prepared for
Introduction
River Flow Loss Analysis
Reach Losses
Minimum Flow Delivery
Worst-Case Minimum Flow
Condition
Maximum Flow Delivery
Analog Model
Duration, Timing, and
Point of Delivery
Stream Gaging Requirements
Footnotes
This technical memorandum describes the manner in which minimum (base) flows and maximum (riparian) flows will be delivered to the Lower Owens River as well as the number and location of gaging stations (tasks II.A.1 and 6 of the work plan). It is important to note that the plan described here is only intended as the initial mechanism for delivering water to the river. The cornerstone of managing the Lower Owens River ecosystem is adaptive management, an approach that will allow us to improve upon stream flow inputs with knowledge gained from long-term monitoring. This initial hydrological plan is established on "worst-case" conditions of water losses, anticipated ecological corridors to link wetland and riverine habitats, and the results of mathematical simulations (modeling) of instream flows. Clearly the information base will change and improve with monitoring results; one must anticipate that the flow delivery system will be modified from time to time to improve the efficiency of delivering minimum and maximum flows and the conservation of water resources.
The controlled flow study performed in 1993 (Go to Footnote 1) provided the basis for (1) establishing optimum flows for target fish species, (2) establishing optimum out-of-channel flows for riparian vegetation and instream habitat, and (3) determining the channel response to different flow levels. The controlled flow study was a mathematical simulation of both fish habitat and channel geomorphological and hydrological response, and, while models are by nature inexact because too few variables can be modeled, the results do establish reliable starting points for ecosystem management. The controlled flow study indicated that a base flow of 40 cfs (a year around minimum flow) will provide optimum habitat for target fish species and a freshet flow of up to 200 cfs will provide optimum water spreading to create and maintain riparian habitat. The controlled flow study also provided a rough insight into water losses and gains in discrete river reaches from the intake to below Keeler Bridge.
River channels are not efficient conduits of water, nor can river channels create uniform flow conditions at all points along its course. Conditions favoring uniform flow in natural channels are rare compared with the well-controlled flow conditions in concrete canals or irrigation ditches. Flow in natural channels is typically varied, unsteady, turbulent, and subcritical (Go to Footnote 2). In non-uniform or varied flow the water volume, depth and/or velocity change over distance (Go to Footnote 3); examples are where the flow moves through a bedrock constriction or passes from a pool to a riffle. Loss of water to groundwater aquifers and streambank storage represents the most significant influence on flow variation. Such is the case in the Lower Owens River where extreme variations in flow conditions (volume, depth, and velocity) were noted for different reaches during the controlled flow study. The most important influence on flow variation in the Lower Owens River is water loss from one reach to another.
Inyo County performed a detailed analysis of flow changes by reach in the Lower Owens using data collected during the controlled flow study in 1993 (Go to Footnote 4). (Figure 1 provides a visual reference for key river locations.) Using discharge data from eight metered sections of the river and flows from various spill gates, Inyo County developed hydrographs, rating curves, wetting front velocities, and peak discharge wave velocities for each reach of the river. Discharge data was then used to estimate a water balance for the river during the 41-day controlled flow study. Results of Inyo County’s water loss analysis are shown in Table 1.
TABLE 1. Water losses in the Lower Owens River as determined by integration of station hydrographs and subtraction of the downstream station total discharge from the upstream station total discharge (from Jackson 1994).
