During the past decade, the systems approach to storage reservoir problems has been heralded as something of a jump from the stone age of mass‐curve analysis into the modern era of science. In reality, however, no such jump ever occurred. There were a number of small ones but, contrary to the common belief, many of them were confined to the staircase of mass‐curve analysis and not all of them were in the upward direction. This paper attempts to put the mass‐curve technique into a proper perspective by clearing out some undesirable semantic underbrush accumulated over the past decades and by showing an intrinsic identity of some mass‐curve and systems‐analytic formulations. It demonstrates that, for the important special case of convex loss functions, both the dynamic and the linear programming formulations of optimum reservoir operation policies as developed over the past decade still have a long way to go to match a 55‐year old mass‐curve technique in terms of exactness, accuracy, as well as computational efficiency. Last but not least, it shows that the mass‐curve technique provides insights into the problems of storage reservoir operation which are entirely out of reach of the systems‐analytic methods and can significantly enhance the art of reservoir design and operation. Copyright 1979 by the American Geophysical Union.
