The hydrologic model describes how water moves through the geologic materials within the study site. The primary elements of the model are recharge, ground water flow and perched aquifer discharge with supporting data from hydrographs, water chemistry and geophysics. Recharge to the model occurs during mid-April through mid-October from irrigation water. Natural precipitation effects surface water flows primarily during December, January and February but is likely not significant source of recharge to the perched aquifers. Ground water and perched aquifer discharge hydrographs exhibit cycles that closely correlate to seasonal irrigation. Water chemistry illustrates how chemical concentrations increase as ground water flows from recharge areas to discharge zones and high electrical responses from geophysics indicate locations of recharge areas.

Irrigation As A Recharge Source

Landuse change on the plateau from sagebrush desert to irrigated farmland coincided with the formation of perched ground water systems, which discharge along the hillsides of the Bruneau Plateau. The Bell Rapids Mutual Irrigation District is located adjacent to the Hagerman Fossil Beds National Monument and operates and maintains an irrigation system that pumps water from the Snake River to irrigate approximately 19,000 acres of land on the plateau. The water license priority for Bell Rapids Irrigation Company is dated December 16, 1963 and construction of the irrigation facilities began in 1969 with the first irrigation season in 1970 (Vector, 1994).

The following paragraphs provide a chronology for irrigation activities. The main pumps lift water from the Snake River up 600 feet in elevation where it is emptied into the Fossil Gulch Canal. Water is pumped from the canal system for field application using sprinkler system technology. The Irrigation season starts in mid April and ends in mid October. Total amount of water pumped into the irrigation system averages 45,000 acre-feet per year and Figure 8 illustrates yearly volumes of water pumped from the Snake River into the irrigation system (U.S.G.S.).

The first irrigation season and recharge to the perched aquifers began in spring 1970. Perched ground water began to discharge on the hillsides during mid 1970's and the first large slope failure occurred in ca. 1979. Slope stability problems are a function of several elements that include ground water, steep slopes up to 70 degrees, poorly consolidated sediments and at some locations over-steeping of slopes from road building. In 1983 a second slope failure nearly destroyed a portion of the Bell Rapids Canal. In spring 1987, a third slope failure occurred that destroyed the Bell Rapids Pumping Station while workers were turning on pumps. During the 1987 irrigation season the Fossil Gulch Canal carried larger volumes of water in response to the non-usable Bell Rapids Canal so lateral line aqueducts could transfer water to the areas previously supplied by the Bell Rapids Canal. After the 1987 season the first one-half mile of Fossil Gulch Canal was realigned to straighten it and lined with concrete.

The 1983 slope failure set in motion a series of actions and canal leakage studies addressing the perched ground water conditions within the plateau. Studies have been conducted by both private and government agencies since that time. Results from these studies indicate the source of recharge to the perched aquifers is the irrigation system.

Figure 8. Total volume of water (acre-feet/YR) pumped into Fossil Gulch Canal.

Summary of Canal Leakage Studies

Recharge to the perched aquifers from the canal and pond has been studied during five different leakage investigations. A discussion of each hydrologic study is provided in Appendix B and Figure 9 illustrates the locations of each study. The first study by Young (1984) measured flow within the Fossil Gulch Canal at two locations and then indirectly calculated the amount of water leaking from the canal based on the difference between the two flow rates. Young (1984) reports about 1,900 acre-feet of water leaks from the first mile of the Fossil Gulch Canal per 120 day irrigation season, before it was lined with concrete in 1988. The second study by Worstell (1985) measured the first half-mile of canal with a seepage meter that is commonly used for such studies. His results indicate a lower leakage rate of 193 acre-feet per 180-day irrigation season.

