The study of paleoliquefaction grew out of (1) the recognition that earthquakes left their imprint on soft sediments as deformational structures primarily through liquefaction, and (2) the need for applying paleoseismology to settings in which active faults were not readily recognizable, accessible, or did not reach the surface. Earthquake-induced liquefaction features are distinctive, and their formation is a result of strong ground shaking that may or may not result in lateral spreading. Paleoliquefaction features include sand blows, and intrusive dikes and sills, as well as less prominent, but equally informative features such as load casts, pseudonodules, ball-and-pillow structures, and convolute lamination. Fluvial depositional environments, with generally easy access and relatively abundant natural outcrops, have been the primary choice for conducting paleoseismic studies. In general, when relying on sand dikes and sills, sand blows, and their related structures common in fluvial sediments, one is restricted to river valleys. Depending on the density of the drainage network and the size of the streams, one may not obtain as much data as would be desired. Therefore, other environments that may contain earthquake-induced liquefaction structures may need to be sought out. Lacustrine and paleolacustrine deposits also have a distinctive suite of liquefaction-induced sedimentary structures, most commonly pseudonodules and load casts, and less commonly convolute lamination. However, these structures are not limited to lacustrine deposits, because they have been observed in paleoliquefaction source beds in the New Madrid seismic zone, liquefaction accompanying the 26 January 2001, Bhuj, India, earthquake, and liquefaction associated with the Charlevoix seismic zone. The earliest earthquake-induced paleoliquefaction features were described and correlated to specific earthquakes using soft-sediment deformational structures in lake sediments and a series of modern earthquakes in California, and deformational structures in prehistoric lake sediments. Ancient lake deposits have been used for paleoseismological studies with some success in the United States in California, Oregon, Washington, and Alaska. Such deposits have also been successfully used in Europe and the Middle East. The most promising of these studies have been in varved glaciolacustrine deposits. Varves are small-scale (centimeter to millimeter) sedimentary units. They form in a variety of marine and lacustrine depositional environments from seasonal variation in clastic, biological, and chemical sedimentary processes. The most common seismically induced structures that occur in varved glaciolacustrine deposits are pseudonodules.

This chapter presents a synthesis of data pertaining to post-early Miocene slip on the San Andreas fault in central California and suggests a three-phase evolition of the San Andreas system. The cricial evidence that supports the three phases of evolution conies from the reversed positions of two exotic rock fragments in the vicinity of Parkfield, California. The three-phase evolution of the San Andreas is also supported by the correlation of other exotic fragments, the basement rocks on which they lie, overlying Tertiary stratigraphic sequences, and distinctive Miocene strata derived from these fragments during their transport along the fault. The 40-km-long section of the San Andreas fault near Parkfield is characterized by exotic blocks composed of Cretaceous hornblende quartz gabbro at Gold Hill and lower Miocene volcanic rocks in Lang Canyon. The gabbro is correlated petrographically with similar rocks near Eagle Rest Peak, 145 km to the southeast, and near Logan, 165 km to the northwest. The lower Miocene volcanic rocks, informally termed the volcanic rocks of Lang Canyon, are correlated with the Neenach Volcanics 220 km to the southeast and the Pinnacles Volcanics 95 km to the northwest. All three fragments of volcanic rocks are unconformably overlain by similar successions of Tertiary sedimentary rocks. The original positions of the bodies of gabbro and volcanic bodies and their overlying sedimentary cover may be reconstructed from these exotic fragments that now lie along the San Andreas fault between San Juan Bautista and the northwestern Mojave Desert. The original undeformed gabbroic body was composed of the hornblende quartz gabbro of Eagle Rest Peak, Gold Hill, and Logan. In its initial prefaulted position, the original gabbroic body lay about 55 km northwest of the early Miocene volcanic assemblage. The undeformed volcanic assemblage was composed of the Neenach Volcanics, Pinnacles Volcanics, and volcanic rocks of Lang Canyon. The original spatial relationship between the undeformed gabbro and volcanic assemblage and their sedimentary cover is preserved in the present position of the gabbro of Logan and the Pinnacles Volcanics. However, in the Parkfield segment of the San Andreas, the gabbro of Gold Hill lies east of the main trace of the San Andreas fault, and the volcanic rocks of Lang Canyon lie 2 km west of the fault. The reversed relative positions of the gabbro of Gold Hill and the volcanic rocks of Lang Canyon suggest a complex history of movement on the San Andreas fault. Consequently, plainspastic reconstruction of these bodies and their overlying sedimentary cover is constrained by the unusual distribution of exotic blocks near Parkfield. The resulting proposed history of movement is divided into three stages that begins with the eruption of the early Miocene volcanic rocks about 24 Ma. The Neenach-Pinnacles Volcanics, erupted after passage of the Mendocino triple junction, were soon cut by the growing San Andreas transform system. During the first phase of movement the Salinian block, which contains the Pinnacles and Logan godies, was detached from the Mojave and Sierran blocks. The Pinnacles and Logan bodies were transported about 95 km northwest from the Neenach Volcanics and the gabbro of Eagle Rest Peak. At the end of the first phase, the Logan and Pinnacles fragments lay adjacent to the west side of what is now the San Joaquin Valley. Concurrently, fan-deltas deposited debris that was derived from the Gabilan Range, the fan-deltas spread across the San Andreas fault into the middle Miocene sea in the San Joaquin trough. During the second phase of movement, the San Andreas—at least locally—stepped eastward and detached a second fragment from the Neenach Volcanics. This fragment consists of the volcanic rocks of Lang Canyon. Slip was transferred to the new trace of the San Andreas fault, and the older trace became completely or largely inactive. After transferral of slip to the new trace of the San Andreas fault, the volcanic rocks of Lang Canyon and the Pinnacles Volcanics remained about 95 km apart on the Salinian Block west of the San Andreas fault. During the third phase, the Gold Hill fragement was slivered off the Logan fragment and was tectonically emplaced on the east side of the San Andreas fault when the Logan fragment lay at the latitude of Gold Hill. The process of slivering off of the Gold Hill fragment was accomplished by deformation of the San Andreas in an eastward bend along what is now the Jack Ranch fault. Bending of the fault was stimulated by the presence of highly sheared Franciscan rocks that crop out near the San Andreas and extend to great depth. Eventually the San Andreas bent to such a degree that slip could not be conducted around the bend, and a new, stable, straight segment was formed. The straightening of the fault resulted in slivering of the Gold Hill fragment from the Logan fragment. After detachment of the Gold Hill fragment, the Salinian block containing the gabbro of Logan, the Pinnacles Volcanics, and the volcanic rocks of Lang Canyon was transported an additional 160 km northwest to its present position. This reconstruction honors the current positions of all the related exotic fragments of gabbro, volcanics, and sedimentary rocks. The timing of the sequence of movements required to reconstruct the original bodies suggests that the three phases of evolution of the San Andreas fault in central California are characterized by increasing slip rates. The rate for the first phase probably averaged about 10 mm/yr over a period of about 8 m.y. The rate for the second phase averaged about 8 mm/yr over a period of about 7 m.y. The rate rate for the third phase averaged about 33 mm/yr over a period of about 5 m.y.

