Geology and Hydrogeology of Lake Merced, San Francisco, California

Yu Shimuzu, Graduate Student, Geography
San Francisco State University

Introduction

Lake Merced is located in northeast of the San Andreas Fault. In the past, seismic activities adjacent to the fault system that stretches along the coastline of California for approximately 1350 kilometers (Harden 1997) have caused a magnitude of damages to nearby cities, including residents in the Bay Area. It is this strike-slip fault system that has contributed to the intriguing geologic history of the area remains active today.

The lake has formed in a geologically interesting area. In addition to the formation of the fault system, evidences of folding and sliding of crusts at a subduction zone are recorded in the underlying bedrock of Lake Merced. Sedimentary deposits have left clues for global climate and sea level changes. Subsequently, a number of physical factors interacting in a complex manner have made the understanding of geology and hydrogeology of the lake difficult.

This section of the geology and hydrogeology of Lake Merced includes details of the formation of Lake Merced, which is a relatively recent event compared to the formation of the Franciscan complex, the Merced Formation, and the Colma Formation. Interactions among these formations are discussed to characterize the physical setting of the lake.

Geologic history

Different types of rocks and sediments that have accumulated over time are represented in the Lake Merced area, although they are not always easily distinguished from one another (Hunter et al. 1984). Materials buried under the ground can be generally classified into two types: bedrock and sedimentary deposits. In the Lake Merced area, consolidated rocks of the Franciscan Complex, which are tightly packed, constitute the bedrock. The Merced and Colma formations are sedimentary deposits found above the Franciscan Complex. Relative vertical position of these rocks indicates their age. Since younger materials are deposited over preexisting surface, the older rocks are found on the bottom. Consequently, the Franciscan rocks are older than the Merced and Colma formations.

However, in the Lake Merced area, the age and origin of rocks are seldom clear because the distribution of rock groups, which represents rock associations according to locational and temporal similarities of their origin, are not always uniform. This phenomenon is a result of vertical or horizontal displacement of blocks of rocks along the fault system (Wallace 1991). The rocks surrounding the San Andreas fault are thus not preserved in an orderly fashion. Therefore, interpretations of the related geologic parameters, especially the determination of age of geologic features, are extremely difficult.

The Franciscan Complex

The Franciscan Complex underlies the surface deposits associated with the Colma and Merced formations. These rocks associated with this complex occur in northeast part of the lake. Another type of bedrock found on the San Francisco Peninsula is the Great Valley Sequence, however this formation is not found directly beneath Lake Merced and therefore will not be included in the scope of this report (Konigsmark 1998).

The Franciscan rocks found below Lake Merced are part of the Central Belt. The Central Belt is one of the Coast Ranges, which as a whole is a tectonic mélange, that uplifted during the Tertiary period along an ancient subduction zone and contains the rocks of Late Jurassic to Early Cretaceous age (Sullivan and Galehouse 1991). However, the oldest Franciscan rocks have been reported to date back to 175 million years ago (mya), around the time when the lower part of the Fallon plate collided with North American plate at a subduction zone (Konigsmark 1998).

Lake Merced rests upon the San Bruno terrain which consist of greywacke sandstone and K-feldspar. Greywacke, the most common rock in the Franciscan Complex, is a variety of sandstone that has a relatively small amount of quartz (SiO₂), and relatively high proportion of feldspar.  Greywacke also contains rock fragments such as grains of chert, volcanic rocks, metamorphic rocks, shale, and fine-grained matrix (Konigsmark 1998) (see Table 1 for a summary of rock types and their characteristics). Two types of feldspars are differentially abundant and reflect the time of formation. The plagioclase feldspar rich in sodium and calcium is older than K-feldspar rich in potassium (Konigsmark 1998). Richness of K-feldspar in the San Bruno terrain eludes to the more recent development of this terrain. The Franciscan Complex also contains other rocks that are less abundant than either greywacke or feldspar. Basalts and other volcanic rocks make up about 10% of the Franciscan Complex, with chert being even less common than basalts and volcanic rocks. Serpentine is rare but is more widely spread in the Franciscan Complex than others (Konigsmark 1998).

Table 1. Common Franciscan Rocks

Igneous Rocks: rocks crystallized from magma

Sedimentary Rocks: rocks deposited in water, usually as grains of other rocks

Metamorphic Rocks: sedimentary and igneous rocks altered by heat and/or pressure.

*Modified from Rock Types, San Francisco Area (Konigsmark 1998).

