Background
The Breede River, in the words of the author, "is the largest in South Africa's Western Province, and plays a significant part in the province's economy," but he says "models predict that flows into it could be seriously affected by climate change." More specifically, he reports that Steynor et al. (2009) used "a form of neural network" that was "trained on historical climate data," which were "linked to historical flow data at five stations in the Breede River valley," in order to ultimately "downscale from a global climate model to the typical area of a catchment," and to thereby determine the consequences of predicted future global warming for Breede River flows, the chief result of which analysis was that Breede River flows were projected to decrease if temperatures rise as predicted by climate models over the next 60 years.

What was done
As a check upon this approach to divining the region's hydrologic future, Philip Lloyd -- a researcher at the Energy Institute of the Cape Peninsula University of Technology located in Cape Town -- used flow data for five sites in the Breede Valley that had been maintained by the Department of Water Affairs to compute historical flow-rate trends over prior periods of warming ranging from 29 to 43 years in length.

What was learned
All of the future flow-rates calculated by Steynor et al. exhibited double-digit negative percentage changes that averaged -25% for one global climate model and -50% for another global climate model; and in like manner the mean past trend of four of Lloyd's five stations was also negative (-13%). But the other station had a positive trend (+14.6%). In addition, by "examination of river flows over the past 43 years in the Breede River basin," Lloyd was able to demonstrate that "changes in land use, creation of impoundments, and increasing abstraction have primarily been responsible for changes in the observed flows" of all of the negative-trend stations.

What it means
Interestingly, Steynor et al. had presumed that warming would lead to decreased flow rates, as their projections suggested; and they thus assumed their projections were correct. However, Lloyd was able to demonstrate that those results were driven primarily by unaccounted for land use changes in the five catchments, and that in his newer study the one site that had "a pristine watershed" was the one that had the "14% increase in flow over the study period," which was "contrary to the climate change predictions" and indicative of the fact that "climate change models cannot yet account for local climate change effects." As a result, he concluded that "predictions of possible adverse local impacts from global climate change should therefore be treated with the greatest caution," and that, "above all, they must not form the basis for any policy decisions until such time as they can reproduce known climatic effects satisfactorily."

Reference
Steynor, A.C., Hewitson, B.C. and Tadross, M.A. 2009. Projected future runoff of the Breede River under climate change. Water SA 35: 433-440.