As is the case with any model, and with soil survey information in particular, ground-truthing at the field scale is necessary to verify results. We acknowledge limitations to our model. It does not consider proximity to a surface water source, which is an issue especially in areas that are irrigated solely from groundwater wells and are not connected to conveyance systems that supply surface water. The SAGBI also does not consider characteristics of the vadose zone or depth to groundwater. In arid regions, deep vadose zones may contain contaminants such as salts or agricultural pollutants that have accumulated over years of irrigation and incomplete leaching. These deep accumulations of contaminants could be flushed into the water table when excess water is applied during groundwater banking events. Furthermore, deep sediment likely contains hydraulically restrictive horizons that have not been documented, creating uncertainty as to where the water travels.Given these issues, SAGBI may be most useful when used in concert with water infrastructure models and hydrogeologic models — which generally do not incorporate soil survey information in a comprehensive way — to develop a fuller assessment of the processes and limitations involved in a potential groundwater banking effort.Selenium received recognition as an environmental contaminant in the 1980s,procona system as a result of the unprecedented events at the Kesterson Reservoir in California , a national wildlife refuge at the time . Large amounts of this trace element had been mobilized through irrigation of selenium-rich soils in the western San Joaquin Valley, transported along with agricultural runoff, and accumulated at the Reservoir.

Toxic selenium concentrations brought about death and deformities for as much as 64% of the wild aquatic birds hatched at the reservoir, including both local migratory species. Within a few years, the habitat of a variety of fish and waterfowl was classified as a toxic waste site . Today, the Reservoir’s ponds are drained and covered beneath a layer of soil fill , yet the mechanisms of selenium release now known as “the Kesterson effect” are still a threat in California and around the world . The environmental and management conditions creating irrigation-induced selenium contamination have been characterized in Theresa Presser’s seminal work . In brief, problems arise when seleniferous soils, such as those formed from Cretaceous marine sedimentary deposits along the Western side of the San Joaquin basin are subjected to irrigated agriculture. Salts, including selenium, naturally present in such soils are mobilized through irrigation, and high evaporation rates concentrate them in the root zone. In order to avoid negative effects on plant growth, subsurface drainage systems are used to export excess salts from the soil. This is particularly necessary in places where deep percolation is inhibited by a shallow impermeable layer. Such subsurface runoff routinely contains selenium in concentrations that exceed the US Environmental Protection Agency designation of toxic waste and thus poses an acute threat to aquatic ecosystems that receive it . The irrigation runoff feeding into the evaporation ponds of the Kesterson reservoir averaged 300 µg Se/L . The discovery of widespread deformities among waterfowl hatched near these ponds in 1983 led to a shift in the perception of selenium. While research had thus far been focused on farm-scale problems related to crop accumulation and toxicity to livestock, it became clear that excessive selenium concentrations in agricultural runoff was a watershed-scale resource protection issue that would greatly complicate irrigation management throughout the Western United States . As a result, California has been a hot spot for global research and management of environmental selenium contamination .

As selenium load management in the San Joaquin basin has made significant progress, new major sites of concern, such as the San Francisco Bay-Delta and the Salton Sea , have emerged in California. Current regulatory standards for selenium as aquatic contaminant are insufficient to be protective of sensitive ecosystems because they do not account for amplified exposure through bio-accumulation . There are many other pathways of anthropogenic selenium contamination – the San Francisco Bay-Delta for example receives half of its input from refineries . However, the diffuse agricultural sources are particularly hard to control , are the principal source of selenium in western US surface waters , and have shaped California’s history like no other selenium source. This paper analyzes what can be learned from the last three decades of seleniferous drainage management and regulatory approaches developed in California. In particular I seek to answer two key questions: 1) What were the greatest achievements and shortfalls of seleniferous drainage management in California? 2) To what extent may the current development of site-specific selenium water quality criteria for the San Francisco Bay and Delta serve as a model for future regulation?Selenium is a naturally occurring trace element heterogeneously distributed across terrestrial and marine environments . On land, seleniferous soils and those marked by selenium deficiency sometimes occur as close as 20 km from one another . Selenium contamination of natural ecosystems is linked to an array of human activities including irrigated agriculture, mining and smelting of metal ores, as well as refining and combusting of fossil fuels. The bio-spheric enrichment factor, which is computed as the ratio of anthropogenic to estimated “natural” emissions of a substance, was found to be 17 for selenium , highlighting the dominance of the anthropogenic component in the modern selenium cycle . Anthropogenic fluxes are expected to keep increasing in the foreseeable future as energy and resource demands increase . Selenium bio-accumulates, with tissue concentrations in animals and plants typically 1-3 orders of magnitude above those found in water.

