Production-Well Withdrawals

The WCWCD has 13 production wells completed in the Navajo Sandstone that are available to capture both pre-existing groundwater (natural recharge) in Sand Hollow and recharge from Sand Hollow Reservoir (fig. 2). The WCWCD and other water users have pumped groundwater from natural recharge in Sand Hollow for many years, with pumping beginning in the 1970s. The WCWCD groundwater pumping totals are recorded monthly from inline magnetic flow meters installed at each well. Since August 2004, monthly pumping by the WCWCD has generally exceeded 150 acre-feet (acre-ft), except for several months during the winters of 2004–05, 2005–06, 2008–09, and 2010 (fig. 3). Through late 2012, most of this pumping was from wells 8 and 9 (fig. 3), both wells are adjacent to the North Dam. Starting in early 2013, some of the pumping shifted to wells 19, 21, and 23 (fig. 3). From 2004 through 2006, there was minimal pumping from these wells during the winter. Since 2006, pumping has been more constant year-round. Monthly pumping from production wells averaged about 280 acre-ft from March 2006 through December 2018. Smaller amounts have been sporadically pumped from wells 1, 2, and 17. Approximately 46,000 acre-ft were pumped from the WCWCD production wells from January 2004 through December 2018 (fig. 3). As of December 2018, pumped totals by the WCWCD at Sand Hollow have been permitted by the Utah Division of Water Rights as natural recharge in Sand Hollow; these withdrawals are governed by different water rights than water recharged from Sand Hollow Reservoir (withdrawal rights for this artificial recharge have not yet been exercised).

Drain Discharge

Because of the steep hydraulic gradients associated with the hydraulic connection between the reservoir and the underlying Navajo aquifer, some land-surface areas downgradient of the North and West Dams became saturated following construction of the reservoir. In response, three French drains (North Dam drain, West Dam drain, and West Dam Spring drain) were constructed for capturing this shallow groundwater (fig. 2). The timing of excavation and spatial dimensions for the three drains are described in Heilweil and Marston (2011).

The water pumped from these drains is measured with a Tigermag totalizing flow meter (Sparling Instruments in El Monte, California). Discharge from the North Dam drain has been pumped at a relatively consistent rate (about 50 acre-ft per month) since January 2008 (fig. 4); about 10,500 acre-ft were pumped from the North Dam drain between 2003 and 2018. Initially, all of this water was returned to the reservoir, but since 2007, most of the water has been used by Sand Hollow Resort (fig. 1) to meet irrigation demands. About 2,500 acre-ft of water was pumped from the West Dam drain back into the reservoir from 2005 through 2018. Beginning in October 2006, pumping of discharge from the West Dam Spring drain was initiated and has largely replaced the need for pumping of the West Dam drain (fig. 2). From 2006 through 2018, about 32,000 acre-ft of water was pumped from the West Dam Spring drain for municipal use or returned back to Sand Hollow Reservoir. Although groundwater discharge to the West Dam Spring drain likely does not vary greatly, pumping from this drain has been intermittent. The intermittent pumping schedule has resulted in high variability in the monthly reported discharge from the West Dam Spring drain (fig. 4).

Groundwater-Level Data and Reservoir Altitude

Groundwater levels measured in an extensive monitoring-well network surrounding Sand Hollow Reservoir were used to document changes in the potentiometric surface associated with recharge from Sand Hollow Reservoir. The WCWCD measures water levels monthly in 21 monitoring wells completed in the Navajo aquifer (fig. 2; the water district [WD] RJ and WD 4 monitoring wells were removed in 2013 and 2017, respectively). These wells were constructed with either 1- or 2-inch (in.) diameter polyvinyl chloride casing, with perforations along the bottom 5- to 20-foot (ft) length of the casing (Heilweil and others, 2005). Three locations have nested pairs of WD monitoring wells: (1) WD 15 and WD 16, (2) WD 17 and WD 18, and (3) WD 19 and WD 20. The vertical distance between well screens for the nested-pair wells are 243, 79, and 227 ft, respectively. Water levels were measured by the WCWCD using electric-tape water-level indicators. Annual independent check measurements were performed by USGS personnel for quality assurance to ensure accuracy (repeatability of measurements) of the instruments.

