October 11, 2000

 Laws Area Hydrology and Canal Operations

Robert F. Harrington, Ph.D.
Hydrologist
Inyo County Water Department

 

The following five questions are addressed in the material below:

What have past practices since 1970 been on the McNally canals?

How has the depth to water fluctuated beneath vegetation parcels that are significantly below baseline?

What is the relation between flows in the McNally canals and water table fluctuations in the Laws area?

How has LADWP’s divergence from past practices affected water levels in the Laws area?

How will operating the McNally canals in the future promote water table recovery?

 

1. Past practices of operation of the McNally canals. Diversions into the Upper and Lower McNally canals from the Owens River (Figure 1) are given in Table 1. The frequency and volume of diversions between 1970 and 1995, and their distribution and use along the McNally canal system define past practices of operation of the McNally canals since 1970. During the interval 1996 through 2000, LADWP has diverged from its past practice by not running a significant amount of water in the McNally canals during years with less than 140% of normal runoff; therefore these years should not be included in the definition of past practices. The nature of the divergence from past practices during 1996-2000 is discussed later in this document (section 4).

Table 1. Diversions from the Owens River to the McNally canals. Runoff year is from April 1 to March 31 of the following year. Data are from LADWP Totals and Means Report and annual runoff summaries for 1997, 1998, and 1999.

*Diversions based on LADWP’s stated intent to not operate McNally canals in 2000

**Estimated from LADWP runoff forecasts.

 

2. Depth to water beneath parcels with vegetation significantly below baseline. The best information regarding depth to water in the Laws area is from LADWP shallow monitoring wells. These data were transmitted to the Inyo County Water Department from LADWP in November, 1999 (letter from Mr. Coufal to Mr. James, November 18, 1999). Figures 2 through 6 show the depth to water beneath vegetation parcels that are significantly below baseline vegetation cover according to the year 2000 vegetation survey. The location of each parcel and monitoring well referred to in Figures 2 through 6 is given in Figure 1. Because not all vegetation parcels contain shallow monitoring wells, and because the depth to water may vary across a parcel, the depth to water beneath each parcel must be estimated from nearby monitoring wells. The hydrograph given for each parcel was interpolated from nearby monitoring wells, as described in Harrington and Howard (2000). For purposes of comparison, for each parcel hydrograph, a hydrograph of a selected nearby monitoring well is shown. The parcel hydrographs and monitoring well hydrographs show similar patterns, attesting that the interpolated hydrographs are a reasonable estimate of depth to water beneath the vegetation parcels. Four features are common to each parcel hydrograph: (1) a period of high water table during 1985 through 1987, (2) a period of decline resulting in a depressed water table in the late-1980’s and early-1990’s, (3) a period of water table recovery peaking in 1999, and (4) a decline from 1999 to 2000. The pattern of a high water table during the mid-1980’s, a sharp decline in the late-1980’s, and gradual recovery in the 1990’s is typical of shallow monitoring wells in LADWP well fields throughout the Owens Valley.

Click on chart to enlarge.

Figure 2. Depth to water beneath parcel LAW052. Note that well 494T is dry for long periods between 1988 and 1998.

---

Click on chart to enlarge.

Figure 3. Depth to water beneath parcel LAW062. Note that well 577T is dry for long periods between 1989 and 1993.

---

Click on chart to enlarge.

Figure 4. Depth to water beneath parcel LAW082. Note that well 577T is dry for long periods between 1989 and 1993.

---

Click on chart to enlarge.

Figure 5. Depth to water beneath parcel LAW085. Note that well V001G is dry for long periods between 1988 and 1998. Note that parcel LAW085 occupies a topographic position intermediate between wells 435T and 492T, hence the intermediate depth to water. Well V001G is within the parcel boundaries.

---

Click on chart to enlarge.

Figure 6. Depth to water beneath parcel LAW112. Note that well 575T is dry between 1988 and 1995.

