Hydroacoustic Signals Generated by Parked and Drifting Icebergs in the Southern Indian and Pacific Oceans
Total Page:16
File Type:pdf, Size:1020Kb
Geophys. J. Int. (2006) 165, 817–834 doi: 10.1111/j.1365-246X.2006.02911.x Hydroacoustic signals generated by parked and drifting icebergs in the Southern Indian and Pacific Oceans Jacques Talandier,1 Olivier Hyvernaud,2 Dominique Reymond2 and Emile A. Okal3 1D´epartement Analyse et Surveillance de l’Environnement, Commissariata ` l’Energie Atomique, Boıteˆ Postale 12, F-91680 Bruy´eres-le-Chˆatel, France 2Laboratoire de D´etection et G´eophysique, Commissariatal’Energie ` Atomique, Boıteˆ Postale 640, F-98713 Papeete, Tahiti, French Polynesia 3Department of Geological Sciences, Northwestern University, Evanston, IL 60201, USA. E-mail: [email protected] Accepted 2005 December 21. Received 2005 December 17; in original form 2005 March 27 SUMMARY Wereport the detection, principally by the French Polynesian seismic network, of hydroacoustic signals generated inside large icebergs, either ‘parked’ along the Wilkes coast of Antarctica in the Indian Ocean, or drifting in the Southern Pacific Ocean between latitudes of 55◦ and 65◦S, during the years 2002–2004. The signals can be classified into two very broad families, based on the nature of their spectra. A first group features prominently monochromatic signals, whose frequency can, however, fluctuate with time during a single sequence of emission (typically lasting a few to a few tens of minutes). Such signals are generally reminiscent of those detected in 2000 in the Ross Sea and are generated principally in the Indian Ocean ‘iceberg parking lot’, between longitudes 144◦E and 156◦E. A new family of signals features a much broader spectrum, superimposed on a number of preferential frequencies suggesting the background activation of a number of resonators; these signals occur both in the parking lot and in the Southern Pacific. Further variations in spectra are documented inside each family. On the basis of similar in situ observations on Ross Sea icebergs under project SOUTHBERG, the first family is generally interpreted as expressing a stick-and-slip process during collisions between large iceberg masses. The second family of signals are observed during exceptional episodes of the otherwise silent drift of the icebergs in the deep Pacific Basin, some of which correlate with their passage over the various fronts defining the oceanographic southern convergence zone. Finally, a most recent episode of activity, generally similar to the above first family, was GJI Seismology detected on 2004 December 3–4, at the ocean entry of the Dibble Ice Tongue, 600 km west of the parking lot along the coast of Antarctica. It is interpreted as resulting from collisions between large drifting icebergs and fragments of the ice tongue calved off during its disintegration, as documented by satellite imagery. Key words: hydroacoustics, icebergs, Southern Ocean, T waves. at the nearby VNDA station in Antarctica, and at stations of the 1 INTRODUCTION Mount Erebus Volcano Observatory network deployed on Ross Is- In a previous contribution (Talandier et al. 2002; hereafter Paper I), land (Aster et al. 2001, 2004). The epicentres were found to correlate we analysed hydroacoustic signals recorded as T phases in late 2000 spectacularly well with the positions of massive icebergs, notably by island-based seismic stations of the Polynesian Seismic Network B-15B, detached from the Ross ice shelf in early 2000, and drifting [R´eseau Sismique Polyn´esien; hereafter RSP] and originating from in the central Ross Sea by the end of the year. sources in the vicinity of Antarctica. Their spectral characteristics While the exact nature of the oscillator producing the hydroa- varied considerably, but usually featured prominent eigenfrequen- coustic signal, as well as the mechanism of its excitation, remained cies in the 4–7 Hz range, often accompanied by overtones. Over speculative, we suggested that the former may involve a resonance the duration of a sequence of activity (typically 2–10 min), the either of the whole iceberg mass or of water-filled cracks in the ice preferential frequency was observed to fluctuate slightly (by 10– sheet itself. As for the latter, we surmised that collisions between 20 per cent), often culminating in a slow upwards or downwards large ice masses drifting in the Ross Sea may act as triggers of evolution upon the termination of the sequence. the oscillators; this model gained support following the deployment Epicentral locations were obtained by combining Polynesian T of portable seismometers in 2003–2004 on iceberg C-16 and the phases with seismic phases interpreted as Lg and recorded both recording of strikingly similar signals during documented episodes C 2006 The Authors 817 Journal compilation C 2006 RAS 818 J. Talandier et al. ployed as part of the International Monitoring System of the Com- prehensive Nuclear-Test Ban Treaty (e.g. Lawrence 1999). In this (a) framework, it is rapidly