The Merced Formation

The Merced Formation, extending from the north of San Bruno to the coast of Marin (Figure 1), is deposited over the Franciscan Complex and overlain by the Colma Formation (Hunter et al. 1984). The formation is of Pliocene and Pleistocene age (Hunter et al. 1984; Sullivan and Galehouse 1991; Yates et al. 1990) making it at least 1.6 to 1.2 mya (Clifton et al. 1988). A total depth of 5,800 ft of sediments were deposited over the bedrock (Geo/Resource Consultants 1993), while Hunter et al. (1984) estimated the thickness of the formation to be 2,000 meters in the Lake Merced area. However, they noted that deposits of the uppermost layer of the Merced Formation and the bottom of the Colma Formation are similar in texture and structure, making the two formations difficult distinguished, unless unconformity was taken into account.

Figure 1. Distribution of the Merced and Colma Formation (Source: Clifton et al. 1987)

Sources of the sediments in the formation vary. Hunter et al. (1984) described the inferred origin of ten facies. The lower sediments are believed to be derived from the Franciscan assemblage and other local sources (Clifton et al. 1987). The sediments of the upper formation were generated at the drainage basin of the Sacramento and San Joaquin rivers (Clifton et al. 1987; Konigsmark 1998). Although these sediments were originally deposited horizontal to the surface they have been deformed due to the presence of the active San Andreas fault. The parts of the deposits that were carried northward after being forced apart by the fault system are found in Marin County, California (Konigsmark 1998). Clifton et al. (1988) found a layer of ash deposition that had been carried from Lassen Peak after it erupted about 400,000 years ago apparently when the Merced Formation was above sea level (Figure 2). Marine sediments are also found. Fossilized bivalves, mollusks, such as Scutellaster spp., and other invertebrates can be considered as evidence for the presence of marine conditions during the sedimentation. These fossils also provide a reference point in the geologic time scale from which paleontologists and geologists can make comparisons to determine the age of other strata. Thin layers of mud and peat are interbedded in the marine deposits (Yates et al. 1990).

Figure 2. Stratiography of the Merced Formation (Source: Clifton and Hunter 1987)



A type of sedimentary deposit in the Merced Formation and a change in the sea level, due to glaciations or a rapid subsidence of basin, are correlated. When the sea level was lower during glacial periods, the shoreline extended seaward from its current position, creating a situation in which non-marine sediments accumulated at the mouth of basins. Subsequently, when the water held in glaciers melted during interglacial periods, the sea level rose and the basins began to accumulate marine deposits (Clifton et al. 1988; Konigsmark 1998).

Similarly, the movement of the shoreline depended on how fast the basin was subsiding. When the basin rapidly subsided the shoreline transgresses landward, whereas the shoreline regressed when subsidence proceeded slowly (Figure 3).

Figure 3. Variation of Sedimentary Deposits (Source: Condie and Sloan 1998).

The Colma Formation

The Merced and Colma Formations are Cenozoic in age. The Colma Formation is younger in its age than the Merced Formation but no younger than the last glacial episode (Wisconsin glaciation) and is dated 0.07-0.13 mya (Clifton et al. 1988; Konigsmark 1998). Hengesh and Wakabayashi (1994) observed an angular unconformity between the Merced Formation and the Colma Formation, meaning that the overlying Colma Formation rested on the top of the Merced Formation at an angle. This has probably resulted from the uplift of the Merced Formation (Hengesh and Wakabayashi 1994).

Sediments of the Colma Formation are not dissimilar to the Merced Formation. The sediments were deposited in either marine or non-marine environment (Clifton et al. 1988; Hengesh and Wakabayashi 1994). Yate et al. (1990) described the texture as "poorly unconsolidated sands" and muds with a thickness of less than 300 ft.

Dune sands overlay the Colma Formation. Some of these dune sands were carried by the Sacramento River system through the Golden Gate and were deposited in eolian environment (Konigsmark 1998). The sands, characterized by excessive drainage of water, extended throughout most of western San Francisco before its development (Sullivan and Galehouse 1991), and supported the native grassland and scrub vegetation that once were widely distributed throughout the San Francisco peninsula.

The Formation of Lake Merced

The formation of the lake began during Pleistocene (Fahy 1974), no earlier than 15,000 years ago (Geo/Reference Consultants 1993). The area surrounding the Lake was formerly a bay or estuary during the last interglacial period prior to the Wisconsin glaciation. The shoreline extended farther westward during the last glacial episode when sea level was lowered. Two natural streams, both flowing westward from the north and south of the current location of the lake, eroded the unconsolidated deposits and creating valleys. As the velocity of the northerly offshore current and sand loading capacity of the current dropped, the sands were deposited on the south of the bay. When more sands were added to the sandbar, it closed the mouth of the bay and formed a lake that was separated from the Pacific Ocean. At first the sandbar was only seasonally present in an earlier stages of the lake formation. But it appears that the channel to the ocean had been closed for a period of time before an earthquake shook the area in 1852 (Fahy 1974). Since that time the lake has remained a freshwater system and with the original lake being artificially separated into three lakes: North, South, and Impound lakes.

