Two-way time–depth relationship for Waka Nui-1 well to aid geological interpretations of the offshore northwestern New Zealand

Abstract

The Waka Nui-1 petroleum exploration well, about 90 km west of Northland, is the only offshore well north of Auckland and provides important lithologic and age control for interpretations of seismic reflection surveys in the petroleum-prospective Northland and Reinga basins. It is especially important because it intersects a nearly complete Cenozoic sequence and Mesozoic Murihiku Supergroup rocks, all of which may be part of the petroleum system in these basins. Correlation with seismic data is uncertain because neither sonic logs nor check shot data were acquired during drilling operations. This short communication documents the development of a two-way time– (TWT–) depth relationship derived by careful analysis of stacking velocities from nearby petroleum industry seismic data and logging-while-drilling logs from Waka Nui-1. Using this correlation, the age and lithology data from the well can be used to infer the age of unconformities and seismic sequences on seismic data in the Northland and Reinga basins. These interpretations provide control on the timing of structural events and the evolution of depositional processes in this region of extensive sedimentary accumulation and are vital for predicting the prospectivity of the basins.

Keywords:

• Northland Basin
• petroleum
• Reinga Basin
• seismic velocities
• Waka Nui-1

Introduction

The Waka Nui-1 wildcat petroleum well lies about 90 km west of Kaipara Harbour in Northland and is the only offshore well north of Auckland (Fig. 1). Information from the well is critical for tying seismic data and constraining interpretations of the tectonic history and depositional environment in the entire offshore region northwest of New Zealand, including the petroleum-prospective Northland and Reinga basins. This is an area of approximately 140,000 km² with sedimentary accumulations up to 8 km thick (e.g. Isaac et al. 1994; Uruski et al. 2008; Stagpoole et al. 2009). Because of the importance of the well for stratigraphic and structural control, most of the recent seismic surveys in the Reinga, Northland and deepwater Taranaki basins included lines that tie to Waka Nui-1.

Figure 1  Location map of Waka Nui-1 and some of the more modern seismic data (red lines) tied to the well.

Early seismic interpretations of the region (e.g. Isaac et al. 1994; Herzer 1995; Gage & Kurata 1996; Herzer et al. 1997) were tied to wells in the Taranaki Basin. Regional depth conversion for these interpretations (e.g. Herzer 1992; Isaac et al. 1994) used stacking velocities, derived during processing of seismic data, to determine sediment thickness. Waka Nui-1 should have provided robust stratigraphic and structural ties to seismic interpretations, significantly better than ties to wells several hundred kilometres away in northern Taranaki Basin. However, no velocity data are available from the well and therefore ties to seismic data from the region are uncertain because of the ambiguities in correlating units on seismic data with formations in the well. Indeed, some recent seismic interpretations in the Northland region still rely on a two-way time– (TWT–) depth relationship developed from check-shot data from northern Taranaki Basin wells (Klue 2008; Johnston 2009).

This short communication describes analysis of the data from the Waka Nui-1 well and proximal seismic data and proposes a TWT–depth relationship that can be used to extrapolate stratigraphic and structural data from the well to seismic data in the Northland and Reinga basins.

Well data

The Waka Nui-1 well was spudded on 30 April 1999 and was drilled in 1455 m of water (Milne & Quick 1999). The well reached a total depth of 3657 m bmsl (depth below mean sea level). No oil shows and only trace amounts of gas were recorded, so the well was plugged and abandoned on 25 June 1999. Down-hole lithology, stratigraphy and logging while drilling (LWD) were recorded (Fig. 2). No post-drilling wireline logs are documented in the well completion report (Milne & Quick 1999). Available well logs are natural gamma ray, resistivity, photo-electric effect, density and neutron porosity. Neither sonic logs nor check-shot data are recorded from Waka Nui-1.

Figure 2  Summary of Waka Nui-1 well data showing gamma ray (GR), resistivity (LWD), density (RHOB) and neutron porosity (TNPH) logs, lithology and stratigraphy (from Waka Nui-1 Well Sheet 1:10,000, version 3.06.2008, Enclosure 1 of Stagpoole et al. 2009).

