by ultimate upstream depletions and is increased by divertible flood flows from unregulated tributaries.

DIVERTIBLE FLOW OF M'KAY CREEK

Flow of McKay Creek is divertible to bottom lands adjacent and also downstream from its mouth and to uplands west of it. These lands total roughly 10,000 acres and have an annual demand of about 26,300 acre-feet in addition to redivertible return flow. Use of McKay Creek flow on lands under Ochoco Reservoir is limited to demand by uplands above the relift pump and bottom lands above the distribution canal, plus peaking requirements of the relift pump. Under this limitation, maximum use of McKay Creek flow in operation of Ochoco Reservoir is 2,300 acre-feet per season. By subtraction, maximum use in operation of Prineville Reservoir is 24,000 acre-feet per season. If the foregoing limitations of use are computed by months, recorded runoff of McKay Creek exceeds the limitation in only 3 months of record since 1928.

UPSTREAM DEPLETIONS

Upstream depletions by 12,190 acres of land for which irrigation storage might be provided were computed from operation studies of 2 reservoir sites on the headwaters of Crooked River. The storage capacities considered were near the maximum justifiable, and would yield a sustained water supply only with long periods of holdover. It is, therefore, unlikely that the depletions computed for each storage season would ever be exceeded under comparable streamflow conditions.

Return flow from these 12,190 acres would offset the gross depletions to some extent. However, this return flow would provide a full water supply, with a small surplus, for an additional 4,745 productive acres lying along Crooked River between the mouth of Paulina Creek and the head of Prineville Reservoir. During the irrigation season, therefore, only a small amount of return flow, including any from the latter acreage, would enter Prineville Reservoir to be credited against the gross depletions. All return flow from these lands during the storage season was assumed to reach Prineville Reservoir. Some 4,180 acres of the lands included in the above acreages have been inadequately irrigated by direct diversions. Historical depletions by these lands were estimated and were deducted from the future depletions to avoid compounding the effect of their diversions.

Depletions by future upstream development above Prineville Reservoir, therefore, include five components: storage retention, winter return flows, direct diversions during the irrigation season, return flows in excess of all diversion requirements during the irrigation season, and depletions by presently irrigated land. Annual upstream depletions computed from these 4 factors range from about 50,000 acre-feet in a year of high runoff following discharge of all reservoir storage, to a slight negative amount in an extremely dry year when winter return flows exceed storable runoff. These depletions are largely from runoff that would exceed allowable reservoir content. In general water resources are adequate to supply the additional lands under the two upstream reservoirs.

RESERVOIR LOSSES

Reservoir losses consist of seepage and evaporation, and are computed from past data recorded for the project.

Seepage

Prior to 1949 all but a negligible part of the seepage from Ochoco Reservoir was collected into well-defined channels and measured by three weirs. The years of combined record for these weirs are shown in table 7 as Ochoco Springs below Ochoco Reservoir. In the first year that the reservoir filled, monthly seepage loss from the right abutment of the dam was about 6 percent of the average content. Patching operations adjacent to the largest boulders in the landslide mass forming the right abutment subsequently reduced this seepage somewhat. The heterogeneous nature of the abutment made these operations difficult, and no further patching has been successful. Seepage with any given stage remained generally uniform for 24 years.

Recorded monthly seepage was plotted for correlation with mean reservoir content. Although there is considerable deviation of points, a mean curve is well defined when excessive seepage in the first 3 years of operation is disregarded. The curve shows mean monthly seepage ranging from about 1 percent of mean content at low reservoir stages to slightly over 4 percent when the reservoir is at present spillway level. Only winter seepage is wasted, as summer seepage enters the existing distribution system. For this reason, and because there is no record of seepage since the dam was rehabilitated, further reduction of this seepage was not assumed. The derived seepage curve was, therefore, adopted for study of the rehabilitated Ochoco Reservoir.

Seepage losses from Prineville Reservoir should be negligible. It has been possible to locate the dam so as to avoid landslide materials on both abutments. Seepage which does exist will be usable for the most part; therefore no further consideration has been given to seepage losses from Prineville Reservoir.

Evaporation

Reservoir evaporation is computed from average monthly temperatures at Prineville for the period 1931-46 by means of a general relationship adapted to the project locality. Seasonal pan evaporation so computed is 41 inches after adjusting slightly for prevailing monthly variation. With a conversion factor 0.7, equivalent gross reservoir evaporation is 29 inches.

