal patterns and turbidite recurrence northeast, the Estremadura Spur to the south, bution (Allin et al., 2016). Larger turbidity cur- limits our understanding of eustatic control over and by a series of seamounts along its west- rents that flush Nazaré Canyon periodically have sedimentation in deep water and highlights the ern margin. The total area of the basin covers been inferred from thick (>20 cm) turbidites in need to more rigorously test the applicability ~107,000 km2 (Weaver et al., 1987). ODP Leg the central Iberian Abyssal Plain, 140 km from of sequence stratigraphic models to individual 149 initial reports have detailed the long-term the mouth of the canyon. In the last 80,000 yr, depositional systems