Energy problems began a few years ago to receive proper treatment, but they are still most of the time dealt with from the supply side. The present research aims at adopting a more global view: this is a prospective study of the energy future of Québec which takes into account the fact that energy comes from and returns to a physical environment, is subject to its lows and in the meantime must contribute efficiently to the economy.

Quantitative work was accomplished using MARKAL, a large size computer model. This energy supply linear program has been adapted to the realities of Québec by the Groupe d'études et de recherche en analyse des décisions (GERAD) belonging to Hautes Études Commerciales of the University of Montréal. The specific contribution of this paper is the introduction and observation of the behaviour of technologies using local renewable resources.

The results of the various exploratory scenarios developed for this analysis indicate in particular that the penetration of renewable energy forms could be considerably more enhanced than is presently proposed, without any substantial changes to the prevailing official assumptions, simply by taking into account more diverse and prosaic forms of energy and technologies hiterto neglected. Such results point out that Québec could enjoy a greater measure of energy autonomy and that by a planned orderly transition to a decentralized system organized around its renewable resources, the Province could increase the flexibility and resiliency of its energy system, and therefore improve its economic structure. It appears that the current surplus hydroelectric capacity represents the greatest obstacle to the breakthrough of new technologies rather than their cost of investment.

The sensitivity of certain techno-economic parameters of electricity-producing equipments is examined in order to determine their robustness. The results of the model demand a careful interpretation by reason of the difficulties inherent in the setting-up of an homogeneous data bank and in the caracteristics of linear programming.