The well was drilled to test Cretaceous coal measures interpreted to be draped across the dip slope of a large tilted fault block. The upper part of the drilled sequence largely matched the well prognosis, but the predicted Cretaceous coal measures were not encountered. The coaly succession identified from the seismic data and encountered at the bottom of the hole are Middle Jurassic in age. It is almost certain that these rocks are part of the Murihiku Supergroup (Isaac et al. 1994; Campbell et al. 2003). The Jurassic coal measures are immediately overlain by Palaeogene transgressive strata (3440–3520 m bmsl), an unconformity representing about 100 million years (Uruski & Stagpoole 2004).

Two-way time–depth analysis

Integrating seismic interpretations with well data requires correlation of TWT on the seismic sections with depth in the well. This is done by converting the velocity and thickness of the strata in the well to TWT. The conversion is usually based on data from sonic logs that measure the seismic velocity of strata in the well, or from check-shot data that measure the travel time of seismic waves in strata adjacent to the well (Sheriff & Geldart 1995). Both sonic logs and check-shot data are typically acquired during drilling operations.

Because of the lack of velocity data in the Waka Nui-1 well, the TWT–depth analysis developed here is based on seismic stacking velocities from the CNL95-034 seismic line that passes directly over the well site, and from the more recent STRATOS-09 line that passes within about 2 km of the well site. Both are good quality seismic lines, recorded with hydrophone streamers > 6000 m long and a source array with > 40 litre airgun capacity. During processing of the data by a commercial seismic processing company, stacking velocities were determined at intervals between 2 and 5 km along each seismic line. Stacking velocities are usually derived from semblance analysis and are part of the seismic data package. Data from 12 velocity analysis locations close to the well (shown as crosses in Fig. 3) were converted to interval velocities using the Dix equation (Dix 1955). The interval velocities and TWT thickness were used to estimate interval thickness in metres, and these were summed to give depth below sea floor. The calculated depths were plotted as a function of TWT (Fig. 3) and used to derive a single TWT–depth relationship for the well site.

Figure 3  TWT–depth relationship for the Waka Nui-1 well (grey line connecting blue dots). Crosses show the stacking velocity data derived during processing of the seismic data on nearby seismic lines and the red line shows the interval velocities calculated from seismic or TWT–depth function.

The TWT–depth relationship was compared to the density log from the well (Fig. 4). Minor adjustments were made to the gradient of the TWT–depth relationship to tie the depth of significant density variations on the density log (RHOB) log with strong seismic events on the CNL95-034 seismic line. The adjustments improved the correlation of the brightest seismic reflector on line CNL95-034 at 3300 ms TWT with a significant Early Eocene decrease in density about 2930 m bmsl, and other significant density variations on the log correlate well with seismic horizons on the line. With the minor gradient adjustments the TWT–depth curve still lies within the range of stacking velocity values (Fig. 3), particularly with data from the most recent STRATOS-09 line.

Figure 4  Seismic data from line CNL95-034 at the Waka Nui-1 well with overlays showing the density log (RHOB converted to TWT-left of well), a synthetic seismogram (overlaid on the well) and depths of formations (to the right of well).

Interval velocities (Fig. 3) were combined with the density log to create a synthetic seismogram using the synthetic generation tool in the SeisWare seismic interpretation package. The synthetic seismogram using a 30 Hz Ricker wavelet correlates well with the seismic data from line CLN95-034, reproducing the most significant seismic events at the well location (Fig. 4) such as the significant Early Eocene event at 3300 ms TWT. Because the velocities used to create the synthetic seismogram are derived from the seismic data rather than a sonic log, the synthetic seismogram is only useful for identifying significant errors in the TWT–depth relationship.