A third canal leakage study, conducted by Montgomery (1987), involved constructing a water impoundment in the first one-half mile of Fossil Gulch Canal. The calculated leakage was 299 acre-feet per season. The fourth canal leakage study, performed by Martin (1994), tested three sections of the Fossil Gulch Irrigation Canal and two irrigation holding ponds. The results from his study show 360 acre-feet of water per irrigation season is lost from the first one-half mile segment of unlined canal. Vector Engineering, Inc. (1994) performed the fifth leakage study. This study was designed to measure the canal flow at 500-foot intervals for the entire length and use this flow rate data to indirectly calculate canal leakage similar to Young's study in 1984. The results of the Vector study were stated as inconclusive, but a leakage rate of 5,000 acre-feet/season for the entire canal system was reported. Canal leakage estimates for an average irrigation season are summarized in Table 2. Assuming a constant leakage rate along the canal may over estimate leakage because of changing geologic conditions. For instance, the Tuana Gravel Formation is below the canal at the inlet area near the edge of the bluffs where most of the leakage tests were performed. The canal likely crosses the contact between the Tuana Gravel and the finer grained Glenns Ferry Formation within the first mile segment of canal. The changing geologic condition of the canal base, described in the geologic model, suggests that extrapolating the leakage studies tested at one location to the entire canal length is probably not correct.

Figure 9 Locations of canal leakage studies (omitted from on-line edition)

Table 2. Volume of water lost to leakage from the Fossil Gulch Canal system determined from different studies.

The Fossil Gulch Irrigation Pond leakage study performed by Martin (1994) indicated a water leakage of 140 acre-feet per season. If this value is correct it would be roughly equal to 33% of the annual discharge from the perched aquifer as estimated by Young (1984). A degree of uncertainty is associated with the reported pond leakage data due to surveying problems during data collection.

Monitor Wells

Ground water monitoring in the Fossil Gulch area started in 1986 with the installation of six monitor wells. In 1994 six additional wells were constructed for a total of 12. Table 3 provides construction information and Figure 10 illustrates the location of the monitor wells. All of the wells drilled in 1986 were constructed with six inch diameter PVC casing with 20 feet of 0.030-inch slot sized screen.

The sand packs extend up to within 30 feet of ground surface then a cement grout surface seal was placed (Young, 1997). The 1986 wells that encountered basalt only penetrated a short distance into the upper surface of the basalt while the 1994 wells drilled completely through the basalt into the underlying Glenns Ferry Formation.

Table 3. Monitor well information.

The wells constructed in 1994 are more complex and variable than the 1986 wells. The 1994 wells were constructed with six-inch diameter PVC casing with variable lengths of 0.040-inch slot sized screen. Gravel pack lengths are also more variable ranging from 35 feet in NPS-3 to 190 feet in NPS-4. The rest of the 1994 wells have gravel pack intervals of about 90 feet. All of the 1994 wells penetrated below the Shoestring Basalt to varying depths ranging from 72 feet in NPS-4 to 38 feet in NPS-5 with the rest averaging about 14 feet below the basalt. Bentonite grout was used to seal from the sand pack up to about 10 to 20 feet from ground surface and then a cement grout seal placed up to ground surface.

Figure 10 location of wells (omitted from on-line edition)

Well completion diagrams and drill logs for the wells are located in Appendix C. Long effective screen intervals have likely interconnected different perched aquifers producing a composite affect for some of the monitor well hydrographs. The U.S.G.S. collected water level measurements in wells from March 1986 through May 1987 and January 1992 through April 1996. National Park Service staff has collected water level data since April 1996. Hydrographs for each of the wells are located in Appendix C.

Ground Water Hydrographs and Contours

Ground water levels are responding to temporal changes in recharge due to seasonal irrigation practices. Monitor well hydrographs for NPS-5, NPS-4 and NPS-3 (Figure 11) have been chosen to demonstrate how changes in recharge conditions effect water levels in the middle perched aquifer. These wells are completed in the Shoestring Basalt flow (geologic model layer 3). Other monitor wells in the Fossil Gulch area were not used in the demonstration because some wells (such as NPS-6) have nearly identical hydrographs of other wells (NPS-3) while others have complex hydrographs attributed partially to well construction and design. Appendix C provides a discussion of these wells.

