Distribution and depth of bottom-simulating reflectors in the Nankai subduction margin
KeywordsSurface heat flow Bottom-simulating reflector Methane hydrate Shallow thermal structure Nankai Trough
base of gas hydrate stability
Philippine Sea plate
Integrated Ocean Drilling Program
Ocean Drilling Program
advanced piston corer temperature tool
The development of gas hydrates, the guest molecules consisting of almost pure methane (e.g., Kvenvolden 1988), has been confirmed in marine sediments using seismic reflectors, sediment cores, and downhole logging data (e.g., Shipley et al. 1979; Kvenvolden and McDonald 1985; Cook et al. 2010). The range in depth of the methane hydrates has been determined by subseafloor temperatures and pressures (e.g., Shipley et al. 1979; Dickens and Quinby-Hunt 1994). Therefore, the presence of methane hydrate can be used to obtain subseafloor thermal information by taking advantage of the hydrate’s known stability characteristics under low-temperature and high-pressure conditions.
The base of methane hydrates has been generally confirmed from acoustic reflectors called bottom-simulating reflectors (BSRs), which, in seismic reflection images, is characterized by high-amplitude reverse-polarity waveforms paralleling the seafloor (e.g., Markl et al. 1970). Because of the accumulation of methane hydrate above the BSRs (e.g., Stoll et al. 1971) and free gas below the BSRs (Miller et al. 1991; Bangs et al. 1993), BSRs are thought to correspond to the base of gas hydrate stability (BGHS), which is a phase boundary between the methane hydrate and free gas. Therefore, many previous studies have investigated the subseafloor thermal regime based on BSR characteristics (Yamano et al. 1982, 1984; Ashi et al. 2002; Harris et al. 2013). Hyndman and Wang (1993) provided constraints on the seismogenic zone with a thermal model that utilized heat flow values from BSR, probe, and borehole measurements.
The heat flow value is generally dependent on the age of the incoming plate (Stein and Stein 1992). The advection of the subducting plate results in substantially reduced heat flow from the margin wedge and forearc relative to the incoming plate. There are locally high or low heat flow observations superimposed on the regionally low heat flow on the wedge and forearc. These are thought to be mainly due to advective fluid flows (e.g., Wang et al. 1993, 1995), rapid sedimentation or erosion (Hutchison 1985; Wang et al. 1993), or fault activity (e.g., Kinoshita et al. 2011). In addition, topographic effects that arise in convex-upward and convex-downward topographies also cause locally high or low heat flows (Ganguly et al. 2000; Chen et al. 2014; Li et al. 2014). Identifying the phenomena that cause local high or low heat flows is important because surface measurements of heat flow are utilized to estimate thermal properties in plate subduction zones, including frictional strength along plate boundaries and radioactive heat production.
The Nankai Trough, formed by the northwestward subduction of the Philippine Sea Plate (PHP in Fig. 1) beneath the Eurasian plate at approximately 4 cm/year (e.g., Seno et al. 1993), is a plate boundary where earthquakes of magnitude 8 have occurred at intervals of 100–200 years (e.g., Satake 2015). Recent megathrust earthquakes with magnitude classes of 8 in the Nankai subduction zone include the Tonankai earthquake in 1944 and the Nankai earthquake in 1946 (e.g., Kanamori 1972). The geological structure in the Nankai accretionary margin can be divided from the southeast to the northwest into a deformation front, a prism slope with slope basins, an outer ridge associated with a megasplay fault, and a forearc basin. The Kumano Basin, one of the major forearc basins along the Nankai Trough, is approximately 100 km from east to west and approximately 70 km from north to south (Morita et al. 2004). The Nankai Trough is characterized by a shallow trench due to the subduction of the young plate and contains thick trench-fill sediments at a water depth of approximately 4000 m ranging from south of Suruga Bay to the northern tip of the Kyushu-Palau Ridge.
Previous heat flow studies
Many previous studies estimated heat flow values using probes and BSRs in the Nankai Trough (e.g., Yamano et al. 1982). Heat flow values gradually decreased from the trough floor to the forearc basin, as Yamano et al. (2003) estimated through a heat flow probe. Although the heat flow values tended to decrease landward, as reported above, they were also found to vary with topography on a local scale, especially in convex-upward and convex-downward seafloor regions, as indicated by Ganguly et al. (2000), who concluded that heat flow could increase by as much as 50% on an undulating seafloor. Harris et al. (2013) found broad agreement between the heat flow measured by probes and those derived from BSRs in the Nankai margin. Offshore Muroto (offshore southeast of Shikoku), high heat flow values of approximately 200 mW/m2 were observed by probes in previous studies (e.g., Yamano et al. 2003) and were thought to be caused by vigorous hydrothermal circulation within aquifers in the oceanic crust (Spinelli and Wang 2008; Harris et al. 2013, 2017).