San Bruno Fault

A hypothetical fault was expected to be present through the bottom of the lake at some depth (Camp Dresser & McKee Inc.  1999).  Bonilla and his fellow USGS scientists found no fault (Jorgenson 2000).

Soils of Lake Merced

The majority of the ground surfaces in the Lake Merced area have been altered. Surfaces are often paved as roads or parking lots, making the natural characteristics of soils non-apparent in many areas. Regardless, the two major types of soils identified in areas surrounding Lake Merced are Orthents and Sirdrak (USGS 1991). Both of these soil types contain sand, loamy sand, and/or sand loam, with little or no development of pedogenic horizons, meaning that they are susceptible to erosional forces and landslides. Orthents are entisols found on recent erosional surfaces, or can simply be thought as a grab bag for entisols lacking specific characteristics required to be classified into a particular suborder. Inceptisols have altered horizons that have lost bases or iron and aluminum and lack an alluvial horizon enriched with silicate clays (Pond 1989).

Hydrogeology of Lake Merced

The geology of the Lake Merced area governs the characteristics of aquifers directly beneath the lake itself and its surroundings. When a continuous horizontal layer of relatively impermeable materials, such as clays, underlies aquifer materials that are more permeable due to their larger pore spaces (i.e. gravel), groundwater is retained with the impermeable layer retarding the vertical flow of the water (Yates et al. 1990). The impermeable clay layers also act as effective barriers separating the underlying aquifer from the water table (Rogge 2000). These clay layers are present in both the Merced and Colma formations.

However, extensive and thick clay layers are only found in the Merced Formation and are either absent from the dune sands or rare in the Colma Formation (Yates et al. 1990). An aquifer found at a greater depth, at about 100 ft deep, across the northern part of the Lake Merced area has a thicker layer of clays. This layer isolates the shallower water table aquifer that predominantly consists of Holocene dune sands and Colma Formation from the underlying confined part of the aquifer system in the Merced Formation. Therefore, the horizontal continuity of the clay layers controls the extent of aquifer confinement. Discontinuous clay layers on the other hand are attributed to unconfined aquifer systems and the vertical flow of water is less restricted relative to the confined system. Unfortunately, the extent and distribution of the clay layers are not completely understood, and a question of whether the wells are drawing water from confined or unconfined sources remains unanswered (Rogge 2000).

Due largely to a recent rapid decline in the groundwater level in the Lake Merced area, public awareness toward current water use regimes and the distribution of the resource has increased. However, available data lack details on how pumping water out of the wells affect the water level. Similarly, the existence of a direct relationship between human utilization of the water and the decline in the groundwater level is likely, but it is yet to be confirmed. Concerned residents of the area await farther analyses of the parameters controlling the groundwater level. One setback to the researchers’ efforts to fully understand the aquifer system is the cost of conducting such detailed studies (Rogge 2000). For a more in depth look at the hydrology, water quality, and the utilization of the Lake Merced water resources please refer to the section on "Lake Merced Hydrology and Water Quality".

Size and Depth of Lake Merced

The North, South, and Impound lakes collectively comprise Lake Merced. A total volume of water in the lake was approximately 1,006 million gallons in 1998 (Camp Dresser & McKee, Inc. 1996). The South Lake is the largest by far and has approximately 70 percent of the total volume of water. The Impound Lake is smaller than the North Lake (Table 2).

Table 2. Volume and Depths of the Three Lakes of Lake Merced

Source: Camp Dresser & McKee, Inc (1996).

Depths of the lakes varies. Average depths of the lakes are: 5.5 ft, 9.8 ft, 13 ft for the Impound, North, and South lakes, respectively.

Inflow and outflow of the lake is briefly summarized here. Inflow of the lake is controlled predominantly by precipitation and the lake’s relationship to groundwater. Seepage from the groundwater through fractures of rocks supplies water to the lake. Outflow of the lake is mainly by evaporation (93%), and seepage to ground water (7%). Water is occasionally pumped directly out of the lake, as in 1990 (Camp Dresser & McKee 1996). Details of the inflow and outflow of the lake and related factors will be discussed in "Lake Merced Hydrology and Water Quality".

