In the mid 1990’s, the North Atlantic subpolar gyre has shown a dramatic warming event that has been thoroughly investigated from observations and numerical simulations. Some studies suggest that it is due to an interannual, wind-driven weakening and shrinking of the gyre that facilitated the penetration of warm Atlantic Water, the weakening of the gyre being attributed to changes in the North Atlantic Oscillation (NAO) and the East Atlantic Pattern, which are the two dominant modes of atmospheric variability in the North Atlantic. However, other studies suggest that the warming event is due to a decadal, buoyancy-driven strengthening of the meridional overturning circulation and subsequent intensification of the poleward heat transport, in response to the positive NAO conditions of 1988-1995. To reconcile this discrepancy, the heat budget in the North Atlantic subpolar gyre is reconstructed from four ocean hindcast simulations sharing the same modelling platform but using different settings. The novelty of this work is the decomposition of the subpolar gyre into a western and an eastern subregion, which is motivated by water mass distribution around Reykjanes Ridge and by the fact that deep convection only occurs in the western subpolar gyre. In the western subpolar gyre, the 1995 warming event is the decadal, baroclinic ocean response to positive NAO conditions from 1988 to 1995. The latter induced increased surface heat loss in the Labrador Sea that intensified deep convection hence strengthened the meridional overturning circulation and the associated poleward heat transport. In the eastern subregion, a concomittant warming was induced by an interannual, barotropic adjustment of the gyre circulation to an abrupt change from positive to negative NAO conditions in the winter 1995-1996. Indeed, the gyre response to negative NAO conditions is a cyclonic intergyre-gyre which increases northward volume and heat transports at the southeastern limit of subpolar gyre. Therefore, the discrepancies found in the literature about the 1995 warming event of North Atlantic subpolar gyre are reconciled in the present work, which suggests that the atmospheric drivers, the mechanisms at stake and the associated timescales are different to the east and to the west of Reykjanes Ridge.