Seismic data and BSR picking
The data for this study were derived from sixteen MCS reflection survey cruises by the research vessels (R/Vs) Kairei and Kaiyo (JAMSTEC, Japan), and the R/V Polar Princess (GC Rieber Shipping, Norway), for a total of 142 survey lines (Fig. 1). The seismic surveys referred to as KR9702 (acquired in 1997), KR9704 (1997), KR9806 (1998), KR9810 (1998), KR9904 (1999), KR0108 (2001), KR0114 (2001), KR0211 (2002), KR0413 (2004), KR0512 (2005), KR1011 (2010), KR1109 (2011), and KR1212 (2012) were conducted by the R/V Kairei; KY0314 (2003) and KY1311 (2013) were conducted by the R/V Kaiyo; and ODKMPP03 (2003) was conducted by the R/V Polar Princess. Data acquisitions from the KR9702, KR9704, KR9806, and KR9810 surveys were performed with an airgun array with a volume of ~ 66 L fired at 50-m intervals and a streamer with a length of ~ 3.5 km containing 120 receivers. Data from KR9904, KR0108, KR0114, KR0211, and KY0314 were acquired with an airgun array with a volume of ~ 200 L fired at 50-m intervals and a streamer with a length of ~ 4 km containing 156 receivers. Data from KR0413 and KR0512 were obtained with an airgun array with a volume of ~ 200 L fired at 50-m intervals and a streamer with a length of ~ 5.5 km containing 204 receivers. Data from KR1011, KR1109, KR1212, and KY1311 were obtained with an airgun array with a volume of ~ 128 L fired at 50-m intervals and a streamer with a length of ~ 6 km containing 444 receivers. Data from ODKMPP03 were acquired with an airgun array with a volume of ~ 70 L fired at 50-m intervals and a streamer with a length of ~ 6 km containing 480 receivers.
Heat flow derived from BSR depth
Heat flow is generally measured through penetration of a 4.5–6.0-m probe with temperature sensors and occasionally via in situ borehole temperature measurements in deep-sea drilling. However, BSR depths constrained by the pressure–temperature conditions have also been used as an effective method for heat flow estimation (e.g., Yamano et al. 1982). Both the probe and BSR methods have advantages and disadvantages that are related to the regions of interest. In general, the probe method can measure heat flow values at any location where the probe can penetrate the seafloor, whereas the BSR method is restricted to areas where BSRs are present, such as continental margins. In areas of bottom water temperature fluctuation, short thermal probes are more susceptible to uncertainties arising from these variations (Davis et al. 2003), unless the effect is removed through long-term observations of the bottom water temperature (Hamamoto et al. 2005). In contrast, values estimated from the BSRs are less affected by the bottom water temperature variations because, in most areas, the BSR depths are located hundreds of meters below the seafloor. Therefore, the values calculated from the BSRs can be used to estimate the geothermal regime up to approximately 1 km below the seafloor. Moreover, because the presence of BSRs in the Nankai subduction zone has been widely confirmed (Ashi et al. 2002; Baba and Yamada 2004; Otsuka et al. 2015), heat flow values can be calculated over wide areas when physical properties, such as seismic velocity, density, and thermal resistance, are estimated from seismic surveys and deep-sea drilling.
In this study, we used the relationship between the two-way travel time and the depth obtained from the check shot survey at Site C0002 of the Integrated Ocean Drilling Program (IODP) Expedition 314 (Expedition 314 Scientists 2009) to convert the two-way travel time into depth in all MCS reflection images. The IODP Site C0002 was selected because many of our survey lines were densely distributed near IODP Site C0002 (Fig. 1) and because this site afforded the most reliable P-wave velocity, including uncertainty, although some of our MCS reflection images were obtained farther away from the IODP site. The velocity in the overlying seawater column was assumed to be 1500 m/s.