---

3. Linkage between McNally Canal operations and water table fluctuations. When water is diverted from the Owens River into the McNally canals, seepage from the canal channel contributes an important amount of groundwater recharge to the Laws area (Green Book, p. 106-108 and Table IV.7). The recharge source for the Laws area differs from other LADWP well fields in the Owens Valley. Most well fields are recharged primarily by seepage from stream channels issuing from the Sierra Nevada (e.g., Green Book, Table IV.13), but in the Laws area, the only streams that contribute recharge are the relatively small streams of the White Mountains and Fish Slough. Green Book Table IV.3 attributes annual average rates of 4.025 cfs to the upper McNally canal and 3.330 cfs to the lower McNally canal, for a total recharge rate of 7.355 cfs recharged to the groundwater system from canal seepage when both canals are operated, which equates to a rate of 5315 acre feet per year. Another estimate of the magnitude of recharge to the Laws area can be calculated from Danskin’s (1998, p. 51) statement that large canals of the Owens Valley such as the upper McNally canal can sustain seepage rates of 1.1 cfs per mile for periods of several months. The upper and lower McNally canals comprise a system of about 20 miles of canal channel, which results in a seepage rate of 22 cfs for the system. If the canals produced this seepage rate for an entire year, 15,900 acre feet of recharge would result, which represents an approximate upper bound to the recharge possible due to canal seepage. Not all of the water diverted into the McNally canals is consigned to raising the water table and satisfying phreatophyte consumptive demands; a portion of the water returns to the river/aqueduct system by groundwater flow or via the Laws Return Ditch. When water tables in the Laws area are raised, the hydraulic gradient from the Laws area to the Owens River is increased, which increases the rate of groundwater flow into the river channel, thus it is incorrect to assume that the entire volume of recharge induced by operating the McNally canals represents a loss of that entire volume from the river/aqueduct system.

A rigorous water budget for the Laws area has not been assembled, but to put these figures in context, they can be compared with other components of the water budget for the Laws area. Underflow from Chalfant Valley is estimated to be approximately 2500 acre feet per year (Green Book, Table IV.12), recharge from streams in the Laws area in a year of normal runoff is about 1270 acre feet per year (Green Book, Tables IV.1, IV.11, equation IV.1), and runoff-year groundwater pumping in the Laws area has ranged as high as 38,841 acre feet (in 1988). Water spreading and irrigation are also part of the distribution and use of canal flows within the canal system, and these activities also contribute to recharge beyond seepage directly from the canal channel (Green Book, Tables IV.10 and IV.12).

When change in depth to water (the difference between the fourth and second columns in Table 2) is compared to diversions into the McNally canals, there is revealed a clear tendency for the water table to rise when the canals are operated (Figure 7). This indicates that there is a connection between canal operations and recovery of the water table. Variables such as groundwater pumping, irrigation, water spreading, runoff, and river stage also influence the water table, but it is clear from Figure 7 that canal operations exert a strong influence on the water table in the Laws area.

Click on chart to enlarge. 

Figure 7. Change in depth to water (or water table elevation) in Laws area indicator wells (April to subsequent April), and runoff-year diversions from the Owens River to the McNally canals. The data are derived from Table 2 by subtracting the second column from the fourth and converting to meters.

Relationships between hydrologic variables such as shown in Figure 7 allow predictive models to be developed based on past observations of hydrologic variables. The County and LADWP have used a network of shallow monitoring wells as indicator wells to assess the effect of groundwater extraction and recharge on water table elevation (Harrington, 1999; Harrington, 1998; Jackson, 1996; Kavounas, 1993; Jackson, 1992). These wells were chosen based on their ability to predict the water table’s response to pumping stress and recharge. Elsewhere in the Owens Valley, LADWP and Inyo County have developed multiple linear regression models by regressing April water table elevation against pumping within a wellfield and Owens Valley runoff. However, in the Laws area, recharge is largely derived from seepage from the McNally canals or from water spreading from diversions from the McNally canals; therefore diversions into the McNally canals from the Owens River is a better predictive variable in the Laws area than Owens Valley runoff.

Indicator wells in the Laws area are 107T, 436T, 438T, 490T, 492T, and 493T (Figure 1). The data used to develop the regression models are given in Table 1. These data were taken from LADWP’s Totals and Means Reports and Monthly Well Reports.

The multiple linear regression model for each indicator well has the form:

where h₁ is the predicted water table elevation at the end of the runoff year, h₀ is the observed water table elevation at the start of the runoff year, P is the pumping in the Laws wellfield during the runoff year in acre feet, and Q is the flow diverted from the Owens River into the McNally canals during the runoff year in acre feet. a₀, a₁, a₂ and a₃ are regression coefficients (Table 3). The regression coefficients are derived by the widely used technique of multiple linear regression (Holder, 1985). All the regression coefficients presented in Table 3 are significant at a level of P < 0.05, except for the a₂ for well 490T, which is significant at a P<0.08 level. Prediction uncertainty is assessed using Monte Carlo simulation as described in Harrington (1998). The largest component of the uncertainty is due to the standard error of the regression, which is given for each model in Table 3.

