Klamath Marsh is a wetland complex that lies in the rain shadow of the Cascade Range in the Williamson River sub-basin of the Klamath Basin. The marsh lies directly east of Crater Lake in an area inundated by pyroclastic-flow and -fall deposits from the Holocene eruptions of Mount Mazama. The physical characteristics of rocks of Pleistocene versus Pliocene age combined with NNW -striking fault systems divide the Williamson River basin into two distinct hydrogeologic regimes. The northwestern regime includes the east slope of the Cascades and consists of at least 150 m of interbedded sand, gravel, and stacks (15 to 45 m) of thin (3-5 m) and vesiculated basalt lava flows. Mean annual precipitation ranges from 150 cm near the crest of the Cascades to 50 cm near Klamath Marsh. Moderate to high yield (100 to 4000 gpm) water wells, springs and flowing wells suggest high permeability and ground water potential. The southeastern regime is underlain by Pliocene pyroclastic flows (∼ 40 m) and lava flows (>30 m). Mean annual precipitation ranges from 70 cm in the highlands to 50 cm in the lowlands. Low-yield (20-100 gpm) water wells and perched unconfined aquifers in Holocene pumice deposits suggest low permeability and low ground water potential in areas underlain by the pyroclastic flows. Volumetric analysis of inflows and outflows in Klamath Marsh for 2000 indicates approximately 86% of inflow is from groundwater and 14% from surface water, with nearly 200 x 10⁶ m³ of water removed by evapotranspiration

The town of Mosier, Oregon, is located near the east, dry end of the Columbia River Gorge, and the local area is known for cherry orchards that rely heavily on groundwater for irrigation. The CRBG groundwater system in Mosier has experienced groundwater declines of up to 60 meters due to over-pumping and or commingling. Declining groundwater levels have led to concerns over the sustainability of the resource, as it is the principle water source for irrigation and domestic use. Despite numerous previous studies of groundwater flow in CRBG aquifers here and elsewhere in the Columbia River basin, an aspect that has received relatively little attention is the interaction between groundwater and surface waters at locations where interflow zones are intersected by the surface waters. The objective of my research is to investigate how CRBG interflow zone exposures in Mosier Creek may be controlling groundwater elevations in the area. The methods used include: (1) geochemical analysis of well cuttings and detailed geologic mapping along area streams to identify interflow zones of individual CRBG flows, (2) analysis of stream discharge data and groundwater elevation data to confirm exchange of groundwater and surface waters, and (3) collection and analyses of 31 water samples from area wells, streams, and springs, to determine if waters from individual CRBG aquifers can be hydrochemically identified and to further constrain understanding of surface and groundwater interactions. My study confirms that the general elevation of the Pomona Member and Basalt of Lolo interflow zone creek exposure is coincident with the elevation where a change in slope of the decline trend in 2004 is seen in Mosier area well hydrographs. Furthermore, the results of stream discharge data indicated a close connection between drawdown from groundwater pumping during irrigation season and groundwater- surface water interaction. At the time of drawdown in the upper-most CRBG aquifer (Pomona), the stream transitions from gaining to losing water into the groundwater system. Elemental chemistry data indicates the Frenchman Springs Sentinel Gap aquifer waters are the most evolved waters in this study. Stable isotopic data reinforced this determination as the Sentinel Gap waters are the lightest, or most negative, with regard to δD and δ18O. Sentinel Gap samples were more depleted than other aquifer samples by 4.38 to 6.89 0/100 for δD and 0.39 to 0.59 0/100 for δ18O. The results of the general chemistry and isotope data reveal a more evolved chemical signature in lower watershed groundwater versus a less evolved signature for waters from wells located higher up on the Columbia Hills anticline. This was interpreted to be the result of the major structural features in the area providing for a more regional pathway of recharge in lower watershed groundwaters, versus a more local source of recharge for upper watershed groundwaters. There was also a pronounced commingled signature in the elemental ratios of lower watershed aquifer waters. The suspected mechanism of recharge to lower watershed wells is through younger Cascadian deposits upslope from the local watershed. The findings of this study reveal the importance of a detailed understanding of CRBG stratigraphy and its relation to surface waters, especially for other areas within the Yakima Fold Belt or Oregon and Washington. Studies that do not consider the influence that individual CRBG flows can have on groundwater-surface water interactions, and the groundwater system as a whole, run the risk of improperly assessing the groundwater resource for a region.