becoming impossible to fully catalogue all observations of cryosignals, and the present paper has no pretence of completeness in this respect. Rather, we focus here on some of the most spectacular and HS-08 intriguing observations, most of them selected primarily on the HS-01 RSP basis of their amplitude, and reflecting the fluctuating detection and location capabilities resulting from variations in availabil- ity of data from various networks. The identification of hydroa- coustic records as cryosignals proceeds from a combined assess- ment of their epicentral location, of their waveforms and, when- ever possible, of the spaciotemporal correlation of their sources with the tracks of major icebergs, as documented from remote sensing. Ross Sea South Pacific Parking Lot Dibble I.T. 2000 2002-03 2002-04 Dec. 2004 Location procedure Our location procedures have been previously described in several 200˚ 210˚ 220˚ publications (Paper I; Reymond et al. 2003). Their algorithm is km based on the propagation tables of Levitus et al. (1994), adjusted MTV VAH Tuamotu (b) locally in the extreme southern part of the ocean, where a velocity -15˚ 0 400 800 Society of 1467 m s−1 is used south of 60◦S, to reflect the disappearance HAO of the SOFAR channel, and the rise of the layer of minimum sound Cook PAE TVO ANA velocity to the ocean surface. The validity of the algorithm was ver- -20˚ ified (north of the convergence zone) based on the recording at the Austral Indian Ocean hydrophones of signals from the 2003 calibration ex- RAR RKT TBI periment on board R/V Melville (D. Blackman, personal communi- Gambier cation, 2003). As discussed in detail in Talandier & Okal (1996), the -25˚ -160˚ -150˚ -140˚ times inverted by the location procedure are not arrival times, which are usually difficult to pick on emergent phenomena, but rather dif- Figure 1. (a) General reference map sketching the relative position of the ferential times between stations obtained by cross-correlating their sources of cryosignals and of receivers used in Paper I and the present study. records. (b) Detailed map of receiver stations used in Polynesia (with codes). The All events discussed here were recorded at land-based stations names of the principal island chains are shown in italics. Circles identify of the RSP (Fig. 1b), principally VAH (on the atoll of Rangiroa), permanent stations of the RSP and upward-pointing triangles temporary stations of the PLUME network; the downward-pointing triangle is the IRIS TBI and RKT (on the small coral-fringed islands of Tubuai and station at Rarotonga. Mangareva) often complemented by PAE and TVO on the large island of Tahiti, and occasionally RAR (the IRIS station on the small reefed island of Rarotonga). In addition, we frequently used records from the portable stations operated as part of the PLUME of collision and ramming of C-16 by another fragment of the original network on the Polynesian atolls of Mataiva (MTV), Hao (HAO), iceberg B-15 (MacAyeal et al. 2004). An alternative trigger could and Anaa (ANA) (Barruol et al. 2002). involve scraping against bathymetric irregularities in the continental In general, events detected only at the Polynesian stations remain shelf of the Ross Sea, whose depth may be locally comparable to poorly located, with little control on epicentral distance. This results the thickness of the floating ice sheets, estimated at ≈300 m. in elongated Monte Carlo ellipses (Wysession et al. 1991) (run with In this general framework, the present paper reports a much en- a standard deviation of Gaussian noise σ = 1 to 2 s) extending larged data set of ‘cryosignals’, that is, hydroacoustic signals orig- G several hundred kilometres, and up to 1000 km if the extreme stations inating in icebergs, mostly drifting farther North in the Southern RAR or RKT are absent. Because of the small nature of the events, Pacific, in the 50◦–65◦S latitudinal band, or ‘parked’ in the vicinity they were not recorded as seismic (or hydroacoustic) phases at the of the Wilkes Land coast of East Antarctica (Fig. 1a). These events distant seismic stations in Antarctica which had constrained the offer a variety of spectral characteristics, some of them duplicating 2000 epicentres at much shorter ranges [Paper I]. Such relatively the resonant features of the 2000 Ross Sea series, and others show- poor locations will constitute a supplemental or ‘secondary’ data ing more complex spectra, requiring a different nature of acoustic set of epicentres. Although only tentative in nature, their sources source. can nevertheless be valuable as they shed additional light on the origin of cryosignals. The situation is quite different when data are available from the 2DATASET hydrophone stations deployed in the past few years at Diego Garcia Since our initial identification of cryosignals in Paper I, a large num- (HS08) and Cape Leeuwin (HS01) in the Indian Ocean, as part of ber of additional observations have been reported (Talandier et al. the International Monitoring System. As shown on Fig.