The success of the regression models depends on there being a correlation between water table fluctuations, groundwater pumping, and operation of the McNally canals. Table 4

gives the Pearson’s correlation coefficient for change in depth to water in Laws area indicator wells (Table 2), groundwater pumping in the Laws area, and diversions from the Owens River into the McNally canals. Groundwater pumping and water table fluctuations are significantly and negatively correlated, i.e., groundwater pumping is correlated with water table declines. The correlation between water table fluctuations and McNally operations is positive and slightly larger than the correlation between pumping and water table fluctuations, attesting to the importance of canal operations in providing recharge. These results confirm what common sense and elementary hydrology would suggest: that groundwater pumping reduces groundwater levels and recharge increases groundwater levels.

Table 2. Data used for development of multiple linear regression models. Initial water table is the elevation above sea level of the water table measured during April of the year in the first column; pumping is runoff-year (April 1 through March 31) pumping for the Laws wellfield (acre feet); diversions to canals are the runoff-year diversions from the Owens River into the upper and lower McNally canals (acre feet); final water table is the water table elevation at the end of the runoff year. The fifth column is regressed against the second, third and fourth columns.

Monitoring well 107T; RP elevation: 4156.1 ft; Land surface elevation: 4154.8 ft
Year	Initial water table	Pumping	Diversions to canals	Final water table
1972	4126.75	28345	0	4117.09
1973	4117.09	15974	5234	4119.23
1976	4126.55	16285	0	4121.95
1977	4121.95	15038	679	4119.32
1978	4119.32	945	18367	4126.80
1979	4126.80	17933	6861	4120.55
1980	4120.55	1251	29132	4132.28
1981	4132.28	25313	3290	4121.01
1982	4121.01	1388	41570	4135.52
1983	4135.52	1113	32492	4137.25
1984	4137.25	7403	6589	4134.80
1985	4134.80	17369	4117	4129.60
1986	4129.60	8600	28204	4131.90
1987	4131.90	38241	0	4119.00
1998	4121.60	483	29410	4132.60
1999	4132.60	1674	0	4129.60
2000	4129.60

 

Monitoring well 436T; RP elevation: 4107.5; Land surface elevation: 4106.3
Year	Water table elevation	Pumping	Diversions to canals	Final water table
1976	4095.23	16285	0	4091.86
1977	4091.86	15038	679	4090.75
1978	4090.75	945	18367	4095.60
1979	4095.60	17933	6861	4093.65
1980	4093.65	1251	29132	4099.62
1981	4099.62	25313	3290	4094.73
1982	4094.73	1388	41570	4102.24
1983	4102.24	1113	32492	4101.53
1984	4101.53	7403	6589	4100.00
1985	4100.00	17369	4117	4097.40
1986	4097.40	8600	28204	4099.90
1987	4099.90	38241	0	4094.20
1988	4094.20	38841	3743	4090.30
1989	4090.30	34785	173	4088.70
1993	4089.10	12618	18082	4092.80
1994	4092.80	16187	26	4092.10
1995	4092.10	8249	22797	4096.60
1996	4096.60	11199	0	4094.80
1997	4094.80	2951	4892	4096.00
1998	4096.00	483	29410	4100.50
1999	4100.50	1674	0	4098.70
2000	4098.70

 

Monitoring well 438T; RP elevation: 4142.1 ft; Land surface elevation: 4138.9 ft
Year	Initial water table	Pumping	Diversions to canals	Final water table
1976	4129.44	16285	0	4127.66
1977	4127.66	15038	679	4126.96
1978	4126.96	945	18367	4131.21
1979	4131.21	17933	6861	4128.16
1980	4128.16	1251	29132	4133.26
1981	4133.26	25313	3290	4128.09
1982	4128.09	1388	41570	4136.37
1983	4136.37	1113	32492	4134.97
1984	4134.97	7403	6589	4133.2
1985	4133.2	17369	4117	4131.9
1986	4131.9	8600	28204	4132.4
1987	4132.4	38241	0	4125.3
1988	4125.3	38841	3743	4122
1989	4122	34785	173	4122.4
1990	4122.4	16933	0	4122.9
1991	4122.9	10949	0	4123.7
1992	4123.7	10562	0	4124.6
1993	4124.6	12618	18082	4124.7
1994	4124.7	16187	26	4124.7
1995	4124.7	8249	22797	4127.5
1996	4127.5	11199	0	4126.3
1997	4126.3	2951	4892	4125.1
1998	4125.1	483	29410	4131.3
1999	4131.3	1674	0	4128.1
2000	4128.1

