Qualitative and quantitative interpretation of airborne gravity and HRAM targets structures for prospect-scale 2D seismic: Southern Maranon Basin, Peru

In conjunction with our partners Ecopetrol, Talisman flew a 12,220 km2 aerogravity/aeromag survey (Sander) in Blocks 134/158, Southern Maranon Basin, Peru in mid-2008. The primary objective of the survey was to reduce exploration cycle time and costs<br>by eliminating a round of regional reconnaissance 2D seismic, moving directly to prospect scale 2D seismic over the most promising anomalies. The survey coverage included areas to the north and west of the area of interest with existing structures<br>defined on 2D seismic, some with well control. 2D and simple 3D modelling studies were used to evaluate airborne gravity vs airborne gravity gradiometry. A careful analysis of expected signal to noise performance over the wavelengths of interest suggested that both methods would perform well. Airborne gravity was selected as the most cost-effective, and for reduced contamination of signal from near-surface geologic noise. Qualitative interpretation was based primarily on bandpass filtered Bouguer anomaly maps. In areas with 2D seismic, there was a strong correlation between structural closures mapped from seismic and anomalies identified on specific filter products. The spatial scale-length of structures appears to be consistent and extends into the area of interest where there is no previous subsurface mapping. The revised structural interpretation agrees with preferred tectonic models. Existing structures on key seismic lines were modelled to determine the 2D anomalous Bouguer gravity response, and compared with profiles extracted from filtered products. Anomaly amplitudes and corresponding contour intervals were chosen as proxies for prospect areas. The prospect areas associated with mapped anomaly contours were then used to design a prospect-scale 2D seismic program. Constrained 3D gravity inversion has since confirmed the prospect locations and scales. The dataset was decimated and reprocessed to estimate signal to noise ratios, which confirming the signal to noise estimates.