Daily reservoir water-level altitude (stage) was recorded from August 2003 through December 2018 using a pressure transducer installed by the WCWCD in the reservoir along the North Dam. Because of periods of transducer malfunction from 2005 through 2007, and from August 2011 to December 2011, daily reservoir stage was interpolated based on monthly measurements recorded at the boat ramp by WCWCD and Sand Hollow State Park personnel, and then correlated with trends from the transducer data.

Recently measured (January 2014 through December 2018) and previously reported (Heilweil and others, 2005, 2009a; Heilweil and Susong, 2007; Heilweil and Marston, 2011, Marston and Heilweil, 2013, 2016; Marston and Nelson, 2018) groundwater levels and reservoir water-level altitude are shown in figure 5. The reservoir stage rose from about 2,980 ft at the beginning of March 2002 to a maximum of about 3,060 ft in May 2006, when the reservoir was first filled to capacity. The reservoir altitude receded to about 3,040 ft in December 2007 as a result of reduced inflows and evaporative losses and then fluctuated between about 3,040 and 3,060 ft from 2008 through 2011. From 2012 through 2014, the reservoir altitude dropped to a fluctuating level between about 3,030 and 3,050 ft; from 2014 to 2016, the reservoir altitude rose again to a fluctuating level between about 3,040 and 3,055 ft; and from 2016 to 2018, the reservoir maintained an altitude between about 3,050 and 3,060 ft.

Based on water-level measurements in the WCWCD 20 monitoring well network as well as 17 public supply or irrigation wells located near Sand Hollow on the Hurricane Bench, altitude of the water table (or potentiometric surface) near Sand Hollow Reservoir during March 2018 ranged from 2,868 to 3,044 ft (fig. 6). The reservoir altitude during this same period was about 3,050 ft. The lines on figure 6 represent estimated potentiometric contours in the aquifer which indicate lines of equal groundwater-level altitude, and the arrows indicate generalized directions of horizontal groundwater movement away from the reservoir. Horizontal hydraulic gradients are calculated by dividing the difference in water-level altitude between two points by the distance separating these locations and indicate the potential horizontal direction of groundwater movement from high to low groundwater altitudes. The steepest horizontal hydraulic gradients are beneath the North and West Dams and generally decline with increasing distance from the reservoir. The steep hydraulic gradients beneath the dams are the combined result of low-conductivity materials at the core of the dams and pumping from wells and drains at the base of both dams. For example, the horizontal hydraulic gradient between Sand Hollow Reservoir (reservoir altitude of 3,050 ft) and WD 1 (groundwater altitude of 2,976 ft) in 2018 was 0.150 foot per foot (ft/ft), whereas the gradient between WD 6 (groundwater altitude of 2,960 ft) and WD 20 (groundwater altitude of 2,881 ft) was 0.020 ft/ft. In 2018, the resulting regional hydraulic gradient due to managed aquifer recharge between WD 9 (groundwater altitude of 3,051 ft) and WD 17 (groundwater altitude of 2,922 ft) was 0.020 ft/ft.

Pumping rates increased at wells WD 19 and WD 23 starting in early 2013. Increased withdrawal has induced a drawdown cone around these production wells north of the reservoir; the drawdowns are apparent in nearby monitoring wells (WD 4, WD 19, and WD 20). Monitoring well WD 19 went dry in March 2016 as a result of the local water table falling below the depth of the screened interval.

Surface-Water Inflow to and Outflow from Sand Hollow Reservoir

Surface water is pumped into and flows out of Sand Hollow Reservoir through a 60-in. diameter pipeline that enters through an inlet structure at the North Dam (fig. 2). This pipeline is part of a network of pipelines that connect the Virgin River, Sand Hollow Reservoir, Quail Creek Reservoir, and the Quail Creek Water Treatment Plant. The WCWCD has the capability to move water within this network of pipelines by using gravity-induced flow and inline pumping. Sand Hollow Reservoir is managed to maximize groundwater recharge and little surface water has been removed from the reservoir. Annual surface-water inflow to and outflow from Sand Hollow Reservoir from 2002 to 2016 is shown in table 1. The “Monthly pump station inflow or outflow” column for 2017 and 2018 in table 2 is the amount of Sand Hollow Reservoir surface water coming in from the Virgin River or going out to the Quail Creek Reservoir, Quail Creek Water Treatment Plant, or other facilities (fig. 1). These data were collected at the WCWCD pump station located about 1 mile (mi) north of the North Dam. Five pumps, each with Sparling Tigermag totalizing flow meters, are linked to a computer system that combines and records total daily discharge in gallons. The flow meters have electronic modules on which calibration diagnostics are performed monthly by the WCWCD. Each module is removed annually for factory recalibration.