The geologic history and sedimentary till of the Tualatin Basin after Columbia River Basalt Group (CRBG) emplacement is assessed and related to groundwater characteristics. The 334 m deep HBD-1 core from the Hillsboro Airport, provides the primary information for sediment characterization and is supported by over 2400 well logs and cores, and four seismic lines. The sedimentary section above the 26 m thick paleosol on the CRBG in HBD-I is divided into two main groups: a 25 m thick section of Missoula flood sediments called the Willamette Silt overlies a 263 m thick finegrained sequence of fluvial Neogene sediments. Pollen, diatom and paleomagnetic data support dividing the Neogene sediments into a 230 m thick Pleistocene package and an underlying, 75 m thick Pliocene to upper Miocene unit. Heavy mineral and INAA chemical analyses indicate that the Neogene sediments were primarily derived from local highlands surrounding the Tualatin Valley. The structure of the top CRBG in the Tualatin Basin exhibits two provinces, a larger northern subbasin with few faults cutting the Neogene sediments above the CRBG and a smaller, more complexly faulted, subbasin south and east of the Beaverton Fault. Neogene sedimentation rates increased ten fold from the late Miocene-Pliocene to the Pleistocene, concomitant with increased basin subsidence. Comparison of Neogene basin evolution among Willamette Valley depositional centers reveals similarities among gravity and seismic reflection characters and subsidence timing between the Tualatin Basin and the northern Willamette Basin and out of phase with the Portland Basin. The Tualatin River CRBG nickpoint near the river's mouth has remained essentially unchanged since the Missoula floods filled the basin 12,700 years ago. This has kept the river from cutting back into the valley resulting in the low gradient evident today. Elevated orthophosphate levels in the upper 140 m of the Neogene sediment section indicate that the sediments are a natural source of phosphorus supplied to groundwater. Groundwater conditions in the lower Neogene sediments promote stabilization of phosphorus as vivianite. The unconfined Willamette Silt aquifer and the underlying confined Neogene aquifers are distinct, separate hydrogeologic units and usually yield less than 40 1pm.

Dutch Canyon is located directly west of the City of Scappoose in Columbia County Oregon. This area is proximate to Highway 30, a major access corridor to downtown Portland, and is experiencing a population increase, which is expected to continue and likely accelerate. As a result, there is growing pressure on water resources. Individual and community efforts to utilize groundwater resources have been hampered by generally poor groundwater yields and water quality concerns outside of the Columbia River corridor and a lack of published hydrogeologic information for the region. The intent of this study is to identify the water-bearing units present in Dutch Canyon and to characterize water quality within these units. The physical hydrogeology of Dutch Canyon was assessed mainly through the collation of 196 local well reports that contained lithologic information from which individual hydrostratigraphic units were identified and characterized. Hydraulic parameters for individual units were estimated using pump rates and drawdowns provided in select well reports. Water quality for the units identified was assessed through the collection of 48 samples of well, spring, and stream water from Dutch Canyon. Measurements of pH, specific conductivity, temperature, dissolved oxygen, reduction potential, and alkalinity were recorded in the field and samples were analyzed for major ions, arsenic, and stable isotopes. The major water-bearing units of Dutch Canyon were separated into five physically distinct hydrostratigrapic units: the lower, middle, and upper units of the sedimentary Lower Miocene Scappoose Formation, and the Wapshilla Ridge and Ortley members of the Lower to Middle Miocene Grande Ronde Basalt. Groundwater flow likely occurs in discrete, relatively thin (~2- to 10-m thick) zones within the Grand Ronde Basalt members. These units only occur along the slopes and ridges of Dutch Canyon west of the Portland Hills Fault, which parallels the eastern margin of the study area. The Scappoose Formation units contain clay- and silt-rich layers and lenses that limit the useable aquifer volume and vertical movement of groundwaters. In general, all hydrostratrigraphic units east of the Portland Hills Fault have low transmissivities and water wells completed in each of them are commonly low- yielding wells, though there are some exceptions. Geochemically, the lower and middle units of the Scappoose Formation were similar to one another with many wells yielding groundwater with high total dissolved solids (TDS) contents (mean TDS = 330 mg L-1; n = 27). Nearly 20% of the wells sampled that were screened in these units (5 of 27) yielded groundwater that exceeded the U.S. Environmental Protection Agency's National Secondary (non-enforceable) Drinking Water Regulation standard of 500 mg L-1 TDS. The upper unit of the Scappoose Formation and the overlying Grande Ronde Basalt members generally yield water with lower TDS contents (mean < 200 mg L-1; maximum = 342 mg L-1; n = 20). Groundwater resources in Dutch Canyon are limited and low well yields are common. The primary water quality concern is saline water, which is generally found in the lower and middle units of the Scappoose Formation near the valley floor. Low recharge rates determined from hydrograph analysis of stream discharge measurements are consistent with the geology and steep terrain of the area and further limit the available groundwater and the degree of flushing of what may be connate waters in the deeper units.

