Purpose and Objective

Check dams were designed to stop downcutting of the channel and raise the water table to provide a more suitable moisture regime for riparian and desirable upland vegetation. The objective of the water table monitoring study was to evaluate the effect of the project on water table dynamics based on results from field studies (Sagraves 1993). In the absence of pre-project data, results of groundwater and surface water modifications to treated sites were compared to control sites to establish trends in the recovery process due to implementation of the project (1986 to 1993).

Methods

In 1986, 7 stream gages and 24 piezometer wells (16 wells in the study area and 8 wells in the control area) were established after check dam installation to monitor surface and groundwater elevations. The wells and gages were located along seven transect lines crossing perpendicular to the channel both inside and outside the demonstration enclosure (Figure 6-1). Three control areas were established outside the study area for comparison, one upstream and two downstream of the check dam enclosure. A piezometer installation plan was developed for each sample location using soil horizon information obtained from exploratory drilling (Patzkowski 1987). Depth-to-groundwater levels and instream water levels were collected at each well or gage on a bimonthly basis for seven years to monitor the entire groundwater table. Analysis of water table data included use of computer contouring software to generate groundwater contour maps (Figure 6-2). Streamflows were not monitored.

The groundwater aquifer in the project area is unconfined and shallow, averaging approximately 20 feet in depth. The alluvial material in the surrounding valley is relatively well sorted and consists mostly of sand, sandy-silts, and gravel, with isolated bedrock outcrops (Patzkowski 1987).

With the exception of 1986 and 1993, precipitation during the seven monitoring years was well below normal. Average annual precipitation at nearby Portola (which is somewhat higher than Red Clover Creek) was compared to yearly totals from 1987 to 1992. Through 1992, precipitation was almost 30 inches cumulative below normal (Figure 6-3). Therefore, study results represent project influences in the driest of situations. It is suspected that drier-than-normal conditions create groundwater conditions

that exaggerate the recharge of the aquifer by the ponds. Normal or above-normal runoff conditions would maintain higher groundwater levels later into the season.

• Check dams had the immediate effect of raising the water table, creating a high soil moisture regime near the ground surface in the test area. This set the stage for rapid growth of riparian and floodplain vegetation—especially evident at Ponds #3 and #4, the two upstream impoundments (Figure 6-1; Sagraves 1991). Channel downcutting is discussed in Section 7.
• Average depth to groundwater in the test area adjacent to the ponds decreased substantially between 1989 and 1990. This was due to the increase in the elevations of the #4 Check Dam crest following maintenance work. Trends also indicate further reduction in depth to groundwater through 1993 but at a slower rate. The test area overall had relatively shallow depth-to-groundwater values, which increased the availability of soil moisture and encouraged the growth of riparian vegetation. This was demonstrated in results on sampling line #4, where the average depth to groundwater was decreased from 6.9 feet below the surface in 1987 to 1.6 feet in 1993 (Figure 6-4). These stations were within the demonstration exclosure, ungrazed, and relatively close to the channel.

Figure 6-4. Mean annual depth to groundwater for test wells 4A-E, 1987–1993.

• Compared to the test area adjacent to the ponds, control site depth to groundwater remained relatively stable between all years, had large depth-to-water values, and therefore sustained only sparse dry-site vegetation. Results along control sampling line #7 did not reflect the large increases in water table elevation evident in the test area (Figure 6-5). The average depth to groundwater remained relatively unchanged, fluctuating between 5 to 7 feet below the surface on sampling line #7. Control stations farther from the stream showed greater variation between years than stations nearest the stream.
• Results show a noticeable difference in the shallow groundwater flow gradient attributable to the effect of the check dams on the local groundwater regime. Control stations exhibited a groundwater gradient that was generally perpendicular and toward the stream course, causing these areas to drain quickly as runoff subsided. In contrast, water levels in the test area exhibited a flat groundwater gradient, with the water table broadening and flowing outward from the ponds during low flow periods, creating localized areas of elevated groundwater levels (Figure 6-2; Sagraves 1994). Though not quantitatively demonstrated, this effect may be responsible for increased meadow storage and subsequent increases in baseflow, extending the magnitude and duration of flow.

Figure 6-5. Mean annual depth to groundwater for control wells 7A-D, 1989–1993.

• The overall groundwater regime in the test area was not affected by either wet or dry years. Runoff from the surrounding watershed may have been sufficient to completely recharge the shallow aquifer system, even in dry years in both test and control areas. The difference is the rate at which the groundwater storage was reduced after the peak runoff, which occurred very quickly in the control areas.
• Following two years of recharge from check dams, test wells showed small fluctuations in water level each season and from year to year—the range was typically less than one foot. The wells farther away from the ponds showed variations up to several feet, nearly an order of magnitude greater than the near-stream wells (Figure 6-4). The control stations returned to non-runoff conditions quickly after a storm event because the downcut channel drained faster here than in the test area (Figure 6-5). This indicates that the near-field water table is directly influenced by the water level maintained in the ponds (Sagraves 1994).
• Seasonal variation was evident in all years. These changes were influenced by variations in dam crest levels and infiltration from local runoff, as well as fluctuations in streamflow and precipitation.
• In general, pond water levels tended to be greatest in the late summer and early fall due to accumulation of natural drift and beaver activity on the upper dam (Sagraves 1993). Groundwater levels were usually lowest during this period as a result of declining runoff; therefore, artificial recharge of the water table by the ponds occurred at this time (Figure 6-2).
• In areas influenced by the check dams, we observed recharge of springs and near-surface ponds in the floodplain within 200 feet of the channel. This encouraged the growth of vegetation in some areas far outside the exclosure. We attribute this to the ponds and the raised water table in the area, soils and geology, and the size and shape of the shallow aquifers.
• Due to insufficient flow data collected prior to and during the demonstration project, the effects of changes in water table and vegetation recovery on summer baseflows could not be quantitatively identified. This is in part due to the complex pattern of water movement (input and output) through Red Clover Valley and the lack of a permanent weir in the channel above and below the project. Case studies in Oregon and Utah suggest, however, that groundwater recharge and increased meadow water storage moderate runoff by absorbing peak flows and increasing dry season flow (Ponce and Lindquist 1990; see Section 13). Additional research is needed to quantify this effect
Table of Contents.