Table 2. Number of mallards dying from apparent white phosphorus poisoning

Table 2. Number of mallards dying from apparent white phosphorus poisoning while exposed to con-

taminated sediments within pens in ERF. (Data courtesy of Patty Pochop, DWRC.)

Season and

Hours of

Total

year of

exposure

Number of mallards

exposure

per day

(hr)

Pen 1

Pen 2

Pen 3

Pen 4

Pen 5

Pen 6

Total

No. mallards* exposed

* Six ducks per experiment.

(less then 1 g/g) detected in most samples are

concentrations (part per trillion) of white phos-

phorus are prevalent and persistent in field-

probably from colloidal, dissolved, or molecular

contaminated sediments at ERF.

white phosphorus sorbed to sediment surfaces.

White phosphorus is a solid at environmental

During cleanup at hazardous waste sites, cri-

temperatures. It has low water solubility and a

teria are required to determine if a specific area

moderately high octanolwater partition coeffi-

requires remediation and if the area is clean after-

cient. Therefore, we do not think that it would be

wards. Cleanup activities are designed to elimin-

mobile in the environment. Field and laboratory

ate risk. In the case of chemical contamination,

studies showed that solid pieces of white phos-

data from toxicological studies are considered

phorus are persistent in water and saturated sed-

with site characteristics to determine a cleanup

iment (Walsh et al. 1996a), but once dissolved in

level that is usually expressed in soil concentra-

water or in the vapor phase, white phosphorus is

tion terms. If the average contaminant concentra-

not persistent. However, we have found that low

tion, within a given level of confidence, in sam-

ples taken within an area falls below this

6801320

level, the area is judged to be clean. This

approach assumes that the spatially aver-

aged soil concentration and sample vari-

6801315

ance may be used to estimate a long-term

daily dose to receptors (USEPA 1992).

However, when a contaminant is acutely

6801310

toxic and heterogeneously distributed as

solid particles over an area, an average

6801305

concentration will not represent true

exposure risk.

The problem of estimating exposure

6801300

risk from white phosphorus became

apparent at ERF when Denver Wildlife

>1

0.1 to 0.99

Research Center (DWRC) conducted

6801295

0.01 to 0.099

studies to test the chemical bird repellent,

0.001 to 0.0099

Not Detected

methyl anthranilate (Clark and Cum-

mings 1994). As a component of this

6801290

355020 355025 355030 355035 355040 355045 355050

study, mallards were exposed to contami-

East (m)

nated sediment by confining them within

Figure 1. DWRC pens 5 and 6 showing white phosphorus concen- six pens. The pens were located in an area

tration ranges for discrete samples taken from 19911994 (upper where previous sampling had shown

rectangle is pen 6).

widespread white phosphorus contami-

2