This thesis evaluates the hydraulic relationship between the Blue Lake gravel aquifer, the Columbia River, and Blue Lake. Hydrogeology, water levels, and stable isotopes were used to establish these hydraulic relationships.

Determination of the vertical and horizontal groundwater hydraulic gradient within a landfill is the first step in determining the potential of groundwater contamination from the landfill leachate. The length of a study and the frequency at which measurements are recorded can greatly affect the description of the local groundwater environment. A more comprehensive analysis can be preformed for longer periods of study and greater measurement frequency. The intent of this study was to install a continuous groundwater level monitoring system around the st. Johns Landfill for a minimum study length of one year. This would allow a more thorough study of the seasonal character and behavior of the groundwater system beneath the landfill than in previous studies. Particular interest was paid to groundwater level changes resulting from seasonal weather changes. Additional attention was paid to other forcing mechanisms which could be perturbing groundwater levels, and variations in the geochemical groundwater constituents. Included throughout this report is a literature review of various studies pertinent to the analysis of groundwater level variations. Seasonal variations in vertical groundwater hydraulic gradients were reviewed and time averaged vertical seepage rates were estimated. Areal plots of groundwater levels were used to view expected horizontal groundwater hydraulic gradients during seasonal maximum and minimum groundwater levels. A computer model was developed to study the effects temporal variations in slough water levels had on groundwater seepage rates through the perimeter dike separating the landfill from the sloughs. The modeling provided an estimate of the average horizontal leachate seepage rate into the sloughs. Comparison plots of monitoring well groundwater levels were used to analyze potential swash zones beneath the landfill and potential effects of lowered water levels in Bybee Lake. Spectral analysis techniques were imployed to determine the dominant frequencies observed in the groundwater levels, allowing determination of the type of forcing mechanism driving the fluctuations. Geochemical groundwater constituents were statistically analyzed to determine the significance of observed trends in the data: areal plots of chloride concentrations and electrical conductivity were made to view constituent distributions within the underlying aquifers. Estimated vertical and horizontal groundwater seepage rates into the local waters showed that horizontal leachate seepage is insignificant compared to vertical leachate seepage. Groundwater level comparison plots indicated no significant swashing beneath the landfill occurred. The statistical studies on groundwater forcing mechanisms indicated that either the slough or the Columbia River water levels could be perturbing groundwater levels. Trend analyses on the geochemical groundwater constituents indicated significant, positive trends in chloride concentrations, and undeterminable trends in electrical conductivity.

