Kesterson Crisis

The events leading to selenium contamination at the Kesterson Reservoir, a man‐made marsh in the Central Valley of California, began in the early 1900s, with the diversion of water to irrigate saline soils containing naturally high concentrations of selenium. With the advent of a systematic plan to collect and dispose of subsurface drainage water from these lands, problems associated with concentration and biological cycling of selenium became apparent at Kesterson, a series of shallow‐drainage‐water disposal ponds that was also managed as a wetland habitat. When chicks of hundreds of shorebirds failed to survive and deformed embryos were discovered, the California State Water Resources Control Board initiated a series of regulatory proceedings. This led to the closure of the Kesterson Reservoir and a series of remedial measures which went through numerous iterations as a team of scientists provided a more thorough understanding of processes leading to selenium transport and biologic exposure in this environment. In particular, field experiments and intensive monitoring demonstrated three factors leading to a cost‐effective resolution of the Kesterson cleanup. First, ground water under Kesterson was protected from selenium contamination by naturally occurring biogeochemical immobilization. Second, extensive monitoring of pore water in the vadose zone demonstrated that the favored soil‐excavation plan would be ineffective in reducing the elevated selenium concentration in ephemeral pools that occurs each winter at Kesterson. Finally, extensive biological monitoring demonstrated that selenium concentrations in the dominant species of upland vegetation at Kesterson was near or equal to “safe” levels. The combination of these factors led to a risk‐based remedial strategy that would leave the selenium inventory in place, and filling the low‐lying areas of Kesterson to prevent formation of ephemeral pools that attract wildlife. Observations that may be applicable to other areas experiencing similar environmental problems include the following: (1) Naturally occurring biogeochemical processes can be the cornerstone of effective remedial strategies; (2) although “safe levels” of toxic constituents based on total elemental analysis may be misleading and species‐specific targets may be more appropriate, seasonal cycles in speciation and analytical uncertainties make species‐specific goals difficult to define; and (3) scientific investigations and fast‐track remedial activities must go hand in hand to achieve cost‐effective, risk‐based remedial strategies.