As the record of inflow to Ochoco Reservoir is based on runoff of influent streams above the reservoir, or on outflow corrected for changes in storage, reservoir evaporation for use in operation studies is gross evaporation, less precipitation. On this basis, using published normal precipitation at Prineville, net reservoir evaporation is as follows, in feet:

Recorded inflow to Prineville Reservoir site is based on the record of discharge at the dam site. Net reservoir evaporation is, accordingly, the increase in consumptive uses within the reservoir site when present

conditions change to reservoir conditions. Present consumptive use is supplied from precipitation, irrigation, and withdrawal of moisture from soil storage. Present storage of soil moisture takes place in winter, and its elimination will probably be only sufficient to offset the slight reservoir seepage expected. Below spillway level, a portion of the reservoir is now irrigated, and the remaining area is dry farmed or in native grass and sage. The nonirrigated areas use nearly all growing season precipitation, in addition to soil moisture. The irrigated area has only a flood season supply, with a resulting high early consumptive use and a use thereafter only about equal to precipitation. These uses were estimated from general crop consumptive use and precipitation at Prineville, and were weighted for the area irrigated to obtain average consumptive use which, deducted from gross reservoir evaporation, results in net evaporation from Prineville Reservoir site as follows, in feet:

Evaporation in the nonirrigation season was disregarded for reservoir operation, as it would usually amount to less than 50 acre-feet per month. Records for Cold Springs Reservoir near Hermiston, Oreg., indicate that only about 20 percent of the annual evaporation would occur in this season.

MULTIPURPOSE STORAGE

The project reservoirs would be used for flood control, recreation, and preservation of fish and wildlife, in addition to their primary use for irrigation. Except for irrigation and flood control, no exclusive storage capacities are allocated for these other uses, although a release of 10 second-feet from Prineville Reservoir is planned for benefit of downstream fish life in months of no spillway or irrigation discharge. Dead storage capacities of 1,400 acre-feet at Ochoco Reservoir and 2,000 acre-feet at Prineville Reservoir site provide a conservation pool, which in part is based on requirements for recreation and fish and wildlife.

Ochoco Reservoir would be operated with an exclusive reservation of 16,500 acre-feet for flood control. Preliminary operation studies indicated that 12,000 acre-feet of the remaining 30,000 acre-feet of active capacity could be used jointly for irrigation and flood control. Early in the storage season, the reservoir, if not depleted by irrigation releases, would be evacuated to provide this amount of additional space until January 31. During February, 11,000 acre-feet of this space would be reevacuated if used to retain flood inflow, or the reservoir would be filled to this extent near the end of the month if no floods develop. During March, 7,500 acre-feet of this space would be operated in a similar manner. During April, 2,000 acre-feet of this space would also be used similarly, and if no floods are in prospect after May 1, the reservoir would be filled to its total irrigation use capacity of 30,000 acre-feet.

Prineville Reservoir has no exclusive flood-control reservation. Preliminary operation studies indicated a joint-use capacity of 60,000 acre-feet for irrigation and flood control. Between November 15 and

February 28, no storage within this space would be permitted, except for temporary retention of floods. During March, 36,000 acre-feet of this space would be reevacuated if used to retain flood inflow, or the reservoir would be filled to this extent near the end of the month if no floods develop. The reservoir would be filled during April.

This plan of operation would depend upon short-term forecasts of runoff during the snowmelt period. Such forecasts would probably enable reduction of the occasional wastes indicated by present operation studies. This condition results in a year when the reservoirs do not fill to irrigation and joint-use capacity. Ordinarily, any waste resulting from the schedule merely replaces waste that would otherwise occur later when irrigation and joint-use capacity fills.

SEDIMENT PROBLEMS

Problems of sedimentation of reservoirs in the area do not appear to be serious. Stream sediment samples have been taken at the gaging station near Prineville but do not cover a prolonged period of time. During 1911 and 1912, samples were taken at the gaging station on Crooked River at Stearns' Ranch, 51⁄2 miles southeast of Prineville. During the 1951-52 water year, samples were taken at the Prineville gaging station on Crooked River. These samples indicate that suspended matter is generally less than 50 parts per million during the period May through February. During March and April, the suspended matter increases with a maximum reported load of 1,570 parts per million on March 27, 1952, with a discharge of 7,300 acre-feet. The previously reported maximum discharge was 9,080 second-feet on March 1 and 2, 1910, at Stearns' Ranch, about 51⁄2 miles downstream from the present gage.