Well NPS-5 is located nearest to the source of recharge and has the earliest rise in water levels, highest water level elevation and greatest hydrograph amplitude of wells completed in the basalt flow. Figure 11 illustrates water levels starting to rise during the early part of June for NPS-5. Well NPS-4 is located further from the recharge area than NPS-5 and water levels start to rise during July as the effects of the recharge pressure wave propagate down gradient. Well NPS-3 is located furthest down gradient near the perched aquifer discharge zones. Water levels in this well begin to rise in August. Figure 12 illustrates the interpreted progression of seasonal recharge events moving through the basalt system noted by hachured lines located at NPS-5, NPS-4 and NPS-3. The water level response of these wells indicate a source area near the Fossil Gulch Irrigation Pond. The pressure response takes about 100 days to travel a horizontal distance of 15,000 to 20,000 feet from the recharge area to the discharge zones; this equates to a range of 150 to 200 feet/day.

Figure 11. The hydrologic model illustrates a cyclic relation between the irrigation system (bar graphs) and ground water hydrographs for NPS-5, NPS-4 and NPS-3.

Figure 12 Contour lines (omitted from on-line edition)

Ground water contours are illustrated in Figure 12, which are based on hydrographs for monitor wells NPS-5, NPS-4, NPS-3 and NPS-6. These wells were selected because they are completed in the middle basalt aquifer. Water level data from October 15, 1997 was used for the contours; this is when the hydrograph for NPS-5 is at its greatest height each year. There is a 36-foot loss in head from NPS-5 to NPS-3 over a distance of 9,000 feet. This equates to a hydraulic gradient of 0.004 (0.4 percent) compared to a gradient of 0.006 (0.6 percent) for the basalt flow. The downward hydraulic gradient from NPS-5 to NPS-3 indicates water flowing from the northwest to the southeast in the basalt aquifer.

Ground Water Chemistry

Water chemistry data for the Fossil Gulch Canal, monitor wells NPS-5 and NPS-4, and aquifer discharge are consistent with the model. As water enters the ground water systems it dissolves ions. An increase in chemical concentrations occurs as ground water flows further from the recharge area. The U.S.G.S. collected water quality data in 1993 and the trilinear diagram in Figure 13 illustrates water chemistry for the Fossil Gulch Canal, NPS-5, NPS-4 and a discharge stream. Total dissolved solids progressively increase from the lowest concentration in canal water, to NPS-5, to NPS-4 and ultimately the perched aquifer discharge, which has the greatest concentrations. The increase in chemical concentrations as ground water flows further from a recharge area to a discharge zone is consistent with the hydrologic model.

Figure 13. Trilinear diagram illustrating an increase of dissolved ions noted by the increase in circle diameters as water flows from recharge area to discharge area. (omitted from on-line edition)

Geophysics

Two surface geophysical studies were conducted in 1994 by Vector Engineering to investigate relations between the irrigation canal system and the perched aquifers. Misa-a-la-masse and Schlumberger studies (Figures 14 and 15) were employed to determine if there is a correlation between canal water and the perched aquifers. Appendix A provides a detailed discussion for each study. The locations of each Misa-a-la-masse array station are marked with a triangle on Figure 14. Voltage potential was measured lateral to the canal while the canal water was charged with an electrical current. High electrical potential responses are inferred to be indicative of ground water recharge from the canal and low potential charges equate to little or no canal water connection with ground water. Pelton (1995) states that Vector (1994) used an incorrect method of moving the current electrode during the Misa-a-la-masse data collection and because of this the data are suspect. The high electrical response located near the Fossil Gulch Pond (Figure 14) is interpreted to reflect recharge in this area that is consistent with the hydrologic and geologic model from this study.

