Production Analysis of Offshore Gas Wells (UK North Sea) Using Bootstrapped Arps Method: Reserves Estimation Without Reservoir Model

Abstract

Abstract This study analyzes recently released proprietary daily production data from 31 decommissioned dry gas fields located on the UK Continental Shelf (UKCS). First, the hydrocarbon fields in the UKCS region were assigned to five groups, based on their geological source origin. The data analyzed here are from two lease areas comprising gas fields, sourcing from two of the five groups distinguished. One set of wells produced from Carboniferous Westphalian sandstones, the other from Permian Rotliegendes sandstones: both reservoir types remain active sub-salt targets. The production profiles of the 31+ legacy wells analyzed were grouped into four types: (1) Wells part of multi-well clusters producing from large fields, with production output controlled by facility access allocation; (2) Problem wells, with poor production and long downtime; (3) Wells with frequent shut-in days but excellent recovery; and (4) Fast clean producers; single well production profiles showing continuous decline without major interruptions, other than slight adjustments of choke settings. Two Permian gas fields (Mimas and Tethys) and one Carboniferous gas field (Kelvin) were produced with fast and clean producer wells. These wells appeared also equipped with bottomhole gauges which together with choke-settings and wellhead pressures, gives a comprehensive record of historic well performance. The historic well performance was analyzed using various bootstrapped data sets (6, 12- and 18-months production), which were least-mean-square-error (LMSE)-fitted with a traditional Arps decline curve analysis (DCA) method. The EUR forecast based on the bootstrapped Arps DCA (6, 12, 18 months) was compared with the known actual EUR based on the full historic well performance record. Based on our evaluation, a fast LSME fit of Arps DCA on early bootstrapped well data for 18 months appeared to give reliable estimates for the actual long-term EUR/well for two fields (Kelvin and Tethys). However, a third field (Mimas) required 5 years of bootstrapped data to achieve reliable EUR estimates.