Stormflow-hydrograph separation based on isotopes: the thrill is gone--what's next?

Beginning in the 1970s, the promise of a new method for separatingstormflow hydrographs using18O,2H, and3Hprovedanirresistibletemptation, and was a vast improvement over graphical separationand solute tracer methods that were prevalent at the time. Eventu-ally, hydrologists realized that this new method entailed a plethoraof assumptions about temporal and spatial homogeneity of isotopiccomposition (many of which were commonly violated). Nevertheless,hydrologists forged ahead with dozens of isotope-based hydrograph-separation studies that were published in the 1970s and 1980s.Hortonian overland flow was presumed dead. By the late 1980s,the new isotope-based hydrograph separation technique had movedinto adolescence, accompanied by typical adolescent problems suchas confusion and a search for identity. As experienced hydrologistscontinued to use the isotope technique to study stormflow hydrol-ogy in forested catchments in humid climates, their younger peersfollowed obligingly—again and again. Was Hortonian overland flowreally dead and forgotten, though? What about catchments in whichpeople live and work? And what about catchments in dry climatesand the tropics? How useful were study results when several of theassumptions about the homogeneity of source waters were commonlyviolated? What if two components could not explain the variation ofisotopic composition measured in the stream during stormflow? Andwhat about uncertainty? As with many new tools, once the initialshine wore off, the limitations of the method became a concern—oneof which was that isotope-based hydrograph separations alone couldnot reveal much about the flow paths by which water arrives at astream channel during storms.