We assumed hydrostatic pressure within the sediment columns above the BSRs; the differences in the results that arise from formation pressures between hydrostatic and lithostatic conditions were discussed in the following section. Using a phase diagram for methane hydrate (Tishchenko et al. 2005), the estimated pressure indicated the temperature at the BSR depth. Although other gases such as CO2 and H2S affect the phase boundary of methane hydrate (Claypool and Kaplan 1974), core samples revealed that methane hydrate in the Nankai subduction zone consists of almost pure methane (~ 99%) (Tobin et al. 2015).
The seafloor temperature was derived from the hydrographic data of the bottom water temperature from the Japan Oceanographic Data Center, which was then used in part to compute the geothermal gradient. The thermal conductivity of the formation between the seafloor and the BSR was estimated from shipboard measurements of discrete samples obtained at Site C0002 of IODP Expeditions 315 and 338 (Expedition 315 Scientists 2009; Strasser et al. 2014). Values of thermal conductivity corresponding to the subseafloor depth were adopted. As the burial depth increased, sediment compaction occurs and porosity decreased. Consequently, the P-wave velocity and thermal conductivity increased with increasing burial depth. Therefore, integrated values of thermal resistance were adopted at depths between the seafloor and the BSR. Finally, the heat flow was calculated from the geothermal gradient and the thermal resistance.
Uncertainty in BSR-derived heat flow
When we converted the two-way travel time into depth and calculated thermal resistance, we applied the physical properties from the sedimentary rocks at Site C0002 of the IODP Expeditions (Kinoshita et al. 2009) to the other BSR areas, allowing the heat flow values to be calculated wherever BSRs are recognized. However, errors may become large in cases where the physical properties of the sediment deviate greatly from those of the IODP Site C0002 (Kinoshita et al. 2009).
Results of error evaluation
1450 (in seawater)
1500 (in seawater)
1550 (in seawater)
Heat flow values calculated from BSRs have large uncertainties reaching up to 20–25% (Townend 1997; Ganguly et al. 2000; Henrys et al. 2003; Marcaillou et al. 2006; Kinoshita et al. 2011). Grevemeyer and Villinger (2001) concluded that uncertainties in heat flow from BSRs fall within 5–10% if probe measurement data can constrain the temperature at BSR depths, while estimated uncertainties can reach 50–60% if probe data are absent. This large error can be produced by picking error in seismic reflection images (Marcaillou et al. 2006) and by the discrepancy between the true value and the adopted value of the P-wave velocity, as well as the thermal conductivity of the sediment (Ganguly et al. 2000; Henrys et al. 2003). Neither double BSRs (Foucher et al. 2002; Chhun et al. 2017) nor foldback reflectors (Otsuka et al. 2015) were found in the survey lines used for our study, preventing errors in the calculation of the heat flow from the BSRs due to misinterpretation of the BSRs.
Our error evaluation showed that the heat flow value reached its maximum when the average P-wave velocity at depths between the seafloor and the BSR was set at its lowest value (Table 1). This could be easily explained by the fact that the BSR depth below the seafloor becomes shallower after the time-to-depth conversion when adopting the low P-wave velocity and because of the associated increase in geothermal gradient. Although both the thermal resistance and the P-wave velocity changed with sediment consolidation, the resultant heat flow was found to be influenced more by the change in geothermal gradient due to P-wave velocity variation than by the change in thermal resistance.
In the same manner, the heat flow exhibited a minimum value when the average P-wave velocity was set at its highest value. Because the wide range of P-wave velocities (1700–1900 m/s) ensured an appropriate error evaluation, the error of the heat flow produced by the maximum and minimum P-wave velocities must be within 11%. The different physical properties of the forearc basin at IODP Site C0002 and the prism slope at ODP Site 1178 caused an average error of 5.2%.
In addition to the error associated with the P-wave velocity of the sediments, the P-wave velocity in the seawater column was affected by temperature and salinity (e.g., Wagner and Pruß 2002; Feistel 2008). However, we considered a P-wave velocity variation of 1450–1550 m/s in the seawater column and confirmed that the heat flow values change by less than 3.9% in the studied area (Table 1). Considering the vertical resolution of the seismic reflection images (< 20 m), the BSR picking also produced an error of no more than 1%. Therefore, the total heat flow uncertainty from the BSR caused by the changes in the P-wave velocity should fall within 25% (± 12.5%) of the value presented in this study.
