New technique for accurate porosity estimation from logging-while-drilling nuclear magnetic resonance data, NGHP-02 expedition, offshore, India

Abstract

Abstract Drilling and acquiring high-quality downhole log data in the gas hydrate–bearing sedimentary sections, found mostly in deep marine environments and at low temperatures, is often challenging because of the disseminated gas hydrates and mostly unconsolidated nature of the host sediments. Logging while drilling (LWD) technology was used in National Gas Hydrate Program Expedition 02 (NGHP-02) in 25 wellbores to acquire data needed for the comprehensive analysis of gas hydrate occurrences offshore eastern India. The LWD tools deployed included a nuclear magnetic resonance (NMR) tool along with a density measurement tool for porosity computation, a high-resolution resistivity imaging tool for fracture evaluation and depositional system identification, and a borehole acoustic tool for gas hydrate identification and pore-pressure monitoring (by measuring deviation of sonic velocity from normal compaction trend). The NMR LWD tool measures transversely relaxing signal (echoes) decay, which is calibrated and inverted into transverse relaxation times or a T2 distribution (Figure 2). The T2 distribution is then summed to compute NMR porosity. Due to the solid nature of gas hydrates at in situ conditions, they are invisible to NMR. It is fundamental for gas hydrate identification to look for zones with differences between the NMR porosity and density porosity. This paper describes a new evaluation technique that addresses the porosity “deficit problem” due to the fast T2 relaxation observed in NMR data acquired during NGHP-02. In the new algorithm, the T1-T2 distributions are jointly inverted, compared to conventional NMR processing, which inverts a T2 distribution from the echo signal using a constant T1/T2 ratio. It is observed that with the conventional method which assumes a constant T1/T2 ratio, the NMR porosity in gas hydrate bearing zones is underestimated by about 3–6 porosity units, and the derived gas hydrate saturations are overestimated by ∼8–10%.