The Yakima Basin National Water-Quality Assessment Program collected fecal-coliform bacteria samples during three synoptic samplings to identify and quantify the cause, source, transport, and effects of fecal-indicator bacteria in Yakima River Basin streams. The August 1999 synoptic sampling targeted the Yakima River main-stem and tributary sites, while the July and October-November 2000 synoptic samplings targeted small- and intermediate-sized agricultural watersheds during irrigation and nonirrigation season, respectively. Quality-assurance results indicated that variability in fecal-coliform concentrations is large and, therefore, a difference of an order of magnitude or more between sites or between times is required for the values to be significantly different 90 percent of the time.

The August 1999 synoptic sampling results indicated that (1) 44 percent of the sites visited, including all the main-stem Yakima River sites, met the Class A fecal-coliform 90th percentile standard of 200 colonies per deciliter, (2) tributaries were the likely source of fecal contamination to the main stem, and (3) tributaries with high fecal-coliform concentrations typically also had high suspended-sediment concentrations. Results of the July and October-November 2000 synoptic samplings indicated that (1) 36 and 81 percent of the sites sampled, respectively, met the standard, (2) during the nonirrigation synoptic sampling, four of the six sites not meeting the standard were from the Granger and Sulphur subbasins, and (3) fecal-coliform concentrations during the irrigation season were generally higher than during the nonirrigation season.

Several levels of temporal variability were examined. The short-term variability observed during a synoptic sampling was found to be site specific, with some sites fairly consistent, while others were rather variable. Seasonally, most sites from the 2000 synoptic samplings showed higher concentrations during irrigation than during nonirrigation. Historically, 13 of the 22 sites sampled during both the July 1988 and August 1999 synoptic samplings had higher concentrations in 1999. The three sites with the highest concentrations in July 1988, however, all had decreases in August 1999. When compared against historical (1972-85) minimum and maximum summer-month medians, the August 1999 synoptic-sampling concentrations generally were between these values.

Instantaneous fecal-coliform bacteria loads were calculated for the August 1999 synoptic sampling in an effort to study the dynamics of bacterial transport. Tributaries affected by agricultural, urban, and hobby farm activities were generally the major sources of bacteria to the main-stem Yakima River during this time. When these August 1999 synoptic-sampling loads in the lower basin reach from the Yakima River at river mile 72 to Kiona (river mile 29.9) were compared to those from the July 1988 synoptic sampling, most sites had higher loads in 1999.

A nonparametric Spearman test was used to detect correlations between fecal-coliform concentrations and physical and chemical data collected during the synoptic samplings. Results for the August 1999 synoptic sampling, which included many mouths of tributaries, showed strong significant correlations with almost every variable. In contrast, only some of the nutrient concentrations showed strong significant correlations during the July and October-November 2000 synoptic samplings, which included small and intermediate- sized agricultural streams.

Looking forward relative to future monitoring goals, research needs, and best management practice development, four hypotheses that deal with processes and sources of bacteria were identified: (1) overland runoff transports bacteria from land surfaces to streams, (2) bacteria in the water column tend to associate with suspended matter, (3) with increasing densities of warm-blooded animals, the likelihood of fecal-coliform contamination in streams also increases, and (4) identifi- cation of bacterial sources is difficult, but must be attempted for remediation to be possible.