We also compared the BSR-derived values obtained by assuming hydrostatic and lithostatic pressure regimes to the heat flow HFAPCT3 from an advanced piston corer temperature tool (APCT3) that penetrated 159 m below the seafloor (mbsf) at Site C0002 of IODP Expedition 315 (Expedition 315 Scientists 2009; Harris et al. 2013) and the heat flow HFprobe from the probe used at the Kumano Basin near Site C0002 (Hamamoto et al. 2011). The difference (HFh − HFAPCT3)/HFAPCT3× 100 fell within 12%, and the difference (HFl − HFAPCT3)/HFAPCT3× 100 reached approximately 18%. Because the heat flow comparison may reflect different thermal conductivities, we also calculated the difference using the geothermal gradient in the same manner as the heat flow and obtained values of − 1.9 and + 2.9%, assuming hydrostatic and lithostatic pressures, respectively. In contrast, the difference (HFh − HFprobe)/HFprobe× 100 reached approximately − 16%, and the difference (HFl − HFprobe)/HFprobe× 100 reached approximately − 10%. The minimum difference was the absolute difference between the geothermal gradient value assuming hydrostatic pressure and the APCT3 geothermal gradient. Because the penetration depth of the APCT3 was much deeper than that of the probe, differences between the BSR-derived heat flow and heat flow obtained with the APCT3 were more reliable. Therefore, we constrained the thermal structure using the temperature at the BSRs by assuming hydrostatic pressure.
Correction for influence of topography
Surface heat flow values are sensitive to seafloor topography. This topographic effect was first considered in the Cascadia margin (Ganguly et al. 2000; Hornbach et al. 2012; Phrampus et al. 2017), the Costa Rica margin (Harris et al. 2010), and the Nankai Trough (Harris et al. 2011). Li et al. (2014) suggested that three-dimensional topographic effects should be taken into account to evaluate heat flow variations in the complex topographic region of Cucumber Ridge off Vancouver Island, where the dip angle locally reaches ~ 45°. Variations in heat flow can also be explained by the magnitude and pattern of two-dimensional topographic effects, although the difference between the BSR-derived heat flow and the modeled heat flow reached 20 mW/m2 over the steep slope of Cucumber Ridge (Li et al. 2014). Because our modeling sections had gentler slopes falling within 20° in the dip direction and relatively little topographic variation, within 5° (3° on average) across the modeling sections, our calculation, which considers two-dimensional topographic effects based on Blackwell et al. (1980), was also considered to provide a reasonable model for shallow thermal structures.
Parameters used for thermal modeling in the A–B cross section southwest of the Kii Peninsula
Free parameter, M
Number of topographic data, N
Distribution of BSRs
BSR depths are controlled by in situ temperatures and pressures, and a gas supply to the formation is required for the development of BSRs. Although the trough floor and the prism toe were theoretically deep enough for BSRs to develop, BSRs were generally not observed in these regions, even at great water depths. This suggested a lack of methane in these regions, owing to young horizontal trench sediments consisting of alternating permeable and impermeable layers (e.g., Ashi et al. 2002). In addition, picking BSRs that parallel the reflectors of sedimentary layers was technically challenging with low-amplitude reflectors.
Heat flow estimation
In convex-upward seafloor regions, the BSR was situated at a deeper depth; therefore, the heat flow values there were lower than in neighboring areas (Fig. 7). Rapid sedimentation causes low surface heat flow values, apparently due to a delay in recovery from the disturbed thermal equilibrium. However, sedimentation in the convex-upward seafloor regions was not likely to occur because of their geometry. Therefore, we considered that this low heat flow was caused by topography rather than rapid sedimentation. In contrast, in the convex-downward seafloor regions, the BSR was shallower and the values were higher than those in the surrounding areas (Fig. 7), apparently because rapid erosion caused high surface heat flow values. However, erosion in the convex-downward seafloor regions was not likely to occur because of their geometry. Therefore, we considered that this high heat flow was caused by topography rather than rapid erosion.