Figure 14 Misa-la-a-masse Study (omitted from on-line edition)

Figure 15 Schlumberger Study (omitted from on-line edition)

Stations for the Schlumberger array (Figure 15) were recorded with the same grid established during the Misa-a-la-masse study and data were collected using an expanding electrode array to map subsurface resistivities. The crosshatch pattern in Figure 15 is interpreted as high water content. There is a general northwest/southeast trend to the patterns which head towards the Fossil Gulch Pond area. Pelton (1995) states that Vector (1994) used a unique computer program to process data collected from the survey and this should be taken into account while interpreting the results. This study reviewed the Schlumberger data and generally concurs with Pelton's (1995) statements. The computer program did not take into account the geology of the plateau or discriminate raw data between different perched aquifers. The program incorrectly connected high responses at different elevations into one layer and interpreted this layer as one aquifer. Other cross sections appear to be reasonable and so the results of the Schlumberger study are debatable.

Perched Aquifer Discharge

Perched aquifer discharge rates have been measured by the U.S.G.S. and N.P.S. using weirs and flumes placed in surface water flows below aquifer discharge zones. Figure 16 illustrates weir and flume locations and Figure 17 illustrates the hydrographs for surface flows and the Fossil Gulch Canal. The purpose of collecting discharge measurements is to aid water balance calculations, record surface flow trends, correlate with recharge events and monitor well data.

The U.S.G.S. used a weir and staff gauge to collect measurements for spring 16ABB2s and a flume for 16ABA1s. Measurements were collected roughly every month from January 1992 through August 1996. The N.P.S. staff started collecting surface water measurements in the same streams as the U.S.G.S. in October of 1996 using volumetric measurement methods instead of weir and staff gauge. The N.P.S. placed metal square notch type weirs into the streams and then used a calibrated bucket and stopwatch to collect measurements.

The majority of aquifer discharge comes from the basalt aquifer. Young (1984) calculated a discharge rate of 420 acre-feet (AC-FT) of water per year. Included in this value are aquifer discharge locations noted in Figure 16 but also two discharge locations 2 miles south of the study area in sections 29 and 32. Aquifer discharge in sections 29 and 32 are not part of the same perched aquifer system located within the study site. If these two locations are subtracted from the 420 AC-FT calculated by Young (1984) then discharge from the study site area is 360 AC-FT per year. Low surface water flow rates occur in April, May and June and the high flow rates occur during November, December and January. Most of the aquifer discharge hydrographs in Figure 17 show a cyclic pattern associated with the seasonal irrigation. The irrigation system starts filling with water in mid-April and four months later in August, discharge flows 16ABB2s and 16ABA1s respond with increased surface water flow. The drop in the Hydrograph for 16ABA1s in late 1995 is due to a channel shift from debris flows shedding off the 1991 landslide. Some flow rates change because mud/rock debris flows originating from the landslides alter the stream channels and flow regime.

Figure 16 weir location map (omitted from on-line edition)

Figure 17. Hydrographs of perched aquifer discharge streams and Fossil Gulch Canal illustrating a cyclic relation between recharge events and aquifer discharge.

Conclusions

The hydrologic model illustrates a hydraulic correlation between the seasonal irrigation, ground water flow within the Shoestring Basalt flow and perched aquifer discharge along the hillsides. Hydrographs support a ground water pressure response occurring from seasonal irrigation, which starts in mid-April. It reaches monitor well NPS-5 in June, NPS-4 in July, NPS-3 August and causes increased surface water flow in August. The lag time between the start of the recharge event in April and a response in perched aquifer discharge is about four months. Ground water contours indicate a hydraulic gradient from NPS-5 to NPS-3, which supports water flowing from the northwest to the southeast. The Misa-a-la-masse, Schlumberger and water chemistry data also indicate recharge occurring near the pond area and flowing down gradient to the discharge zones.

Table of Contents
Chapter 1 | 2 | 3 | 4 | 5
Appendix A | B | C | D

Questions or Comments?