Return flow from a 82,030‐ha (202,700‐acre) tract of calcareous silt loam soils irrigated with water diverted from the Snake River in southern Idaho increased the downstream total soluble salt and NO3‐N loads, but decreased the downstream PO4‐P load. Under the existing water management practice, 50% of the total input water returned to the Snake River as subsurface drainage. Net total soluble salt output was 2.4 metric tons/ha (1.0 English ton/acre) and, on the average, was considerably greater than necessary to maintain a salt balance. Net NO3‐N output was 33 kg/ha (30 lb/acre). Only about 30% as much PO4‐P left the tract via drainage water as entered the tract in irrigation water. As water passed through the soil, PO4‐P was removed by chemical reactions in the soil, thus decreasing the concentration in the subsurface drainage water and decreasing the downstream PO4‐P load. Applied P fertilizer was not leached into the drainage water.

Phosphorus was measured in irrigation and surface drainage waters for two large irrigation tracts, and inorganic, water‐soluble PO4‐P and total P inputs and outputs were computed. The present irrigation practices on both irrigation tracts conserve P by removing more P from the Snake River in irrigation water than is returned in drainage water. Even greater P conservation could be attained by implementing new practices. Approximately 90% of the P in waters diverted to irrigate the Northside Tract remained in the tract. About 50% of the amount diverted remained in the Twin Falls Tract. Particle size segregation takes place in some drainage streams and the finer sediments returning to the river contain higher total and NaHCO3‐extractable P concentrations than the soils from which they were eroded. The importance of defining sampling methods and procedures for P analyses is illustrated by comparative data on irrigation and drainage waters. The PO4‐P concentration in samples filtered through a 0.45 µm membrane filter was independent of the sediment concentration in waters, whereas the total P concentration of unfiltered samples was proportional to the sediment concentration.

Sediment inputs from Snake River irrigation water and sediment losses back to the river were measured for two large irrigated tracts in southern Idaho. There was a net sediment accumulation of 0.69 metric tons/ha onto the 65,000 ha Northside tract but a net sediment loss of 0.46 metric tons/ha from the 82,000 Twin Falls tract. Differences in sediment losses from the two tracts result from the difference in sedimentation in the drain‐ways of the two projects. Sediment deposited in drains on the Northside tract amounted to 4.5 metric tons/ha compared to 0.95 metric tons/ha for the Twin Falls tract. Drains on the Northside tract were constructed to grade whereas most drains on the Twin Falls tract are natural channels with steeper gradients. The net amounts of sediment eroded from farms within each tract were 4.0 metric tons/ha for the Northside tract and 1.42 metric tons/ha for the Twin Falls tract. This erosion loss from farms could be reduced within each tract by more careful use of water and construction of on‐farm sediment retention ponds. This would also reduce the amount of sediment returned to the river and lower costs of mechanically removing sediment from drains and canals. Construction of sediment retention ponds along main drains and reducing the amount of surface runoff returning to the river would also reduce the amount of sediment returning.

EDIMENT and phosphorus (P) removal efficencies of a sediment-retention pond with a capacity of about 3400 m" receiving surface water runoff from 4050 ha of irrigated land, were measured for five years. Average daily flow through the pond, during the irrigation runoff period, was 347 Lis, with a pond retention time of 2.7 h. The pond removed 65 to 76 percent of the sediment, and 25 to 33 percent of the total P entering the pond. Sediment and phosphorus removal efficiencies depended upon the flow rate and the sediment concentration of surface return flow entering the pond. Sediment and phosphorus were most efficiently removed when the stream flow was 340 to 453 L/ s and the sediment concentration was in the range of 20 to 750 mg/L. Sediment removed from the pond was used to cover protruding basalt to improve and expand a golf course. Article has been reviewed and approved for publication by the Soil and Water Division of ASAE. Contribution from USDA•SEA•AR, in cooperation with the Water

Irrigation and subsurface drainage waters sampled from an 62,150-hectare (203,000-acrel irrigation district in southern Idaho were evaluated for bacteriological quality. The soils in the district Are wind deposited over fractured basalt, calcareous, and have a pH near 7.8. Drainage, where needed, is provided by horizontally mined tunnels or by tile drains connecting shallow relief wells that flow the year around. For the 12 months ending September 30, 1969, a 2-meter (6.5-foot) depth of water for the entire irrigation tract was diverted, and 50% of the water passed through the soil becoming subsurface drainage. The irrigation water and seven subsurface drains were sampled at 2-week intervals during the summer of 1969. Conform, fecal streptococci, starch hydrolyzers, and bacteria able to grow at temperatures from 0 to 55C were counted. The diverted irrigation water contained from 140 to 3,300 coliform per 100 ml, but 86% of the subsurface drainage samples contained 5 or fewer conforms per 100 ml. Numbers of other microorganisms were also low in the drainage waters. The outflow samples were oxygen saturated and the temperatures were 13.0 ± 1.1C for all samples. Percolation through the soil improved the water quality almost to domestic water standards.Additional Index Words: pollution control, water purification, ground water.Water diverted from the Snake River in southern Idaho for irrigating agricultural land is contaminated with coliforms and other microorganisms normally associated with human pollution (7). This contamination originates from sewage outfalls and food processing wastes upriver. The coliform count occasionally exceeds standards for swimming and other body contact sports, but it has not been determined that a health hazard for agricultural uses exists. In a survey of water use on the Twin Falls irrigation district during 1969, Carter, Bondurant, and Robbins (3) determined from measurement of water diverted, rainfall, surface drainage, and calculated evapotranspiration, that about 50% of the input water, or 100 cm passed through the soil and emerged from subsurface drains.Portneuf silt loam is the predominant soil in the irrigation district. It is well drained and the nearly level phase is typical of the series, although the slope ranges from 0 to 12% with an estimated 50% of the land sloping less than 3% and 30% sloping from 3 to 6%. The depth of bedrock is over 1 m and the parent material is Aeolian (silt). Soil permeability is moderate, ranging from 2 to 6 cm/hour. Typical ranges in mechanical analyses are 15 to 20% clay, 55 to 65% silt, and 15 to 30% very fine sand (0.1 to 0.05 mm). Cation exchange capacity is approximately 21 meq/100 g soil and is made up of calcium 12, magnesium 5, sodium < 1, potassium 0.5, and hydrogen 3 meq/100 g soil. Characteristic of much of the soil is a lime-cemented hardpan approximately 35 to 50 cm below the surface. Bedrock is undecomposed basalt that is fractured, and interlayered with cinders and ash that acts as a reservoir and conveyance for a large volu...