Discussion

Interpreted interval velocities (Fig. 3, Table 1) increase with depth to the top of a high velocity layer (c. 3200 m/s) at a depth of 2555 m bmsl, corresponding to the Oligocene and Late Eocene Tikorangi Formation limestone in the well. Below 2740 m bmsl (base of the limestone), velocities range from 2200 to 2700 m/s and increase to 4200 m/s in Murihiku Supergroup rocks at the bottom of the well.

Table 1  Depth, two-way time and interval velocity data derived for the Waka Nui-1 well site.

	Depth (BRT)	Depth (m bmsl)	TWT from sea surface (ms)	Interval velocity (m/s)
Sea surface	24	0	0
Seabed	1479	1455	1940	1500
Top Ariki Formation	2171	2147	2685	1858
Top Miocene	2214	2190	2723	2263
Early Middle Miocene (unconf.)	2427	2403	2910	2278
Top Tikorangi Formation	2579	2555	3021	2739
Near top Tikorangi Formation	2610	2586	3040	3263
Base Tikorangi Formation	2764	2740	3135	3242
Near top Middle Eocene	2840	2816	3206	2141
Mid Eocene (Top Porangan)	2920	2896	3278	2222
Top Early Eocene	2951	2927	3301	2696
Top Palaeocene	3020	2996	3359	2379
Base Waipawa Formation	3118	3094	3437	2513
Top Kapuni Formation	3464	3440	3700	2631
Top Kapuni Conglomerate	3515	3491	3736	2833
Top Murihiku (unweathered)	3576	3552	3775	3128
TD	3681	3657	3825	4200

The interval velocities for geological formations were compared with velocities used by Herzer (1992) to map the offshore Northland Basin and with those derived for coeval formations in the Taranaki Basin (Thrasher 1988; Stagpoole 1997). Both these interpretations show a general increase in interval velocities from the seabed to the base of the Tikorangi Formation, lower velocities in the Eocene and Palaeocene formations and increasing velocities in Cretaceous and older strata. In the Taranaki Basin, the marked velocity reversal in Eocene and Palaeocene formations has been attributed to fluid over-pressure in the sandstone and mudstone formations below the Tikorangi Limestone (Humphris & Ravens 2000). However, no significant overpressure intervals were recorded in the mud-weight log from Waka Nui-1, with recorded mud weights less than 11 ppg (pounds per gallon) throughout the drilling operations (Milne & Quick 1999). Lithology also contributes to the velocity contrast, with limestone generally having a higher seismic velocity than either sandstone or mudstone (Sheriff & Geldart 1995).

The interval velocities derived for Waka Nui-1 well below the Tikorangi Formation limestone are on average lower than those in Taranaki, but they are within the range of velocities reported for these rocks in Taranaki wells (2350–4600 m/s; Stagpoole 1997). Explanations for the relatively low interval velocities include less compaction due to less sedimentary overburden in Waka Nui-1 (in general about one kilometre thinner than in Taranaki) and/or regional lithology variations.

The TWT–depth relationship in this study is an advance on previous work, but there are still ambiguities in the tie between seismic data and the stratigraphy in the well. This correlation will be further improved and the regional controls on seismic velocities better understood when more wells are drilled in the region and velocity data are collected during drilling operations.

Conclusions

Ties of seismic reflection data with the Waka Nui-1 well are uncertain because of the lack of velocity data from the well. This short communication documents a method that uses well data and petroleum industry seismic reflection data to derive a TWT–depth relationship for the well site (Table 1). The resulting relationship, based on stacking velocities and LWD logs from the well, provides a good correlation between significant density contrasts in the well and strong seismic horizons on the seismic data and is similar to relationships derived from previous studies in Northland and Taranaki.

The relationship can help constrain interpretations of seismic reflection data in the Northland and Reinga basins. The age and lithology correlations with the Waka Nui-1 well will aid geological interpretations of the offshore region northwest of New Zealand.

Acknowledgements

The author thanks Chris Uruski, Ray Wood and two anonymous reviewers for reviewing this manuscript and providing helpful advice. Information on the SeisWare seismic interpretation package is available from http://www.seisware.com.