 

Monitoring well 490T;RP elevation: 4078.3 ft; Land surface elevation: 4077.3 ft
Year	Initial water table elevation	Pumping	Diversions to McNally canals	Final water table elevation
1976	4059.85	16285	0	4058.39
1977	4058.39	15038	679	4057.86
1978	4057.86	945	18367	4061.81
1979	4061.81	17933	6861	4061.19
1980	4061.19	1251	29132	4064.33
1981	4064.33	25313	3290	4062.48
1982	4062.48	1388	41570	4067.34
1983	4067.34	1113	32492	4068.26
1984	4068.26	7403	6589	4065
1985	4065	17369	4117	4063.6
1986	4063.6	8600	28204	4067.1
1987	4067.1	38241	0	4064.1
1988	4064.1	38841	3743	4061.1
1989	4061.1	34785	173	4058.2
1990	4058.2	16933	0	4056.9
1991	4056.9	10949	0	4056.4
1992	4056.4	10562	0	4056.5
1993	4056.5	12618	18082	4057.9
1994	4057.9	16187	26	4058.2
1995	4058.2	8249	22797	4061.9
1996	4061.9	11199	0	4061.1
1997	4061.1	2951	4892	4060.3
1998	4060.3	483	29410	4062.8
1999	4062.8	1674	0	4062
2000	4062

 

Monitoring well 492T; RP elevation: 4130.1 ft; Land surface elevation: 4128.4 ft
Year	Initial water table	Pumping	Diversions to canals	Final water table
1977	4074.47	15038	679	4073.76
1978	4073.76	945	18367	4093.66
1979	4093.66	17933	6861	4082.93
1980	4082.93	1251	29132	4100.18
1981	4100.18	25313	3290	4086.39
1982	4086.39	1388	41570	4104.65
1983	4104.65	1113	32492	4106.91
1984	4106.91	7403	6589	4102.7
1985	4102.7	17369	4117	4091.3
1986	4091.3	8600	28204	4097.7
1987	4097.7	38241	0	4081.2
1988	4081.2	38841	3743	4072.5
1993	4079.6	12618	18082	4088.2
1994	4088.2	16187	26	4085.8
1995	4085.8	8249	22797	4092.5
1996	4092.5	11199	0	4090
1997	4090	2951	4892	4093.5
1998	4093.5	483	29410	4102.9
1999	4102.9	1674	0	4101.1
2000	4101.1

 

Monitoring well 493T; RP elevation: 4133.2 ft; Land surface elevation: 4131.6 ft
Year	Initial water table	Pumping	Diversions to canals	Final water table
1977	4093.08	15038	679	4090.55
1978	4090.55	945	18367	4107.35
1979	4107.35	17933	6861	4102.39
1980	4102.39	1251	29132	4116.78
1981	4116.78	25313	3290	4107.12
1982	4107.12	1388	41570	4122.72
1983	4122.72	1113	32492	4122.66
1984	4122.66	7403	6589	4118.6
1985	4118.6	17369	4117	4110.8
1986	4110.8	8600	28204	4117.8
1987	4117.8	38241	0	4103
1988	4103	38841	3743	4094.2
1989	4094.2	34785	173	4082.1
1990	4082.1	16933	0	4083.7
1991	4083.7	10949	0	4088.3
1992	4088.3	10562	0	4087.4
1993	4087.4	12618	18082	4103.2
1994	4103.2	16187	26	4097.7
1995	4097.7	8249	22797	4106.5
1996	4106.5	11199	0	4100.5
1997	4100.5	2951	4892	4105
1998	4105	483	29410	4118.8
1999	4118.8	1674	0	4114.2
2000	4114.2

 

Table 3. Regression coefficients and statistics for multiple linear regression models. n is number of data, r² is the adjusted coefficient of determination, and SE is the standard error of the regression.

	Well
Coefficient	107T	436T	438T	490T	492T	493T
a0	1.94E+03	1.84E+03	1.94E+03	1.14E+03	2.09E+03	1.54E+03
a1	5.31E-01	5.51E-01	5.30E-01	7.19E-01	4.89E-01	6.26E-01
a2	-3.39E-04	-1.25E-04	-1.04E-04	-3.60E-05	-4.99E-04	-3.31E-04
a3	2.41E-04	1.51E-04	1.55E-04	1.41E-04	2.90E-04	4.26E-04
n	16	21	24	24	19	23
Multiple r	0.97	0.97	0.94	0.98	0.97	0.97
SE (feet)	1.86	0.98	1.47	0.83	2.68	3.01

Table 4. Pearson’s correlation coefficients for change in depth to water in Laws area indicator wells (Dh), groundwater pumping in the Laws well field (P), and diversions from the Owens River into the McNally canals (Q).

