Seismic velocity and reflectivity analysis of concentrated gas hydrate deposits on the southern Hikurangi Margin (New Zealand)

Highlights

• Recently acquired high-resolution seismic data and existing low-resolution industry data are presented.

• Two large concentrated hydrate deposits are identified beneath Glendhu and Honeycomb ridges.

• A novel method involving analysis of seismic velocity and reflectivity is used to obtain estimates of hydrate saturations.

• Hydrate saturations peaks of >80% are estimated locally.

• The main driving mechanism for hydrate accumulations is inferred to be along-strata gas migration.

Abstract

In the southern Hikurangi subduction margin, widespread gas hydrate accumulations are inferred based on the presence of bottom simulating reflections and recovered gas hydrate samples, mainly associated with thrust ridges. We present a detailed analysis of high- and medium-resolution seismic reflection data across Glendhu and Honeycomb ridges, two elongated four-way closure systems at the toe of the deformation wedge. High-amplitude reflections within the gas hydrate stability zone, coincident with high seismic velocities, suggest the presence of highly concentrated gas hydrate accumulations in the core regions of the anticlinal ridges. A novel method involving combined seismic velocity and reflectivity analysis and rock physics modelling is used to estimate hydrate saturations in localised areas. The effective medium model consistently predicts gas hydrate saturations of ~30% of the pore space at Glendhu Ridge and >60% at Honeycomb Ridge, whereas the empirical three-phases weighted equation likely underestimates the amount of gas hydrate present. We note that our estimates are dependent on the vertical resolution of the seismic data (5–14 m), and that the existence of thin layers hosting gas hydrate at higher concentrations is likely based on observations made elsewhere in similar depositional environments. A comparison between the two ridges provides insights into the evolution of thrust related anticlines at the toe of the accretionary wedge. We propose that the main driving mechanism for concentrated hydrate accumulation in the study area is along-strata gas migration. The vertical extent of these accumulations is a function of the steepness of the strata crossing the base of gas hydrate stability, and of the volume of sediments from which fluid flows into each structure. According to our interpretation, older structures situated further landward ofthe deformation front are more likely to host more extensive concentrated hydrate deposits than younger ridges situated at the deformation front and characterised by more gentle folding. The method introduced in this work is useful to retrieve quantitative estimates of gas hydrate saturations based on multi-channel seismic data.

Introduction

Gas hydrate systems on continental margins are relevant to a range of scientific issues related to climate change and ocean acidification (Buffett and Archer, 2004), geological hazards (Dillon et al., 1998; Mienert et al., 2005), and energy supply (Makogon et al., 2007; Johnson and Max, 2006). The upper few tens to hundred metres of marine sediment on many of the world's continental margins, typically below water depths of about 500 m, are partially saturated with natural gas hydrate, an ice-like crystalline lattice of water molecules trapping gas molecules inside (Collett et al., 2009). The gas hydrate stability zone (GHSZ) is the vertical extent within sediments over which gas hydrates can exist if there is sufficient supply of methane for hydrate formation (Sloan, 1990). Deposits of gas hydrates can occur in several forms, but they are usually concentrated within the pore spaces of coarse-grained strata, or they occur in lenses, fractures or nodules in less porous, fine-grained sediments (Clennell et al., 1999; Collett et al., 2009). Commercial production of gas hydrates as an alternative resource to conventional oil and gas has not yet been achieved, although a first offshore gas hydrate production test was carried out by Japan in 2013 (Yamamoto et al., 2014), followed by a second Japanese production test in 2017, and the first Chinese production test in 2017 (Li et al., 2018).

