Arka Dyuti Sarkar , Mads Huuse

The North Viking Graben (NVG) is part of the mature North Sea Basin petroleum province and designated as a major carbon storage basin for NW Europe. It has been extensively drilled over five decades with an abundance of well and seismic data in the public domain. As such it serves as an excellent setting to demonstrate the efficacy of a reflection seismic data led approach to predicting subsurface temperatures using a state-of-the-art full waveform inversion velocity model covering the entire NVG. In a forward modelling approach, an empirical velocity to thermal conductivity transform is used in conjunction with predefined heatflow to predict subsurface temperature. The predefined heatflow parameters are set based on the range of values from previous studies in the area. Abundant well data with bottom hole temperature (BHT) records provide calibration of results. In the second step of inverse modelling, BHT’s as well as the velocity derived thermal conductivity are used to evaluate a 1D steady state approximation of Fourier’s Law for heatflow. In this way heatflow is estimated over the 12000 km2 model area at a km scale (lateral) resolution, highlighting lateral variability in comparison to the traditional point-based heatflow datasets. This heatflow is used to condition a final iterative loop of forward modelling to produce a temperature model that is best representative of the subsurface temperature. Calibration against 139 exploration wells indicate that the predicted temperatures are on average only 0.6 °C warmer than the recorded values, with a root mean squared error of 5 °C. BHT for the recently completed Northern Lights carbon capture and sequestration (CCS) well 31/5-7 (Eos) has been modelled to be 97 °C, which is 6 °C below the recorded BHT. This highlights the applicability of this workflow not only towards enhancing petroleum systems modelling work but also for use in the energy transition and for fundamental scientific purposes.