Fish-tissue and bed-sediment samples were collected from 20 sites in the upper Snake River Basin in Idaho and western Wyoming as part of the National Water-Quality Assessment Program to determine the occurrence and distribution of organochlorine compounds. During 1992-94, 41 samples were analyzed for 28 different organochlorine compounds in whole-fish tissue and 32 compounds in bed sediment. Sites sampled were third- through seventh-order streams that represented three environmental settings: reference conditions, agricultural land use, and mixed (agricultural and urban) land use. Fourteen organochlorine compounds were detected in fish tissue and nine in bed sediment. All compounds detected in bed sediment also were detected in fish tissue. Fish-tissue and bed- sediment samples from agricultural and mixed land-use sites contained one or more organochlorine compounds. The most frequently detected compound at all sites was p,p'DDE, which was present in 80 percent of the fish-tissue and 30 percent of the bed-sediment samples. A maximum of three compounds were detected, all in fish-tissue samples from reference (forest and rangeland) sites. The highest number of compounds was detected in fish-tissue (nine) and bed-sediment (eight) samples from mixed land-use sites. No clear relation was apparent between the occurrence of external anomalies and fish-tissue contaminant concentrations or land use.

The distribution of organochlorine compounds in the basin was related to land use. Total DDT was detected at sites in all land uses; total PCB was detected at only agricultural and mixed land-use sites. Total chlordane was detected in fish- tissue samples from primarily mixed land-use sites; samples from six of the eight sites contained detectable concentrations. Median concentrations of p,p'DDE in fish-tissue samples from mixed land- use sites were significantly higher (p<0.05) than from reference and agricultural sites. Significant positive relations between percent agricultural land and concentrations of total DDT (r² = 0.41) and lipid-normalized total DDT (r² =0.48) were observed.

Concentrations of p,p'DDE, total PCB, total DDT, and toxaphene in fish-tissue samples from three mixed land-use sites equaled or exceeded national guidelines established for protection of fish-eating wildlife: Portneuf River at Pocatello, total PCB; Rock Creek at Twin Falls, p,p'DDE, total DDT, and toxaphene; and Snake River near Buhl, p,p'DDE and total DDT. Concentrations of total DDT and p,p'DDE in 32 and 34 percent of fish-tissue samples, respectively, analyzed during this study exceeded the 1980-81 U.S. Fish and Wildlife Service/National Contaminant Biomonitoring Program (USFWS/NCBP) geometric mean concentrations. Concentrations of total PCB in samples from the Portneuf River at Pocatello and concentrations of toxaphene in samples from Rock Creek at Twin Falls also exceeded the USFWS/NCBP geometric mean concentrations. Comparisons of 1970-84 USFWS/NCBP concentrations of total DDT and total PCB in fish-tissue samples from the Snake River near Hagerman with concentrations measured during this study indicated a decreasing trend. Concentrations of p,p'DDE in all sediment samples from Rock Creek at Twin Falls exceeded the Canadian Probable Effect Level guideline. Total PCB was detected in sediment from only one site, Portneuf River at Pocatello, which was also the only site where concentrations of total PCB in fish tissue were elevated. Because organochlorine compounds are lipophilic and tend to bioaccumulate in tissue, fish are a better indicator of organochlorine contaminant occurrence and distribution than are bed sediment or water in the upper Snake River Basin.

Some of the highest concentrations of organochlorine contaminants in tissue and sediment in the basin were detected at sites receiving irrigation- return flows. Results of this study support the importance of controlling sediment erosion on irrigated land to reduce the quantity of contaminants entering streams that receive irrigation-return flows.