The Dallas-Monmouth area, located in the west-central Willamette Valley, Oregon, consists of Tertiary marine and volcanic bedrock units which are locally overlain by alluvium. The occurrence of groundwater with high salinities has forced many rural residents to use public water supplies. Lithologic descriptions from driller's logs, geochemical (INAA), and x-ray diffraction analyses were used to determine alluvial facies distribution, geochemical and clay mineral distinctions among the units, and possible sediment sources. Driller's log, chemical and isotopic analysis, and specific conductance information from wells and springs were used to study the hydrogeologic characteristics of the aquifers and determine the distribution, characteristics, controlling factors, and origin of the problem groundwaters. Three lithologic units are recognized within the alluvium on the basis of grain-size: 1) a lower fine-grained unit; 2) a coarse-grained unit; and 3) an upper fine-grained unit. As indicated by geochemical data, probable sediment sources include: 1) Cascade Range for the recent river alluvium; 2) Columbia Basin plutonic or metamorphic rocks for the upper fine-grained older alluvium; and 3) Siletz River Volcanics from the west for the coarse-grained sediment of the older alluvium. The Spencer Formation (Ts) is geochemically distinct from the Yamhill Formation (Ty) and the undifferentiated Eocene-Oligocene sedimentary rock (Toe) with higher Th, Rb, K, and La and lower Fe, Sc, and Co concentrations. The clay mineralogy of the Ty is predominantly smectite (86%) while the Ts contains a more varied clay suite (kaolinite, 39%; smectite, 53%; and illite 8%). The Ty and Toe are geochemically similar, but are separated stratigraphically by the Ts. The Siletz River Volcanics is distinct from the marine sedimentary units with higher Fe, Na, Co, Cr and Sc concentrations. The Ty and Toe are geochemically similar to volcanic-arc derived sediments while the Ts is similar to more chemically-evolved continental crust material. Wells that encounter groundwater with high salinities (TDS>300 mg/1): 1) obtain water from the marine sedimentary bedrock units or the older alluvium; 2) are completed within zones of relatively low permeability (specific capacities ~5 gpm/ft); and 3) are located in relatively low-lying topographic settings. The poor quality waters occurring under these conditions may be due to the occurrence of mineralized, regional flow system waters. Aquifers of low permeability are less likely to be flushed with recent meteoric water, whereas upland areas and areas with little low permeability overburden are likely zones of active recharge and flushing with fresh, meteoric water. The most saline waters sampled have average isotopic values (6D = -6.7 ° / 00 and 60 = -1.7 ° / 00 ) very near to SMOW, while the other waters sampled have isotopic signatures indicative of a local meteoric origin. The Br/Cl ratios of most (10 of 14) of the waters sampled are within 20% of seawater. A marine connate origin is proposed for these waters with varying amounts of dilution with meteoric waters and water-rock interaction. The problem waters can be classified into three chemically distinct groups: 1) CaC12 waters, with Ca as the dominant cation; 2) NaCl waters with Na as the dominant cation; and 3) Na-Ca-Cl waters with nearly equal Na and Ca concentrations. The NaCl and CaC12 waters may have similar marine connate origins, but have undergone different evolutionary histories. The Na-Ca-Cl waters may represent a mixing of the NaCl and CaC12 waters.

The middle Holocene cataclysmic eruption of Mount Mazama blanketed Walker Rim, in south central Oregon, with 270 cm to 300 cm of pumice, causing capture of surface water systems by groundwater, stream relocation, and the formation of biologically diverse fens and seasonal wetlands. The pumice aquifer at Round Meadow, an 8.6 km2 basin, hosts both a fen and seasonally ponded wetlands. The Round Meadow watershed lies within a closed basin between the upper Klamath and Deschutes river basins. As the highest meadow at Walker Rim, it is a relatively well-constrained system to study the effects of hydrological disruption. A water budget was calculated for the basin, hydraulic conductivity was evaluated for the three main sediment layers in the meadow, recharge sources and evaporative trends were studied using stable isotope analysis, and aquifer residence times were estimated using CFC tracer water age dating. Water year 2014 was a drought year and observation of the system under stressed conditions allowed discrimination of four independently functioning components of the hydrogeologic system. These were the meadow, which is by far the largest component in terms of water storage, the fen where iron cementation and up to 1 m of peat holds water in a berm above the meadow, three springs which are sourced from deeper groundwater hosted in the bedrock which underlies the pumice deposit, and the outflow area. In all cases, the aquifer material is pumice, but the influence of the pre-eruption landscape and post-eruption modifications of the aquifer material have resulted in partial isolation of the components. The water budget analysis indicated that the basin lost 44 cm of water storage during WY 2014. Hydraulic conductivity values of 1x10-6, 2x10-2, and 4x10-5 cm/s, were determined for the diatomaceous silt underlying the pumice, the Plinian pumice fall aquifer, and for the diatomaceous silt overlying the pumice, respectively. The pumice is characterized as a perched, weakly confined aquifer and residence times in the pumice are much longer (decades) than for water near the surface of the meadow. Water discharging at the springs is isotopically different (lighter) than either the surface water or groundwater in the pumice aquifer. The fen at Round Meadow appears dependent on seasonal precipitation to recharge water, and responds to fluctuations in annual precipitation. The wetland meadows are volumetrically the main water-storing features at Round Meadow, and these are not homogenous features, but a combination of discrete components.