Regional BSR depth variation
The estimated BGHS depths and observed BSR depths were expected to agree well. BGHS depths considering the topographic effect (hereafter referred to as “two-dimensional BGHS depths”) were defined as depths at the hydrate-gas phase boundary (Tishchenko et al. 2005) with increasing pressure and modeled temperature from the seafloor, whereas BGHS depths without considering topographic effect (hereafter referred to as “one-dimensional BGHS depths”) were depths at the phase boundary (Tishchenko et al. 2005) with linearly increasing pressure and temperature from the seafloor. As demonstrated in the previous section, BSR-derived heat flow values varied in the regions with a convex-upward or convex-downward seafloor. However, when considering the seafloor topography of the modeling sections, the influence appeared to propagate to only several hundreds of meters in the vertical direction and several kilometers in the horizontal direction. Therefore, the heat flow was predominantly determined by the basal heat flux from the subducting plate. We, therefore, conducted the thermal modeling by taking into account the topographic effect based on the heat flux. The two-dimensional BGHS depths calculated from this thermal modeling fit well with the observed BSR depths in both the convex-upward and convex-downward seafloor regions (Fig. 9d). Although the absolute difference between one-dimensional BGHS depth and the observed BSR depth reached 180 m at a distance of 11–16 km from the trench axis (Fig. 9d), the absolute difference between the two-dimensional BGHS depth and the BSR depth fell within 50 m in this region. Considering the uncertainty in the BSR depth (± 5%), the difference of 50 m was extended to approximately 30–70 m in this region. Similar results could also be found in other investigated profiles (Additional file 1: Figures S4–S23). These results indicated that the topographic effect should be one of the fundamental considerations in heat flow estimations and that our thermal modeling accurately expressed the local heat flow at least at depths between the seafloor and the BSR.
As demonstrated in the section discussing the uncertainties associated with this study, the total uncertainty in the heat flow from the BSRs was less than 25%. We also conducted an error evaluation of the thermal structure by comparing the maximum (+ 12.5%), average, and minimum (− 12.5%) basal heat flow (or temperature gradient, \(\alpha\)) values (Fig. 9c). The uncertainty yielded temperature variations on the BSR within ± 3 °C. These temperature variations also yielded variations in two-dimensional BGHS depths falling within ± 70 m.
BSRs at the prism toe and trough floor off Tokai
We interpreted acoustic reflectors that parallel the seafloor with seismic reflection profiles showing high-amplitude reverse-polarity waveforms at the prism toe and at the trough (trench) floor as BSRs. To confirm whether these reflectors originated in methane hydrate, we calculated the theoretical depth of hydrate formation. The available heat flow values near these areas and those acquired at the trough floor with a thermal probe ranged from 80 to 100 mW/m2 (Yamano et al. 2003; Hamamoto et al. 2011). Assuming a thermal conductivity of 1.0 W/m/K, the one-dimensional BGHS depths were calculated to be 220–270 mbsf, while the observed BSR depth was 300–400 mbsf. Considering the absence of surface heat flow and P-wave velocity in the sediment very close to these areas (within distances of approximately 1 km) where BSRs were found, these reflectors could be interpreted to be methane hydrate BSRs. Artificial noises paralleling the seafloor, such as a waveform induced by airgun bubbles, were not observed throughout the region. Multiple reflectors could also be dismissed because of the discrepancy between the two depths.
Heat flow variation
The heat flow was highly variable at a distance from the trench axis along the Kii Transect (Fig. 8c). The high BSR-derived heat flow value of 97 ± 12 mW/m2 was locally obtained along the lower landward slope in the southwestern region offshore of the Kii Peninsula, where the relatively young plate (Mahony et al. 2011) was subducting. This region is located in a convex-downward seafloor and is subject to the focusing effect induced by topography. The high heat flow value was estimated to be 61 ± 8 mW/m2 after topographic correction. The difference between the topographically uncorrected and corrected values was 36 ± 20 mW/m2 (Fig. 8c). This corrected heat flow value was not anomalously high compared with the surrounding areas. Therefore, the topographic effect was a primary contributor to the very small-scale variation in heat flow in this region.
The minimum BSR-derived heat flow of 25 ± 3 mW/m2 was obtained off southern Hyuga, where the oceanic plate was deeply subducting in the Nankai margin. This region is located in a convex-upward seafloor and is subject to the defocusing effects induced by topography. The minimum value of corrected heat flow was 29 ± 4 mW/m2. The slab deepening and the topography were likely to cause the minimum heat flow value in our study area. Although the sedimentation effect appeared to cause the lower heat flow (Hutchison 1985; Wang et al. 1993), according to our seismic profile, this region of low heat flow corresponded to ridge topography rather than a sedimentary basin.