	Dh	P	Q
Dh	1.00	-0.62	0.68
P		1.00	-0.56
Q			1.00

 

4. The relationship between LADWP’s divergence from past practices in operation of the McNally canals and water table fluctuations in the Laws area. In the context of the Agreement’s provisions for vegetation management and surface water, past practices of surface water conveyances refers to the frequency of operation and the volume of water diverted into conveyances, and the distribution and use of those diversions within the conveyance system.

In Figure 8, canal diversions are plotted against percent of normal Owens Valley runoff. Since 1996, LADWP’s has operated the McNally canals less frequently and has reduced the volume of water diverted. During the interval 1970-1995, it was LADWP’s practice to run the McNally canals when the runoff was more than approximately 80% of normal, with diversions increasing as runoff increased (Figure 8). The period 1996 through 2000 diverges from this practice. No water was diverted into the canals during runoff-year 1996 despite 131% of normal runoff, and a modest 4892 acre feet were diverted into the canals in runoff-year 1997 despite a 120% of normal runoff year. 1999 saw no diversions into the canals and LADWP has stated they do not plan to run the canals during 2000, despite the fact that is has been LADWP’s past practice to run the canals in 80-100% of normal runoff years. The pattern of diversions into the McNally canals since 1996 is consistent with LADWP’s verbal statement that the McNally canals would henceforth be run only when Owens Valley runoff was greater than 140% of normal, i.e., the only year since 1996 that has conformed to past practices is 1998, a 145% of normal runoff year. Examination of Figure 8 suggests that adherence to past practices would result in diversions into the McNally canals of 10,000 to 20,000 acre feet per year for runoff-year 1996 and 1997.

The interval of low diversions from 1987 through 1992 (Table 1) is consistent with LADWP’s past practice of not diverting water into the McNally canals in years where percent of normal runoff was less that about 80% (e.g., 1971, 1972, 1976, and 1977).

Click on chart to enlarge.

Figure 8. Runoff-year (April 1 through March 31) diversions from the Owen River into the Upper and Lower McNally canals plotted against percent of normal Owens Valley runoff (data from LADWP Totals and Means Report).

In order to assess the effect that LADWP’s divergence from past practice had on the water table, regression model simulations of runoff years 1996 through 2000 were conducted to examine water table fluctuations under the extant management versus water table fluctuations that would likely have occurred had LADWP adhered to their past practices. Three scenarios were simulated: the first is the actual pumping and runoff that occurred, the second and third use increased diversions into the McNally canals to examine what would have happened had past practices been adhered to. Simulation #2 uses 10,000 acre feet of diversions in 1996 and 1997, and simulation #3 uses 20,000 acre feet of diversions in 1996 and 1997 as input. These two cases roughly bracket the range of diversions that would have occurred under past practices. The pumping stress in each simulation is the same. Pumping for runoff year 2000 is simulated based on the amount of pumping promulgated in LADWP’s pumping plan for 2000. Well 107T was not simulated because it was dry until January 1998.

Table 5. Input data used in simulations (acre feet).

	Simulation
	#1 Actual conditions		#2 Past practices, low diversions		#3 Past practices, moderate diversions
Year	Canals	Pumping	Canals	Pumping	Canals	Pumping
1996	0	11199	10000	11199	20000	11199
1997	4892	2951	10000	2951	20000	2951
1998	29410	483	29410	483	29410	483
1999	0	1674	5000	1674	5000	1674
2000	0	4000	5000	4000	5000	4000

Figures 9 through 13 show the results of the three simulations and the observed water table fluctuations for wells 436T, 438T, 490T, 492T, and 493T (see Figure 1 for the locations of the wells). The indicator wells do not lie within the boundaries of the affected vegetation parcels, so they do not provide parcel-specific simulations of water table fluctuations beneath each parcel; rather, they provide a sample of how the water table in the Laws area responds to the various management scenarios given in Table 3. Comparison between the observed water table fluctuations and the water table fluctuations simulated using the diversions and pumping that actually occurred (simulation #1) provides a sense of how well the models reproduce the water table fluctuations that actually occurred. Other more general measured of model fit are given in Table 3. Simulations #2 and #3 provide estimates of how the water table would have responded had the McNally canals been operated in accordance with past practices in 1996, 1997, 1999, and 2000.