The actively deforming Clear Lake basin has been shaped primarily by shear and tensional stresses within the broad San Andreas fault system; and it has been modified both by eruption and subsidence of the Clear Lake Volcanics and by depositional processes. Within the San Andreas fault system, shear has been dominantly right-lateral N35°W to N45°W, and maximum tension dominantly east-west. However, the current local maximum tension deduced from focal-plane solutions is N70°W, representing a secular change of unknown, possibly short duration, that could modify tensional basin-forming processes. The Collayomi fault zone (N45°W) partially bounds the basin on the southwest. The seismically active Konocti Bay fault zone (N25°W) across the basin merges into the less well documented N40°W Clover Valley fault zone that includes the faults in Clover Valley and near Lucerne and shorter faults near Clearlake Highlands and Lower Lake. The northeast margin of the main lake and both margins of the Oaks arm are largely fault-controlled; the Highlands arm is partly fault-controlled. East of the basin, sediments now assigned to the Cache Formation—ranging in age from more than 1.8 to about 1.6 m.y. old—were deposited in an earlier basin that was controlled mainly by N35-45°W (Bartlett Springs) and N20°E (Cross Spring) fault zones. The earliest stage of the present Clear Lake basin is dated by the 0.6-m.y.-old rhyolite of Thurston Creek, which is found at or near the base of lacustrine-fluvial deposits. The extensive rhyolite’s thickness and distribution suggest that its eruption triggered subsidence that initiated or accelerated basin formation. Limited data suggest that a dominantly volcanic subsidence feature within the larger Clear Lake Basin could have had dimensions of about 13 by 15 km and either an elliptical shape with major axis oriented west-northwest, or a rectilinear shape. Volcanism occurring between 0.6 and 0.3 m.y. ago partly filled the southern part of the Clear Lake basin with flows, pyroclastic materials, and clastic deposits. Deposition beneath Clear Lake (0.4 mm/yr for the past 0.45 m.y.) has kept pace with subsidence and tilting down to the northeast. Projected maximum sedimentary and volcanic thicknesses could total more than 1 km beneath Clear Lake, suggesting a possible overall subsidence rate of 1.7 mm/yr for the past 0.6 m.y.