The “Monthly drain and spring return flow to reservoir” column in table 2 is the portion of discharge to the three drains that is pumped back into Sand Hollow Reservoir. The “Monthly outflow to Sand Hollow Resort” column (table 2) is the amount of water required by the resort that cannot be met by discharge to the North Dam drain and is fulfilled by outflow of stored water from Sand Hollow Reservoir. Therefore, the “Monthly total inflow or outflow to/from reservoir” column is a sum of the pump station inflow/outflow, the drain and spring return flow, and the outflow to Sand Hollow Resort (table 2).

The “Monthly pump station inflow or outflow” column is unchanged from Heilweil and Marston (2011) and is comparable to the “Total surface-water inflow or outflow” column in table 7 of Heilweil and others (2005) from March 2002 to August 2004, the “Monthly surface-water inflow or outflow” column in table 2 of Heilweil and Susong (2007) from September 2004 to August 2006, and the “Monthly net surface-water inflow/outflow” column in table 2 of Heilweil and others (2009a) from September 2006 to December 2007.

As in Heilweil and Marston (2011), the “Monthly drain and spring return flow to reservoir” and “Monthly outflow to Sand Hollow Resort” columns are included in calculations of total inflow to and outflow from the reservoir. These amounts are added to the “Monthly pump station inflow or outflow” column and summed in the “Monthly total inflow or outflow to/from reservoir” column. Monthly total inflow/outflow amounts from January 2017 through December 2018 ranged from about −1,370 to 4,530 acre-ft. Approximately 288,000 acre-ft of total net inflow were pumped into Sand Hollow Reservoir from the Virgin River from 2002 through 2018.

Meteorological Data

Meteorological data were initially collected at the WCWCD weather station (fig. 2) in Sand Hollow starting in January 1998. Beginning in 2010, the Soil Climate Analysis Network (SCAN) Sand Hollow weather station (https://www.wcc.nrcs.usda.gov/nwcc/site?sitenum=2159) operated by the Natural Resources Conservation Service (NRCS), replaced the collection of data by the WCWCD weather station. The NRCS Sand Hollow SCAN station is located south of the reservoir within Sand Hollow near the site of the decommissioned WCWCD weather station (fig. 2). Data from both weather stations have been used for evaluating evaporation and precipitation, which are required for calculating monthly recharge from Sand Hollow Reservoir. Parameters measured include air temperature, wind speed, wind direction, precipitation, relative humidity, and incoming solar radiation. Instrumentation includes a temperature and relative humidity probe, a wind direction and speed monitor, a tipping bucket rain gage, and a solar radiometer. Sensors collect data every minute, and average hourly and daily values are computed and stored on a data logger, with the exception of precipitation, which is summed rather than averaged. The solar radiation and temperature data were used for calculating evaporation by using the version of the Jenson-Haise method reported in McGuinness and Bordne (1971).

From January 13, 1998, to December 31, 2018, daily average air temperature ranged from −10 to 36 degrees Celsius (°C). The coldest temperatures during the year were typically in December and January, when minimum air temperatures occasionally were below −10 °C. The warmest temperatures were typically in July, when maximum air temperatures occasionally approached 45 °C. The minimum daily averages are typically in December and January, and the maximum daily averages are typically in June and July. Daily average solar radiation ranged from 14 to 990 calories per square centimeter per day.