| REACH | REACH LENGTH (miles) |
TOTAL LOSS (acre-ft) |
LOSS DURATION (days) |
LOSS RATE (cfs) |
LOSS RATE (cfs/mile) |
| Intake to Blackrock | 4.82 | 1300.66 | 37 | 17.72 | 3.68 |
| Blackrock to Goose Lake | 6.56 | 277.11 | 40 | 3.49 | 0.53 |
| Goose Lake to Five Culverts | 5.87 | 1210.52 | 36 | 16.95 | 2.89 |
| Five Culverts to Mazourka | 6.07 | 715.17 | 38 | 9.49 | 1.56 |
| Mazourka to Manzanar Reward | 8.80 | 402.61 | 38 | 5.34 | 0.61 |
| Manzanar Reward to Reinhackle Spring |
6.55 | 887.19 | 38 | 11.77 | 1.80 |
| Reinhackle Spring to Lone Pine Road | 10.94 | 664.47 | 41 | 8.17 | 0.75 |
| Lone Pine Road to Keeler Bridge | 6.05 | 72.47 | 41 | 0.89 | 0.15 |
| TOTAL | 55.66 | 5530.20 | 73.83 |
Table 1 indicates that the principle lossing reaches are from the Intake to Blackrock Ditch and from Goose Lake to Five Culverts; other reaches also lost a substantial amount of flow during the study. However, the water balance during the controlled flow study would indicate that all reaches of the Lower Owens River are losing reaches. In addition to water lost to the stream channel one must factor in water loss from evapotranspiration. A 1912 study of evapotranspiration loss from Charlie’s Butte to Mt. Whitney Bridge (about 53 miles) indicated a total loss of 10cfs or 0.19 cfs/mile for this reach (Go to Footnote 5). The Inyo County study factored in evapotranspiration losses for the river and spill gates using vegetation inventory data bases for the river and adjacent wetlands. Evapotranspiration combined with channel water losses for the river and spill gates amounts to 36,341 ac-ft/yr at a flow of 40cfs.
It is important to understand that these losses are derived from the controlled flow study in which water balance or equilibrium was never reached. Although the data used by Inyo County is empirical it does not represent actual flow losses once a steady-state flow condition is achieved in the Lower Owens River. Quoting from the Inyo County study:
Changes induced by pumping and annual hydrologic variation in shallow groundwater systems adjacent to the river will also affect streamflow. To what degree groundwater pumping affects surface flow remains to be seen. A 1996 review of shallow groundwater levels in the Lower Owens River Valley (Go to Footnote 6) indicated positive net groundwater storage changes in the Blackrock, Independence and Lone Pine reaches of the valley, and a net negative change in the Manzanar and Union Wash portions of the valley. Surface flow in the Lower Owens River will probably improve the volume of stored groundwater adjacent to the river in the Manzanar and Union Wash areas; providing groundwater pumping does not outpace the rate of recharge.
Given that the empirical data from the controlled flow study represents the high end, or worst-case conditions for flow losses, and the 1912 study on river evapotranspiration represents the low end, we can define a range of flow losses as 0.2 cfs/mile to 3.7 cfs/mile. Within this range (as shown in Table 1) there is great variability between reaches; the Lone Pine Road to Keeler Bridge may not even be a losing reach and the river below Independence spill gate is typically a gaining reach in the winter. Neither end of the range of flow loss is acceptable for planning the initial flow delivery system for the Lower Owens River Project
The actual water balance in the Lower Owens River will be known in time and will be a function of the filling of aquifers and bank storage along the channel as well as the climax stage of riparian vegetation and aquatic macrophytes. While it is not possible to accurately predict how long it will take for the river flow to reach equilibrium or steady-state conditions, experience with the Owens River Gorge re-watering project indicate that the Lower Owens River might reach flow equilibrium in a few years. In the meantime an adaptive management approach is needed to ensure that minimum flows are maintained throughout the Lower Owens River.
The goal of providing year-around minimum (base) flow to the river is to achieve as close to 40 cfs as possible in all reaches of the river. The foregoing discussion makes it clear that it is impossible to achieve exactly 40 cfs at all points in the Lower Owens River. Nevertheless, it is possible to meet the minimum flow throughout the river with a reasonable amount of variability in time once bank storage and groundwater aquifers are filled in each reach. The initial goal of water delivery must, therefore, be to fill aquifers in losing reaches and adjust discharge as necessary once a predictable minimum flow equilibrium or steady-state condition is attained.
The first year of flow release will be a "donor" year. Approximately 40-50 cfs will be released beginning in mid-March or the first of April (the first spring frost-free period) to avoid problems associated with icing in the channel (Go to Footnote 7). Flows will be allowed to fill shallow, near-channel aquifers, wetlands, and bank storage areas. An out-of-channel or riparian flow of 200 cfs will be released in the freshet period (typically late May or early June) to recharge upper terrace aquifers. The first year is a water donation year because the goal is not to gage riparian flows to promote seeding of riparian plants but to simply reach flow equilibrium as early as possible.