The lower boundary of the GHSZ, herein referred to as the base of gas hydrate stability (BGHS), is often imaged in seismic reflection data by a distinctive bottom-simulating reflection (BSR) arising from the existence of free gas beneath gas hydrates (e.g. Pecher et al., 1998; Haacke et al., 2007; Rodrigo et al., 2009; Tinivella and Giustiniani, 2013). The seismic response of hydrate-bearing sediments varies significantly according to the nature and stratigraphic architecture of sedimentary units, as well as the degree of saturation, as described in detail by Boswell et al. (2016). For instance, consider a sand unit of low seismic impedance buried between two consolidated (high impedance) mud units. With low levels of gas hydrate saturation in the sand unit, the impedance of that layer will begin to increase and approach the impedance of the neighbouring mud units, which could theoretically result in a suppression of reflectivity from the mud-sand interface (Dillon et al., 1991; Lee and Dillon, 2001; Yoo et al., 2013; Boswell et al., 2016). As saturations increase to higher levels (e.g. above 40%), the presence of hydrate in the pore space can lead to significantly higher impedance in the sand unit than that of the surrounding mud units (Boswell et al., 2016). In this case, strong positive polarity reflections from the hydrate-bearing layer are to be expected. It is also possible that the background (“hydrate free”) impedance of a gas hydrate reservoir layer is higher than surrounding layers, meaning that the reflection marking the reservoir layer is a peak rather than a trough. In such a case, any increase in hydrate saturation within the reservoir layer will simply increase its impedance and thereby increase the amplitude of the positive polarity reflection. In general, high amplitude reflections within the GHSZ, with similar polarity to that of the seafloor are expected to indicate highly concentrated gas hydrate accumulations (several tens %, Nouzé et al., 2004; Bellefleur et al, 2006; Boswell et al., 2016). If a coherent reflection can be traced from beneath the BSR to above the BSR, a clear change in polarity is often identifiable, marking the change from gas-charged strata (negative polarity) below, to gas hydrate-charged strata (positive polarity) above. Within the GHSZ, anomalous reflectivity corresponding to gas hydrate-bearing sediments should reveal positive polarity events, i.e. the same polarity as the seafloor and opposite polarity to the BSR. However, in geological settings where the sediment layers’ thicknesses are below the vertical seismic resolution, tuning effects often occur, and the identification of hydrate-bearing strata may no longer be possible based only on polarity observations.

Offshore concentrated hydrate deposits have been identified widely around the world, including in the eastern Nankai Trough (Fujii et al., 2014; Nouzé et al., 2004; Taladay et al., 2017), in the passive and active margins of the South China Sea (Berndt et al., 2019; Liu and Liu, 2018; Lin et al., 2014; Zhang et al., 2015), in the Gulf of Mexico (Boswell et al., 2009; Frye et al., 2012; Haines et al., 2017; Portnov et al., 2019), offshore South Korea (Lee et al., 2013; Ryu et al., 2013), in the Chilean margin (Rodrigo et al., 2009) and offshoreIndia (Satyavani et al., 2004; Ojha and Sain, 2007; Shankar and Riedel, 2011). Onshore evidence of highly concentrated gas hydrate deposits has been found, for example, in the Mackenzie Delta in Northern Canada (Bellefleur et al., 2006).

The southern Hikurangi Margin east of New Zealand contains a large gas hydrate province. Multi-channel seismic data reveal widespread BSRs in shallow sediments across the Hikurangi subduction margin (Katz, 1982). Moreover, locally intense fluid seepage associated with methane hydrate has been observed across the margin (Greinert et al., 2010; Watson et al., 2020). Based on regional BSR analysis, Pecher and Henrys (2003) estimated 2.285 × 10¹¹ m³ of gas hydrate is present. While most of the gas hydrate may be distributed in low concentrations, several regions along the margin have been interpreted as hosting high local concentrations (e.g. Fohrmann and Pecher, 2012; Wang et al., 2017). Glendhu and Honeycomb ridges are thrust ridges that lie at the toe of the Hikurangi accretionary wedge in water depths ranging from 2100 to 2800 m (Fig. 1). A recent synthesis of 2D industry seismic surveys allowed the identification and mapping of highly reflective features within the GHSZ on the southern Hikurangi Margin (Crutchley et al., 2018b). These data reveal both positive and negative seismic polarity features that have been interpreted as concentrated gas hydrate and free gas accumulations, respectively, within layers that are interpreted as relatively high-permeability units.

In this paper, we identify regions of anomalously high reflectivity above the BSR as accumulations of concentrated gas hydrate, following the approach set out, for example, in Boswell et al. (2016). Our over-arching objective is to provide a detailed characterisation of the gas hydrate system at Glendhu and Honeycomb ridges on the southern Hikurangi Margin. Moreover, we expand on the identification of anomalous positive polarity events by quantifying the amount of gas hydrate present in the pore space of selected sedimentary layers at the top of the concentrated hydrate zones (CHZ). To this end, we relate the seismic velocity and reflectivity of these seismic events to the gas hydrate saturation (S_H) through rock physics models for hydrate-bearing sediments (Lee et al., 1996; Dvorkin et al., 2000). Following this approach, we make first-order local estimations of gas hydrate saturation within the concentrated hydrate zones. The effective medium model for gas hydrate-bearing sediments (Dvorkin et al., 2000) systematically predicts higher gas hydrate saturations than the three-phase weighted equation (Lee et al., 1996), which likely underestimates the amount of gas hydrate present in the pore space. Our results confirm that high gas hydrate saturations (>40%, Yun et al., 2005) can cause significant seismic amplitude anomalies within the GHSZ. Moreover, we show that the combined analysis of seismic velocity and reflectivity is a reliable method to locally estimate gas hydrate saturation based on multi-channel seismic (MCS) data.