Table 6 summarizes the predicted effect that divergence from past practices will have had on the water table at these five monitoring wells at the end of runoff-year 2000. The effect of the reduced recharge over the period 1996-2000 is estimated to result in reductions in the water table elevation of between 0.45 and 1.85 m below what would have occurred if the McNally canals had been operated in a manner consistent with past practices. The degree of reduction in water table elevation depends on the location of the monitoring well and the amount of flow in the canals (i.e., simulation #2 versus simulation #3). These model results are consistent with the Table IV.7 in the Green Book, wherein it is shown that the McNally canals contribute a substantial amount of recharge to the Laws area, and are consistent with the observed correlation between water table recovery and operation of the McNally canals (Tables 3 and 4). Note that a similar analysis of the model results for runoff-year 1999 (rather than 2000) would yield similar results to those presented in Table 6, and that the model results are consistent with the observed fluctuations in the water table (Figures 9 through 13).

 
Table 6. April 2001 predicted difference in depth to water between simulation #1 (actual practices) and simulations #2 and #3 (adherence to past practices).

	D DTW (meters), April 2001
Well	Simulation #2	Simulation #3
436T	0.50	0.66
438T	0.45	0.56
490T	0.59	0.87
492T	0.77	0.91
493T	1.39	1.85

  

Click on chart to enlarge

Figure 9. Simulation results for well 436T. The difference between simulation #1 and simulations #2 and #3 is due to recharge from the McNally canals (see Table 5).

---

Click on chart to enlarge

Figure 10. Simulation results for well 438T. The difference between simulation #1 and simulations #2 and #3 is due to recharge from the McNally canals (see Table 5).

---

Click on chart to enlarge

Figure 11. Simulation results for well 490T. The difference between simulation #1 and simulations #2 and #3 is due to recharge from the McNally canals (see Table 5).

---

Click on chart to enlarge

Figure 12. Simulation results for well 492T. The difference between simulation #1 and simulations #2 and #3 is due to recharge from the McNally canals (see Table 5).

---

Click on chart to enlarge

Figure 13. Simulation results for well 493T. The difference between simulation #1 and simulations #2 and #3 is due to recharge from the McNally canals (see Table 5).

---

5. Relationship between future operations of the McNally canals and water table recovery. The regression coefficients associated with canal flows, a₃, give the amount of water table rise per acre foot of water diverted into the McNally canals. The values given for this regression coefficient in Table 3 indicate that the water table rises between 0.046 and 0.129 meters (0.151 and 0.426 feet) per 1000 acre feet diverted into the canals. Thus, a diversion of 20,000 acre feet into the McNally canals next runoff year would promote 0.92 and 2.58 m of water table recovery. Areas such as parcels LAW052, LAW082, and LAW085 that are near the canals and areas of water spreading and irrigation would have relatively large rises in the water table. Recovery would vary with location within the Laws area, and would be diminished or reversed if groundwater pumping should occur.

 

---

References

Danskin, W. R., Evaluation of the Hydrologic System and Water-Management Alternatives, USGS WSP 2370-H, 1998.

Green Book for the Long-Term Groundwater Management Plan for the Owens Valley and Inyo County, Technical Appendix to the Inyo County/Los Angeles Department of Water and Power Long-Term Water Agreement, 1990.

Harrington, R. F., Multiple Regression Modeling of Water Table Response to Pumping and Runoff, County of Inyo Water Department Report, 1998.

Harrington, R., Updated Regression Models for Forecasting Pumping-Induced Water Table Fluctuations, County of Inyo Water Department Report, June 1999.

Harrington, R., and C. Howard, Depth to Groundwater Beneath Vegetation Reinventory Parcels, County of Inyo Water Department Report, June 2000.

Holder, R. L., Multiple Regression in Hydrology, Institute of Hydrology, 1985.

Jackson, R., Water Level Predicting Multiple Linear Regression Models Developed for the Eighteen Indicator Shallow Test Wells in Owens Valley, California, ICWD report 92-3, 1992.

Jackson, R., Water Level Predicting Multiple Linear Regression Models Developed for Shallow Groundwater Observation Wells in the Bishop Well Field in Owens Valley, California, Report 96-2 (draft), Inyo Co. Water Dept., 1996.

Kavounas, P., Water Level Predicting Multiple Linear Regression Models Developed for the Eighteen Indicator Shallow Test Wells in Owens Valley, California, memorandum to Randy Jackson, June 13, 1993.