When riparian vegetation becomes established and begins serrel stage development we may see seasonal as well as diurnal/nocturnal changes in flow magnitude during plant growth. Consequently, it is important to fill depleted water storage systems along the river channel in the early years so that other influences (vegetation pumping, evapotranspiration, groundwater pumping) can be accounted for independent of recharge.
The second year of flow release will also be a donor year, but the second year will allow a closer evaluation of where the "weak spots" (losing reaches) are and provide better projections of time to fill aquifers. Like the first year, second year releases (minimum and maximum flows) will not be intended to promote vegetation but to further fill aquifers and enhance bank stabilization.
The third year of flow releases will focus on attaining
approximately 40 cfs minimum flow by adjusting the delivery system as needed. This may
entail discharge from some spill gates to augment flows in different river reaches.
Adaptive management using stream gage monitoring results will determine the final minimum
flow delivery system. The worst-case condition that we can anticipate once the river has
reached flow equilibrium is a continuous average loss of 1cfs/mile (Go to
Footnote 8). If that circumstance arises flow augmentation from spill gates will be
required.
Assuming the river loses an average of 1 cfs/mile after the equilibration period, the flows can be maintained as close to 40 cfs as possible in all reaches by balancing augmentation flows from spill gates as shown in Figure 2. This figure is only a hypothetical illustration of how minimum flows can be augmented and is based on a loss of 1 cfs/mile. The majority of the minimum flow is released at the intake and flow losses within reaches are augmented with release from spill gates along the river. The schematic shows the beginning and ending flow in each of eight different river reaches. Flows within reaches will vary between the flow at the top and the flow at the bottom of the reach, but, in most cases, the minimum flow of 40 cfs will be met without sacrificing water conservation. Certainly the balancing shown in Figure 2 can be done many different ways as long as the goals of maintaining 40 cfs with minimal variation within reaches and water conservation are met.
As stated previously, it is reasonable to expect river flows to attain equilibrium within a short period of time, thus precluding the need for long term augmentation flows along the lines described for the worst-case condition. Adaptive management based on monitoring of river discharge will allow changes in augmentation flow volumes and release points as the river reaches steady-state conditions, but our initial effort will be to allow flows to reach equilibrium and then determine the need for augmentation flows in specific reaches.
The maximum flows to be delivered to the Lower Owens River are out-of-channel flows associated with freshet periods (rapid snow melt in the spring). Out-of-channel flows (termed riparian flows) are essential to the life of rivers because they create a disturbance regime that results in instream habitat diversity and the creation and maintenance of riparian habitat (Go to Footnote 9).
The controlled flow study concluded that riparian flows should emulate the natural hydrology to the degree possible (up to a maximum discharge of 200cfs) so that the lower river experiences the same wet year-dry year cycles as the upper river (above the intake). The variability of the natural hydrologic cycle in the river below the intake will be achieved with an analog model that mimics freshet conditions in the river above the intake.
We originally anticipated that pulse flows would be determined as a ratio of annual flows in the Owens River above the intake. Establishing a ratio between flows above and below the intake to meet out-of-channel requirements is dependent upon knowing both bankfull discharge levels and the ratio of frequency of occurrence between Q10 and Q2 flood intervals (Go to Footnote 10). However, analysis indicated that without long-term measurement of bankfull discharge in the Lower Owens River the equations used in determining the annual flow ratio would not be reliable and normal water year maximum flows could be underestimated. Consequently, a maximum flow method adopted in the Mono Basin based on annual runoff forecasting will be used.
The LADWP prepares runoff forecasts each year to assist in
determining the amount of water expected to be available for the aqueduct. The forecasts
correspond to the runoff year which extends from April 1 through March 31. The forecasts
are made near the first of the month in February, March, April, and May. The LADWP uses
precipitation data, snow survey data, and weather forecasts as the basis for their Runoff
Forecast Model.
The basis for decision making on the annual riparian flow discharge
to the Lower Owens River will be projections of runoff year type classification using the
LADWP Runoff Forecast Model. Riparian flows to the Lower Owens River for each runoff year
by classification are as follows:
Extreme Water Year: 300 cfs
Wet Water Year: 250 cfs
Wet-Normal Water Year: 200 cfs
Normal Water Year: 200 cfs
Normal-Dry Water Year: 75 cfs
Dry Water Year: 0 cfs
In wet years extraordinary runoff may provide the opportunity to
release more than 200 cfs to the Lower Owens. In dry years no riparian flows will be
released but the base flow of 40 cfs will be maintained.