Consequently, the predominant selenium uptake pathway for animals is through the consumption of food rather than water. Bio-accumulation and bio magnification are particularly intense in aquatic ecosystems and selenium contamination of such habitats is a global concern . In the Western United States alone, nearly 400,000 km2 of land are susceptible to irrigation induced contamination by the same mechanisms that led to the demise of the Kesterson Reservoir . Other nations where irrigation induced selenium contamination has been observed include Canada, Egypt, Israel, and Mexico . The environmental impacts of selenium depend on the element’s chemical speciation. The element’s primary dissolved forms, selenate and selenite , are mobile and bio-available . They can be sequestered in soils or sediments upon microbial reduction to solid elemental Se, metal selenides, or volatilized to the atmosphere upon reduction to gaseous methylated Se. Both selenate and selenite are toxic at elevated concentration,procona valencia buckets selenite however was found to be more toxic than selenate in direct exposure studies involving invertebrates and fish and also to bio-accumulate more readily at the base of aquatic food chains . Additionally, once any dissolved form of selenium is assimilated by an organism it is converted into highly bio-available organo-selenide species, Se . Exposure studies comparing organo-selenides to selenite in the diets of water birds established lower toxicity thresholds for the former . Organo-selenides are released from decaying organisms and organic matter during decomposition and can then persist in solution or be oxidized to selenite, while the conversion back to selenate does not occur at relevant rates in aquatic environments . Thus, recycling of selenium at the base of aquatic food webs through assimilation and decomposition usually leads to a buildup of the more bio-available and toxic forms over time . This buildup of bio-available selenium species may also explain why tissue concentrations in the upper trophic levels of stagnant or low-flowing ecosystems typically exceed those of fast flowing ecosystems with comparable selenium inputs, but shorter residence times . The complex environmental cycling of selenium has been a major obstacle in creating water quality regulations for this element . Regulatory concentration guidelines vary widely between jurisdictions and there are significant opportunities for new regulatory approaches . The Californian office of the EPA is currently working on site-specific water quality criteria for the protection of wildlife in the San Francisco Bay and Delta . These criteria are to be based on a modeling approach developed by USGS scientists, capable of translating tissue limits to dissolved concentration limits . There is hope among aquatic toxicologists that California’s new site-specific approach may become a model for national standards . For all contaminants regulated since 1985, aquatic life criteria under the Clean Water Act have been defined through separate dissolved concentration limits for longer term “continuous” and short term “maximum” limits . The selenium criteria that were established in 1987 defined continuous concentration limits of 5 µg/L as acid-soluble selenium with maximum concentrations not exceeding 20 µg/L more than once every three years for freshwater environments, but allowed up to 71 µg/L with up to one three-year exceedance of 300 µg/L for saltwater environments.These selenium limits became legally binding for 14 states including California after promulgation with the 1992 Water Quality Standards.

A central problem with the current criteria is that they were predominantly based on data drawn from direct exposure laboratory studies and thus failed to take into account the more ecologically relevant toxic effects due to bio-accumulation and trophic transfer. The freshwater criteria were based on field data from a contamination event , while the saltwater criteria were purely based on laboratory studies which did not account for bio-accumulation. The resulting difference of more than one order of magnitude between fresh- and saltwater criteria is not supported by field data . In fact, the saltwater criteria have widely been regarded as under protective of wildlife, including waterfowl . In addition, the freshwater criteria appear under productive of particularly sensitive ecosystems and species . To be protective of waterfowl in the wetlands of the Central Valley Region, a 2 µg Se/L monthly mean water quality criterion was deemed necessary by the Regional Water Quality Control Board and this objective was officially approved for the region by the EPA in 1990 . For the wetlands of the Central Valley Region, this criterion overrides the statewide criteria promulgated in 1992 and remains in effect today. However, given the wide range of bio-availability between different selenium species and the complex transfer processes between environmental compartments and trophic levels, regulation based solely on dissolved or acid-soluble concentrations has been characterized as inadequate . In response to such criticism the EPA proposed in 2004 a new tissue-based criterion for selenium with a 7.91 µg/g fish tissue limit to supersede the previous national water quality guidelines for selenium. This limit is based on the lowest level of effect in juvenile bluegill sunfish under simulated overwintering conditions . Whereas there is little doubt that tissue concentrations are more representative of exposure than dissolved concentrations for individual species, it is unclear if a single fish tissue limit will be protective across entire food webs including a diversity of fish and waterfowl . The proposed tissue based criteria have to date remained at draft stage due to objection by the US Fish and Wildlife Service. The historic developments that lead to the rise of selenium contamination in the San Joaquin Valley can be traced to the passage of the California Water Resources Development Act of 1960. The Act laid the financial foundation for the State Water Plan providing for the construction of the nation’s largest water distribution system and including also infrastructure measures for “the removal of drainage water” . The State Water Projects funded under this plan began delivering water to 4,000 km2 in the Southern San Joaquin Valley as of 1968 . To prevent salinization and manage agricultural runoff, the Bureau of Reclamation constructed collector drains, a main drainage canal , and a regulating reservoir, Kesterson . Originally, the San Luis Drain was planned to deliver drainage out of the San Joaquin Valley all the way to the San Francisco Bay Delta, however the northern part of the drain was never completed . Instead, from the time of the San Luis Drain’s completion in 1975 until its temporary closure in 1986, all runoff water channeled through the drain was delivered to the evaporation ponds of the Kesterson Reservoir, which had become part of a newly created national wildlife refuge in 1970 . There, in the early 1980s, high rates of embryo deformity and mortality, as well as large numbers of adult deaths among waterfowl were identified as caused by the elevated selenium concentrations in the evaporation ponds . This led to the closure of the Reservoir to all runoff inputs in 1986 .