Possible causes of BSR depth anomalies other than topography
As mentioned in the previous section, the topographic effect was expected to be a primary contributor to the variation in heat flow values in convex-upward and convex-downward seafloor regions (Fig. 8c). After topographic correction, there were some sudden increases in BSR-derived heat flows throughout the modeling section, even far away from the lateral boundaries of the section (Fig. 8c, Additional file 1: Figure S24). These sudden increases may have reflected advective fluid flow where a thrust fault intersects the seafloor, although this fault cannot be confirmed from our seismic profiles. In order to obtain evidence for advective fluid flow, submersible exploration would be required. Therefore, the possibility of advection could not be excluded. Although recent erosion also causes sudden increases in heat flow, we could not confirm erosion from bathymetry features either. Moreover, sudden increases may be due to recent uplift, where thermal equilibrium had not yet been reached. Therefore, we may have actually overestimated heat flow values at some uplifted areas via topographic correction (e.g., at a distance of approximately 15 km from the trench axis in Additional file 1: Figure S14). More surveys that include measurements from probes and boreholes were needed to identify these phenomena.
The differences in BSR depths between our model and observations were large in some places. Although P-wave velocities that were different from the true velocity was one factor generating the differences, the influence was small because the BSR depths below the seafloor were shallow. For example, the error associated with BSR depths was 20 m assuming a P-wave velocity error of 100 m/s and a two-way travel time between the seafloor and the BSR of 0.4 s. We calculated the two-dimensional BGHS depths considering the 25% uncertainty demonstrated in the “Uncertainty in BSR-derived heat flow” section. In particular, we recalculated the thermal structure using the maximum (+ 12.5%) and minimum (− 12.5%) heat flows (geothermal gradient, \(\alpha\)) and calculated the two-dimensional BGHS depths from the recalculated thermal structure. We then calculated the corresponding differences between the recalculated two-dimensional BGHS and BSR. If we adopted the minimum absolute difference between the two-dimensional BGHS and BSR, most of the differences fell within 40 m (approximately one standard deviation of the differences between two-dimensional BGHS and BSR).
The differences that could not be explained by two-dimensional topographic effects, considering the uncertainty of BSR-derived heat flow, may have been due to prominent sedimentation and erosion. Sedimentation and erosion could be identified from the combination of geometries and positive and negative differences between the two-dimensional BGHS and BSR, although fluid flow, salinity variations, higher-order hydrocarbons, and three-dimensional topography could also explain the differences. If the calculated two-dimensional BGHS depths were shallower than the observed BSR depths, then the area was suggested to have suffered from the cooling effect of sedimentation. Likewise, calculated two-dimensional BGHS depths that were deeper than the observed depths suggested that the area suffered from erosion.
Sea level change also alters the depths of BSRs by inducing pressure changes in the water column (e.g., Kremer et al. 2017). However, a sea level change of 100 m produces only a 3.7% variation in heat flow at a water depth of 1500 m, assuming a uniform geothermal gradient of 40 °C/km (Expedition 315 Scientists 2009), if the BSR depths have not yet reached thermal equilibrium after the sea level change. Because the effect decreases with water depth and most of the modeling area was deeper than 1500 m, sea level change was negligible for estimations of sedimentation and erosion.
Validity and significance of our thermal modeling
Frictional heating along a subducting plate interface and radioactive heat production could change the thermal structure, even though the effective coefficient of friction was thought to be low in the Nankai Trough area (Yoshioka and Murakami 2007; Hamamoto et al. 2011; Ji et al. 2016). Most subduction zone thermal models do not consider topography with a wavelength of a few kilometers or less, as was treated by this study. In that case, the results of such a thermal model should be compared with heat flow observations that have been corrected for the small wavelength topography. The frictional heating and radioactive heat production could be estimated more precisely using a wide range of heat flow values in conjunction with our result for the shallow thermal structure because the topographic correction changed up to 36 mW/m2 of the BSR-derived heat flow in this study (Fig. 8c).
The time required to reach thermal equilibrium within subseafloor sequences through thermal diffusion is thought to be affected by the methane hydrate fraction and the latent heat of hydrate formation or dissociation (Kinoshita et al. 2011). Therefore, the sedimentation and erosion that occur within thermal equilibrium can be estimated by assuming the total hydrate fraction and an initial disturbance, such as rapid sedimentation or erosion. Although sedimentation and erosion can be estimated from anomalies in the local heat flow, erosion in convex-upward seafloor regions or sedimentation in convex-downward seafloor regions apparently mitigate local heat flow anomalies. In these areas, it is difficult to discern each phenomenon through only BSR depth data. However, comparing two-dimensional BGHS depths by considering the topographic effect in conjunction with the observed BSR depths can aid investigations of these phenomena. In summary, the shallow thermal structure that considers the topographic effect is of use in estimating the subseafloor temperature and also for understanding past surface geological phenomena, such as active sedimentation or erosion.