It is anticipated that riparian flows will be delivered as intake releases only. In their analysis of wave velocity during the controlled flow study Inyo County determined that the peak flow (155cfs) travel time was about 17 ft/min and required 12 days to reach Keeler Bridge from the intake. However it should be noted that this was done during an extremely dry conditions and with aquifers unfilled. Wetting front velocities in the dry channel were roughly half of the velocity of the peak discharge in a wetted channel. Consequently, it can be expected that in time, riparian flows will require something less than 12 days to travel from the intake to the pumpback station. The duration of riparian flows however will be determined through adaptive management.
Timing riparian flows to enhance willow/cottonwood seed dispersal provides some selective advantage over salt cedar and optimizes the seeding and germination of native riparian plant species. Annual seed development will vary from the upper reaches of the Lower Owens River to the most downstream reaches. Consequently, riparian flow timing should be based on the reach of river where seed development is latest.
The placement of gaging stations in the river is critical to monitoring streamflow. Without an adequate spread of metered sections it will be difficult to determine changes in water loss by reach and to implement adaptive management strategies over time. The initial water release plan will be monitored with a spread of gaging stations covering more channel reaches than will be necessary once the river flow reaches equilibrium. Consequently, in the initial years, temporary gage stations will be established; once it is determined which reaches of the river require long-term monitoring those gage stations will be made permanent and the unnecessary stations removed.
The initial water release plan will begin with 9 temporary river
gaging stations. The locations for gages are as follows:
Gaging stations above and below the pumpback station will monitor the discharge entering the pumping facility and the discharge allowed to flow to the Delta.
1. Hill, M., W.S. Platts, S. Jensen, and G. Ahlborn. 1994. Data base and modeling results for the lower Owens River Project: controlled flow study. LADWP, Bishop, CA.
2. N. Gordon, T. McMahon, and B. Finlayson. 1993. Stream Hydrology. J. Wiley & Sons, NY.
3. Brookes, A. 1994. River channel change. Pages 55-75 In,Calow, P. and Petts, G., eds., The Rivers Handbook, Vol.2. Blackwell Scientific Publ., Cambridge, MA.
4. Jackson, R. 1994. Lower Owens River planning study: discharge data and preliminary estimates of losses for the lower Owens River. Inyo County Water Dept., Bishop, CA.
5. Lee, C.H. 1912. An intensive study of the water resources of a part of the Owens Valley, California. USGS Water Supply Paper 294.
6. Jackson, R. 1996. Shallow groundwater levels in the Owens Valley: 1995 update. Inyo County Water Dept., Bishop, CA.
7. Icing, particularly anchor ice, can accelerate bank erosion when one side of the channel retains frozen soil particles and the other does not; icing conditions also inhibit groundwater recharge. Thus, the first year's flow is initiated in the early spring to allow maximum groundwater and bank storage recharge prior to the onset of winter.
8. Given that the empirical data from the controlled flow study represents the high end for flow losses and Lee's 1912 study on river evapotranspiration represents the low end, we can define a range of flow losses as 0.19 to 3.68 cfs/mile. Within this range (as shown in Table 1) there is great variability between reaches. Neither end of the range of flow loss is acceptable for planning the initial flow delivery system. However, by convention and general agreement between hydrologists familiar with the river (personal communication with S. Keef, Chief Hydrographer for LADWP and R. Jackson, Inyo County Hydrologist), it is usually assumed that the river in its current condition loses about 1 cfs/mile on average.
9. See Hill, M., and W. Platts. 1995 Lower Owens River Watershed Ecosystem Management Plan: Action Plan and Concept Document. LADWP, Bishop, CA., for a detailed discussion of the role of multiple flow regimes and riparian habitat in the functioning of riverine-riparian ecosystems.
10. Methodology described by Remy, M. 1994. In, Calow, P. and Petts, G., eds., The Rivers Handbook, Vol.2, Blackwell Scientific Publ., Cambridge, MA.