Monthly precipitation was recorded at the Sand Hollow weather station or the NRCS station continuously from January 1998 through December 2018, except for two periods when malfunctioning instrumentation resulted in data loss: December 26, 2008, to January 3, 2009, and September 28 to November 16, 2009. Precipitation amounts during these two periods were estimated based on data from the nearby St. George Southgate Golf Course weather station (#427516; http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?ut7516). From January 1998 through December 2018, monthly precipitation ranged from 0 to about 4.3 in. (fig. 7) and averaged about 0.6 in. Average annual precipitation during the 20-year period was 7.7 in.

Changes in Reservoir Storage

Changes in reservoir storage were calculated by using altitude to volume relations from daily reservoir water-level altitudes reported by the WCWCD. After inception of the reservoir in 2002, surface-water storage increased to an average of about 47,000 acre-ft in 2006. Surface-water storage decreased to an average of about 37,000 to 38,000 acre-ft during 2008 through 2010. Following the abnormally wet winter and spring of 2010–11, surface-water storage was kept at a high level during 2011 and 2012, averaging between about 44,000 and 45,000 acre-ft but then declined to an average of about 30,000 acre-ft in 2014 as a result of less than normal precipitation in the region and decreased flows in the Virgin River (table 1). Surface water storage increased again in 2016 to average about 42,000 acre-ft. Monthly surface water storage was variable with levels from January 2017 and December 2018 varying between about 37,000 and 51,000 acre-ft.

Reservoir Evaporation

The McGuinness and Bordne (1971) version of the Jensen-Haise method was selected for calculating evaporation from Sand Hollow Reservoir during this study. A detailed comparison to results using other methods for estimating evaporation is given in Heilweil and others (2005). The McGuinness and Bordne (1971) version of the Jensen-Haise method is based on the relation below:

By using air temperature and solar radiation from the nearby weather stations (fig. 2), monthly evaporation rates were calculated with equation 3. Average annual evaporation rates ranged from 0.41 to 0.54 ft per year from 2002 through 2016. The estimated monthly evaporation rates ranged from 0.07 to 1.15 ft per month from January 2017 through December 2018 (table 2). Multiplying the estimated monthly evaporation rates by average reservoir surface area yields monthly evaporation losses that ranged from about 90 to 1,510 acre-ft between January 2017 and December 2018.

Estimates of Total Volume of Managed Aquifer Recharge from Sand Hollow Reservoir

Monthly estimates of precipitation (P), evaporation (E), inflows (I_sw and I_dr), outflows (O_sw), and changes in surface-water storage (ΔS) were used in equation 1 to calculate recharge to the Navajo aquifer beneath Sand Hollow Reservoir. Annual recharge from 2002 through 2016 ranged from about 5,400 to 18,200 acre-ft (fig. 8; table 1). Monthly recharge from January 2017 through December 2018 ranged from about 40 to 2,000 acre-ft, with 7.3 to 15 percent composite uncertainties (table 2). Higher composite uncertainties in the summer reflect the larger, weighted importance of evaporation losses, which have the highest uncertainty.

Estimated average monthly recharge rates beneath Sand Hollow Reservoir ranged from about −0.0003 to 0.430 ft/d between March 2002 and December 2018 (fig. 9). Although the graph shows large monthly fluctuations, recharge has generally stabilized at an average of about 0.02 ft/d during 2011–18.

Net inflow, evaporation, and groundwater recharge from Sand Hollow Reservoir from 2002 through 2018 are shown in figure 10. Total net inflow during this period was about 288,000 acre-ft, with annual inflow during this period ranging from about 800 acre-ft in 2007 to 56,000 acre-ft in 2005. The general increase in reservoir water-level altitude and area from 2002 to 2007 resulted in a steady increase in the volume of annual evaporation from about 1,100 to 6,600 acre-ft over the same period, and then remained relatively constant ranging between 5,700 acre-ft in 2010 and 8,300 acre-ft in 2017. Total estimated evaporative losses from 2002 through 2018 were about 101,000 acre-ft. Total estimated evaporation losses from 2017 through 2018 was about 16,400 acre-ft. Annual recharge ranged from a low of about 5,400 acre-ft in 2008 to a high of about 18,000 acre-ft in 2005. Total estimated recharge from 2002 through 2018 was about 159,000 acre-ft, with a 2σ uncertainty of 15,100 acre-ft. Of that total, about 18,500 acre-ft of recharge occurred in 2017 and 2018.