Evaluation
75
Humboldt, and Stillwater wildlife management areas in Nevada. Irrigation drainage containing elevated concentrations of selenium was
determined to be toxic to young fish and other aquatic organisms, and
chronic selenosis was diagnosed in some waterbirds (Hallock and
Hallock 1993). The toxic effect of selenium at these sites shows that a
substantial biological hazard existed. The Protocol rating indicated high
hazard for 2 sites (Fernley and Humboldt) and moderate hazard for the
other (Stillwater). The HQ analysis indicated that no hazard was present
at any of the sites.
Example 3.
Belews Lake has a well documented history of selenium-induced teratogenesis and reproductive impairment in fish (Lemly 1985a, 1993a,
1997b). Selenium concentrations were somewhat elevated in bird eggs
but did not reach levels that substantially impair reproduction. However, the toxic effect of selenium on fish was devastating (19 species
were eliminated) and persistent, indicating high hazard for the site.
The Protocol gave a high hazard rating for Belews Lake, whereas HQ
analysis indicated that no hazard was present.
Evaluation
Hazard estimates from HQ analysis were incorrect for two reasons: (1)
shortcomings in the USEPA water quality criterion, and (2) a basic
flaw in the HQ approach as it applies to selenium. Although the USEPA
criterion (5 Ilg SelL) was developed in the mid-1980s using field data
that reflected current (pre-1988) information on bioaccumulation, more
recent information indicates that 5 Ilg SelL is too high to protect sensitive species of fish and aquatic birds from reproductive toxicity and the
effects of Winter Stress Syndrome (Skorupa and Ohlendorf 1991; Lemly
1993a, 1993b, 1996; Skorupa et al. 1996; NIWQP 1998; Hamilton and
Lemly 1999). The USEPA criterion is now well over 10 years old, and
several research studies and reviews published since then (eg, Skorupa
and Ohlendorf 1991; Peterson and Nebeker 1992; Lemly 1993a, 1997a;
Skorupa 1998) indicate that the value should probably be somewhere
in the 1-2 Ilg SelL range, especially for wetlands and impoundments
where bioaccumulation is maximized. Had a value of 2 Ilg SelL, for
example, been used instead of 5 Ilg SelL, the HQ numbers would be
quite different, and the hazard estimates would more closely match
those indicated by the Protocol for sites in Utah and New Mexico. Thus,
as it is currently practiced, HQ analysis will consistently underestimate selenium hazard because it relies on an invalid USEPA water
quality criterion.
75
Humboldt, and Stillwater wildlife management areas in Nevada. Irrigation drainage containing elevated concentrations of selenium was
determined to be toxic to young fish and other aquatic organisms, and
chronic selenosis was diagnosed in some waterbirds (Hallock and
Hallock 1993). The toxic effect of selenium at these sites shows that a
substantial biological hazard existed. The Protocol rating indicated high
hazard for 2 sites (Fernley and Humboldt) and moderate hazard for the
other (Stillwater). The HQ analysis indicated that no hazard was present
at any of the sites.
Example 3.
Belews Lake has a well documented history of selenium-induced teratogenesis and reproductive impairment in fish (Lemly 1985a, 1993a,
1997b). Selenium concentrations were somewhat elevated in bird eggs
but did not reach levels that substantially impair reproduction. However, the toxic effect of selenium on fish was devastating (19 species
were eliminated) and persistent, indicating high hazard for the site.
The Protocol gave a high hazard rating for Belews Lake, whereas HQ
analysis indicated that no hazard was present.
Evaluation
Hazard estimates from HQ analysis were incorrect for two reasons: (1)
shortcomings in the USEPA water quality criterion, and (2) a basic
flaw in the HQ approach as it applies to selenium. Although the USEPA
criterion (5 Ilg SelL) was developed in the mid-1980s using field data
that reflected current (pre-1988) information on bioaccumulation, more
recent information indicates that 5 Ilg SelL is too high to protect sensitive species of fish and aquatic birds from reproductive toxicity and the
effects of Winter Stress Syndrome (Skorupa and Ohlendorf 1991; Lemly
1993a, 1993b, 1996; Skorupa et al. 1996; NIWQP 1998; Hamilton and
Lemly 1999). The USEPA criterion is now well over 10 years old, and
several research studies and reviews published since then (eg, Skorupa
and Ohlendorf 1991; Peterson and Nebeker 1992; Lemly 1993a, 1997a;
Skorupa 1998) indicate that the value should probably be somewhere
in the 1-2 Ilg SelL range, especially for wetlands and impoundments
where bioaccumulation is maximized. Had a value of 2 Ilg SelL, for
example, been used instead of 5 Ilg SelL, the HQ numbers would be
quite different, and the hazard estimates would more closely match
those indicated by the Protocol for sites in Utah and New Mexico. Thus,
as it is currently practiced, HQ analysis will consistently underestimate selenium hazard because it relies on an invalid USEPA water
quality criterion.