H2S was found to be increasing in Gullfaks produced sea water. wells supported by the same injector showed similar development indicating an association between the injector and the H2S. Statoil's biofilm model was used to history match this behaviour to set the input parameters and enable a prediction to be made for each well. Some wells will exceed the process H2S limits and solutions are discussed. Produced water re-injection was shown to be detrimental to souring reservoirs.
The Gullfaks Field (Fig. 1) is situated in the Norwegian Sector of the North Sea in Block 34/10, production started in December 1986 / early 1987 from 3 Lower Brent subsea wells, supported by 1 subsea water injector, and 1 producer in the Cook Formation. By June 1993 it had produced 100 M sm (628 mmbbl of oil, 10330 M sm (36400 bscf) associated gas, 26 M sm (160 mmbbl) water and had injected 127 M sm (800 mmbbl) sea water.
Early water production started in the initial Rannoch wells during 1988 due to the Etive over ride as the Rannoch injectors had thermally fractured into the high permeability Etive. The sea water content of the low produced water cut rapidly increased from 0.5 to over 0.9 as determined from the routine ion concentration analysis.
Over 1 year after sea water breakthrough H2S was found to be increasing by routine surveys in oil, water and gas in some Gullfaks wells to levels significantly higher than native. wells supported by the same injector behave in the same manner indicating that the H2S was associated with the injection well.
Statoil has developed a mathematical injection well biofilm model which assumes a mass of bacteria growing around the injection well feeding on sea water supplied nutrients (ref. 1). The sulphate reducing bacteria (SRB) generated H2S is then transported by the injected sea water towards the supported production wells. However, most of the early generated H2S was adsorbed in the reservoir and therefore absent at sea water breakthrough in the production wells.
Gullfaks is a highly faulted field with relatively short injector producer spacing and direct support of the Lower Brent producers in each fault block by dedicated injectors resulting in numerous sea water pore volumes swept through the reservoir. This has lead to the adsorption capacity of the reservoir being saturated and the eventual production of H2S (Fig.3).
The Statoil biofilm model 'history matched' some of the mature Lower Brent H2S wells and predicted future H2S production profiles using simulated water production profiles and the estimated sea water fraction based on ion analysis. The model matched the rapid increase of H2S such that the biofilm's H2S generation capacity and the reservoir's absorption capacity parameters were fixed (Figs. 3,4,5).