Short period of high temperature, followed by relatively light rainfall, cause the rapid runoff of snowmelt, which causes flooding with accompanying relatively high silt loads. These floods occur during periods when flood-control storage is reserved in the reservoirs and seldom last more than 2 or 3 days. These flood waters are released within a period of a few days, with the resultant lower trap efficiency of the reservoir to catch suspended sediments.

Estimated suspended matter of Crooked River is about 100,000 tons per year. Based on a trap efficiency of 80 percent and an average sediment weight of approximately 70 pounds per cubic foot, the average annual storage capacity filled by sedimentation would be some 52 acre-feet per year.

Although there are numerous records of the quantity of suspended sediment load carried by various streams throughout the United States, the quantitative data relative to the bed load for these streams are meager and usually negligible. Estimates of bed load made for Elephant Butte Reservoir, Rio Grande project, New Mexico-Texas, ranged from 15 to 25 percent of the total sediment. Assuming the bed load of Crooked River to average 20 percent of the total silt load, it would be approximately 13 acre-feet per year.

The estimated sedimentation of Prineville Reservoir, based on the above sketchy information, might be as much as 100 acre-feet per year. Considering the proposed reservoir capacity, which is made up of 2,000 acre-feet of dead storage and 153,000 acre-feet of irrigation and flood-control space, it is evident that the prospective sedimentation

would only decrease the active capacity by a minor amount and thus would not be a serious problem. No storage space is dedicated to sediment control. A sediment sampling program is being continued with a view to providing a firmer basis for estimated sediment encroachment for use in reservoir space allocations when plans are completed for use of the unassigned space.

Ochoco Reservoir, with a much smaller drainage area, most of which is in forest cover, would have an even smaller silt problem than the proposed Prineville Reservoir. Its capacity, consisting of 1,400 acrefeet dead storage, 30,000 acre-feet irrigation and joint use storage and 16,500 acre-feet for exclusive flood control storage, is considered ample to accommodate the small amount of sediment which may be expected.

RESERVOIR PERFORMANCE

Operation studies for Ochoco and Prineville Reservoirs on the basis of past runoff and the foregoing estimates of project demands, upstream depletion, divertible flood flows, reservoir losses, and flood storage reservations show that an active storage capacity of 30,000 acrefeet at Prineville used in conjunction with 30,000 acre-feet of capacity at Ochoco would adequately supply the project area of 20,210 acres. The additional 123,000 acre-feet of active capacity in Prineville Reservoir, 60,000 acre-feet of which would be utilized jointly for flood control, would provide an average additional supply of 49,300 acre-feet in most years. Thus Prineville Reservoir could be operated to meet a total irrigation draft of 80,700 acre-feet per year. Performance of the two reservoirs so determined is shown graphically on the reservoir operation diagram for the period 1928-39, which includes the critical runoff period and a subsequent period containing 1 year of high runoff and an isolated year of low runoff.

Qchoco Reservoir

Ochoco Reservoir would have filled in 5 years during the 12-year period. Spillway waste or flood control releases would have been necessary in 9 years, although they would have exceeded 7,000 acre-feet in only 2 years. Waste for 1938 would have been about 41,500 acrefeet, of which about half may have resulted from the flood in March and April. In addition to the spillway waste and flood control releases, reservoir seepage during the storage season would have wasted an annual maximum of about 2,000 acre-feet, if past seepage conditions prevailed.

Ochoco Reservoir would have supplied full irrigation diversion demand of 13,500 acre-feet in all but 1 year in the critical period. The shortage in 1931 was 2,400 acre-feet. From 1940 to date, the 30,000 acre-feet of active storage at Ochoco Reservoir could have supplied the irrigation demand without shortage.

Prineville Reservoir

During the 12-year period Prineville Reservoir would have filled to capacity in 1928 and 1938, which indicates virtually complete control of the river. There would have been spillway waste or releases to provide reservation of joint-use capacity in only 3 of the 12 years. Maximum annual waste would have been about 217,000 acre-feet, over half of which is controlled release of the April 1938 flood. The waste subsequent to 1938 comes when water supply is plentiful.