Conclusion

The geology and hydrogeology of Lake Merced is a very complex subject, and this statement also applies to any physical, biological, or anthropologic factors analyzed to describe the biogeography of Lake Merced. Also, a lack of information and a difficulty locating sources were problems encountered by the researchers. However, speculating the results may sometimes be necessary in a case of paleontological study because some of the key features may not have been preserved in fossil records. Understanding the hydrogeology is similarly complicated by a great magnitude of speculation involved in the analyses. However, advancements in Geographic Information System have provided easier and more precise ways to do computer modeling. Several modeling projects have been conducted in the past including the Westside Basin groundwater models (CH2MHILL 1997).

This section has summarized key features to characterize some of physical settings of Lake Merced. There is more to be investigated to come to a complete understanding of the geology and hydrogeology the lake. However, this does not mean that we do not have to take a holistic approach toward the preservation Lake Merced and its surroundings. Rather, we should end the abusive use of the ecological services provided by the lake and minimize human impacts.

References

Camp Dresser & Mc Kee. 1999.  Lake Merced water sanitary survey November 1999 Report. San Francisco Public Utilities Commission.

Clifton, H.E. and R.E. Hunter 1987. "The Merced Formation and related beds: A mile-thick succession of late Cenozoic coastal and shelf deposits in the sea cliffs of San Francisco, California". In: Geological Society of America Centennial FieldGuide—Cordilleran. Section. pp. 257-262.

Clifton, H. E., Hunter, R.E., and Gardner, J.V. 1988. "Analysis of eustatic, tectonic, and sedimentologic influences on transgressive and regressive cycles in the late Cenozoic Merced Formation." In: Cordilleran Section of  Geological Society of America Centennial Field Guide. Volume 1.  Paola, C., and Kleinspehn, K.L. Boulder, CO (Eds.): Geological Society of America. Pp.27-262.

Fahy, Neil E. 1974. "Origin of Lake Merced." California Geology: August 1974.

Geo/Resource Consultants, Inc.. 1993. Lake Merced Water Resources Planning Study.  San Francisco Water Department in association with Montgomery/Watson, Jones and Stokes Asso., Inc. Public Affairs Management. San Francisco, CA GRC Project No. 1756-00.

Harden, Deborah R. 1997. California Geology. Upper Saddle River, NJ: Prentice Hall, Inc.

Hengesh, James V. and John Wakabayashi. 1994. Quaternary Deformation Between Coyote Point and Lake Merced on the San Francisco Peninsula: Implications for Evolution of the San Andreas Fault. USGS National Earthquake Hazard Reduction Program, Fiscal Year 1994 Award No. 1434-94-G-2426. [Online] Available: http://erp-web.er.usgs.gov/reports/annsumm/g2426.htm

Hunter, R.E., Clifton, H.E., Hall, N.T., Csaszar, G., Richmond, B.M., and Chin, J.L.. 1984. "Pliocene and Pleistocene coastal and shelf deposits of the Merced Formation and associated beds, northwestern San Francisco Peninsula, California". In: Society of Economic Paleotonlogists and Mineralogists Field Trip Guidebook 3. Midyear Meeting. pp. 1-29.

Jorgenson, Pat. 2000. "The Demise of the San Bruno Fault, or the Fault that Never Was." USGS News Release. [Online] Available: http://www.usgs.gov/public/public_affairs /press_releases/pr1171.html

Konigsmark, Ted. 1998. Geologic Trips: San Francisco and the Bay Area. Gualala, California: GeoPress.

Rogge, Erdmann. 2000. Personal Communication

Steila, Donald and Thomas E. Pond. 1989. The Geography of Soils: Formaiton:Distribution and Management. 2^nd ed. Savage, Maryland: Rowman & Littlefield Publishers, Inc.

United States Department of Agriculture, Soil Conservation Service. 1991. Soil Survey of San Mateo County, Eastern Part, and San Francisco County, California.  In cooperation with the Regents of the University of California (Agricultural Experiment Station).

Sullivan, Raymond and Jon S. Galehouse. 1991. Geological Setting of the San Francisco Bay Area. In: Geologic excursions in Northern California: San Francisco to the Sierra Nevada. Edsitors, Doris Sloan and David L. Wagner. Sacramento, CA: California Dept. of Conservation, Division of Mines and Geology.

Yates, Eugene B, Scott N. and Lisa Horowitz-McCann. 1990. Geohydrology, Water Quality and Water Budgets of Golden Gate Park and the Lake Merced Area in the Western Part of San Francisco, California. U.S. Geological Society, Water Resources Investigation Report 90-4080. Prepared in Cooperation with the S.F. Water Department. Sacramento, CA 1990.

Wallace, Robert E, ed. 1991. The San Andreas Fault System, California. USGS Professional Paper 1515. Washington, D.C.: United States Government Printing Office.