BSRs were present across the area, ranging from the forearc basin to the accretionary prism slope. These were verified for the first time at the accretionary prism toe and at the trough floor offshore of the Tokai region.
BSR depths below the seafloor were larger in convex-upward seafloor regions and smaller in convex-downward seafloor regions than in neighboring areas. These regional variations could be explained by the defocusing and focusing effects induced by the topography. In addition, our model helps detect past surface geological phenomena, such as prominent sedimentation or erosion.
The total uncertainty of the BSR-derived heat flow should fall within 25%, considering allowable changes in the P-wave velocity in seawater and sediment, the geothermal gradient, and thermal resistance.
AO calculated BSR-derived heat flow and thermal structure, and drafted the manuscript. HO interpreted seismic reflection profiles. AK drafted the subroutine code for calculation of BSR-derived heat flow. JA designed the study with AO. All authors contributed to the scientific discussions and manuscript inputs. All authors read and approved the final manuscript.
We are grateful to Y. Kawada for the original code for thermal modeling, J.-O. Park, crew, and scientific parties of R/V Kairei, Kaiyo, and Polar Princess for acquiring MCS data used in this study. We are also grateful to R. Harris for his fruitful comments on an early version of the manuscript. We thank anonymous reviewers for helpful comments. AK is currently funded by the Austrian Science Fund (FWF-P29678). Generic Mapping Tools (GMT) software (Wessel and Smith 1998) is used to draft some figures.
The authors declare that they have no competing interests.
Availability of data and materials
MCS reflection data conducted by R/V Kairei and Kaiyo can be obtained from Crustal Structural Database Site (https://www.jamstec.go.jp/jamstec-e/IFREE_center/index-e.html). Hydrographic observation data of bottom water temperature can be obtained freely from the Japan Oceanographic Data Center (http://www.jodc.go.jp/jodcweb/index_j.html). Other data presented in Results section of this study are available upon request by emailing the corresponding author.
Ethics approval and consent to participate
This work was partially supported by JSPS KAKENHI Grant Number 15H02988.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- Aoki Y, Tamano T, Kato S (1982) Detailed structure of the Nankai trough from migrated seismic sections: studies in continental margin geology. AAPG Mem 34:309–322Google Scholar
- Baba K, Yamada Y (2004) BSRs and associated reflections as an indicator of gas hydrate and free gas accumulation: an example of accretionary prism and forearc basin system along the Nankai Trough, off Central Japan. Resour Geol. 54:11–24. https://doi.org/10.1111/j.1751-3928.2004.tb00183.x CrossRefGoogle Scholar
- Expedition 314 Scientists (2009) Expedition 314 Site C0002. In: Kinoshita M, Tobin H, Ashi J, Kimura G, Lallemant S, Screaton EJ, Curewitz D, Masago H, Moe KT, Expedition 314/315/316 Scientists (ed) Proceedings of the Integrated Ocean Drilling Program, 314/315/316. Washington, DC, pp 1–77Google Scholar
- Expedition 315 Scientists (2009) Expedition 315 Site C0002. In: Kinoshita M, Tobin H, Ashi J, Kimura G, Lallemant S, Screaton EJ, Curewitz D, Masago H, Moe KT, Expedition 314/315/316 Scientists (ed) Proceedings of the Integrated Ocean Drilling Program, 314/315/316. Washington DC, pp 1–75Google Scholar
- Harris RN, Grevemeyer I, Ranero CR, Villinger H, Barckhausen U, Henke T, Mueller C, Neben S (2010) Thermal regime of the Costa Rican convergent margin: 1. Along-strike variations in heat flow from probe measurements and estimated from bottom-simulating reflectors. Geochem Geophys Geosyst 11:Q12S28. https://doi.org/10.1029/2010gc003272 Google Scholar
- Hutchison I (1985) The effects of sedimentation and compaction on oceanic heat flow. Geophys J R Astron Soc 82:439–459. https://doi.org/10.1111/j.1365-246X.1985.tb05145.x CrossRefGoogle Scholar
