TITLE: How to Characterise the Embedded Uncertainty within processed data? A real-world example SPEAKER: Romain Chassagne (Institute of Petroleum Engineering, Heriot-Watt University) ABSTRACT: Field management decisions are based on representative enough model of the underground. As such, the model has to be “reliable”; to increase this reliability and decrease the uncertainty the model is updated; however, the update leads to the question of the data integration. This question is almost a routine for well production data (time-series) but is still an open question when it comes to the seismic integration. This specific assimilation of data, called seismic history matching (SHM) is an unsolved challenge up to now. Many tentatives and approaches have been tried over the decades, without reaching any clear statements or practical solution or good practice to follow for a given real dataset. What makes the SHM so difficult is mainly the seismic data itself, which is very complex by his very nature. Indeed, acquisition, interpretation, processing, make this data embedded with uncertainties, due to the physics and measurement issues. One of the challenges is to be able to extract and quantify the uncertainties carried over by the seismic data. In this talk I will present the seismic history matching workflow and challenges, some approaches to overcome these difficulties will be discussed. Romain Chassagne is a researcher at Institute of Petroleum Engineering, Heriot-Watt University in the UK. He holds MS and PhD degrees from Bordeaux University in France in applied mathematics. He worked on multiphase flow in porous media at Schlumberger Cambridge Research and Institut Français du Pétrole, before to join the Edinburgh Time-Lapse Project as a theme leader. Now Chassagne’s interests include update of the model with emphasis on seismic integration through history matching, along with proxy models, calibration, optimisation algorithms, seismic uncertainty.