- Kinoshita M, Tobin H, Ashi J, Kimura G, Lallemant S, Screaton EJ, Curewitz D, Masago H, Moe KT, Expedition 314/315/316 Scientists (2009) NanTroSEIZE stage 1: investigations of seismogenesis, Nankai Trough, Japan. In: Proceedings IODP/315/316. https://doi.org/10.2204/iodp.proc.314315316.123
- Kremer K, Usman MO, Satoguchi Y, Nagahashi Y, Vadakkepuliyambatta S, Panieri G, Strasser M (2017) Possible climate preconditioning on submarine landslides along a convergent margin, Nankai Trough (NE Pacific). Prog Earth Planet Sci 4:20. https://doi.org/10.1186/s40645-017-0134-9 CrossRefGoogle Scholar
- Kvenvolden KA, McDonald TJ (1985) Gas hydrates of the Middle America Trench—Deep Sea Drilling Project Leg 84. In: Von Huene R, Aubouin J et al (eds) Initial Reports DSDP 84. US Govt Printing Office, Washington, pp 667–682. https://doi.org/10.2973/dsdp.proc.84.123.1985
- Marcaillou B, Henry P, Kinoshita M, Kanamatsu T, Screaton E, Daigle H, Harcouët-Menou V, Lee Y, Matsubayashi O, Kyaw Thu M, Kodaira S, Yamano M, Expedition IODP, 333 Scientific Party (2012) Seismogenic zone temperatures and heat-flow anomalies in the To-nankai margin segment based on temperature data from IODP Expedition 333 and thermal model. Earth Planet Sci Lett 349–350:171–185. https://doi.org/10.1016/j.epsl.2012.06.048 CrossRefGoogle Scholar
- Miller JJ, Lee MW, von Huene R (1991) An analysis of a seismic reflection from the base of a gas hydrate zone, offshore Peru. Am Assoc Pet Geol B 75:910–924Google Scholar
- Moore GF, Taira A, Klaus A, Party SS (2001) Proceedings of the Ocean Drilling Program. Initial Reports, vol 190. Ocean Driling. Program. College Station, TX. https://doi.org/10.2973/odp.proc.ir.190.2001
- Morita S, Ashi J, Aoike K, Kuramoto S (2004) Evolution of Kumano basin and sources of clastic ejecta and pore fluid in Kumano mud volcanoes, Eastern Nanaki Trough. In: Proceedings of the international symposium on methane hydrates and fluid flow in upper accretionary prisms. Engineering Geology Laboratory, Department of Civil & Earth Resources Engineering, Kyoto University, Kyoto, pp 92–99Google Scholar
- Shipboard Scientific Party (1975) Site 298, Initial Reports, Deep Sea Drilling Project, vol 31, pp 317–350. https://doi.org/10.2973/dsdp.proc.31.109.1975
- Shipboard Scientific Party (1986a) Site 582, Initial Reports, Deep Sea Drilling Project, vol 87, pp 35–122. https://doi.org/10.2973/dsdp.proc.87.103.1986
- Shipboard Scientific Party (1986b) Site 583, Initial Reports, Deep Sea Drilling Project, vol 87, pp 123–256. https://doi.org/10.2973/dsdp.proc.87.104.1986
- Shipley TH, Houston MH, Buffler RT, Shaub FJ, McMillen KJ, Ladd JW, Worzel JL (1979) Seismic reflection evidence for the widespread occurrence of possible gas-hydrate horizons on continental slopes and rises. Am Assoc Pet Geol B 63:2204–2213Google Scholar
- Strasser M, Dugan B, Kanagawa K, Moore GF, Toczko S, Maeda L, the Expedition 338 Scientist (2014) Site C0002. In: Proceedings of Integrated Ocean Drilling Program, vol 338. https://doi.org/10.2204/iodp.proc.338.103.2014
- Taira A, Hill I, Firth J, Shipboard Scientific Party (1991) Proceedings of the Ocean Drilling Program. Initial Reports 131. Ocean Drilling Program, College Station, TX. https://doi.org/10.2973/odp.proc.ir.131.107.1991
- Tobin H, Hirose H, Saffer D, Toczko S, Maeda L, Kubo Y, the Expedition 348 Scientists (2015) Site C0002. In: Proceedings of Integrated Ocean Drilling Program, vol 348. https://doi.org/10.2204/iodp.proc.348.103.2015
- Yamano M, Uyeda S, Aoki Y, Shipley TH (1982) Estimates of heat flow derived from gas hydrates. Geology 10:339–343. https://doi.org/10.1130/0091-7613(1982)10<339:eohfdf>2.0.co;2Google Scholar
- Yoshioka S, Murakami K (2007) Temperature distribution of the upper surface of the subducted Philippine Sea plate along the Nankai Trough, southwest Japan, from a three dimensional subduction model: relation to large interplate and low-frequency earthquakes. Geophys J Int 171:302–315. https://doi.org/10.1111/j.1365-246X.2007.03510.x CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.