Clear Lake, California, lies in a volcano-tectonic depression that received nearly continuous lacustrine deposition for the past 500,000 yr and probably longer. The lake has been shallow (<30 m) and eutrophic throughout its history. Sediments beneath the floor of the lake are fine grained (chiefly >7.0φ) and contain fossils of a large lacustrine biota, as well as a pollen record of land plants that lived in the basin. The sediments also contain tephra units of local and regional extent. The ages of the sediments in Clear Lake are determined from radiocarbon dates on the sediments, from correlation of regionally distrbuted tephra units, and from inferred correlation of oak-pollen spectra with the marine oxygen-isotope record. From the chronology of events recorded in the cores from Clear Lake, the late Quaternary history of the lake can be deciphered and the sediments correlated with other basins in northern California. Comparison of cores from Clear Lake with strata of the Kelseyville Formation, exposed south of the main basin, suggests a general northward migration of lacustrine sedimentation, which in turn suggests a northward tilt of the basin. Migration of the lake was a response to volcanism and tectonism. Volcanic rocks erupted from Mt. Konocti, on the southern margin of the lake, and displaced the shoreline to the west and north. Clear Lake is bounded by faults that are part of the San Andreas fault system. These faults strongly influenced the position, depth, and longevity of Clear Lake. Movement on these boundary faults deepened the Highlands and Oaks arms of the lake about 10 ka. Tectonic movement accompanying faulting was probably also largely responsible for hiatuses in the deposits beneath Clear Lake about 17 and 350 ka that have been inferred from the subbottom stratigraphy of the lake. Climate change, although responsible for large variation in the composition of the terrestrial flora of the Clear Lake drainage basin, has not influenced the areal extent, depth, or position of the lake.

We describe the depositional environments of the Cache, Lower Lake, and Kelseyville Formations in light of habitat preferences of recovered mollusks, ostracodes, and diatoms. Our reconstruction of paleoenvironments for these late Cenozoic deposits provides a framework for an understanding of basin evolution and deposition in the Clear Lake region. The Pliocene and Pleistocene Cache Formation was deposited primarily in stream and debris flow environments; fossils from fine-grained deposits indicate shallow, fresh-water environments with locally abundant aquatic vegetation. The fine-grained sediments (mudstone and siltstone) were probably deposited in ponds in abandoned channels or shallow basins behind natural levees. The abandoned channels and shallow basins were associated with the fluvial systems responsible for deposition of the bulk of the technically controlled Cache Formation. The Pleistocene Lower Lake Formation was deposited in a water mass large enough to contain a variety of local environments and current regimes. The recovered fossils imply a lake with water depths of 1 to 5 m. However, there is strong support from habitat preferences of the recovered fossils for inferring a wide range of water depths during deposition of the Lower Lake Formation; they indicate a progressively shallowing system and the culmination of a desiccating lacustrine system. The Pleistocene Kelseyville Formation represents primarily lacustrine deposition with only minor fluvial deposits around the margins of the basin. Local conglomerate beds and fossil tree stumps in growth position within the basin indicate occasional widespread fluvial incursions and depositional hiatuses. The Kelseyville strata represent a large water mass with a muddy and especially fluid substrate having permanent or sporadic periods of anoxia. Central-lake anoxia, whether permanent or at irregular intervals, is the simplest way to account for the low numbers of benthic organisms recovered from the Kelseyville Formation. Similar low-oxygen conditions for benthic life are represented throughout the sedimentary history of Clear Lake. Water depths for the Kelseyville Formation of 10 to 30 m and 12 m near the margins of the basin are inferred both before and after fluvial incursions. These water-depth fluctuations cannot be correlated with major climatic changes as indicated by pollen and fossil leaves and cones; they may be due to faulting in this technically active region.

Clear Lake occupies a structural depression in the northern California Coast Ranges at an elevation of 404 m. Eight sediment cores were taken from the lake in 1973. This paper reports the palynology of cores CL-73-4 and CL-73-7. The former is 115 m long, and is interpreted to cover the entire last glacial cycle; the latter is 27.5 m long and covers at least the last 40,000 radiocarbon yr. The pollen record of core CL-73-4 is dominated by three pollen types (oak, pine, and TCT [Taxodiaceae, Cupressaceae, and Taxaceae]) that together account for between 75 and 99 percent of the pollen in each sample. Core CL-73-7 is similarly dominated by these pollen types, but aquatic and riparian pollen types are also locally important. The present vegetation around Clear Lake consists of oak woodland; mixed coniferous forest is found at higher elevations in the surrounding mountains. The present pollen rain into Clear Lake is dominated by oak pollen. During the cooler parts of the last glacial cycle, oak pollen influx to the sediments of Clear Lake was largely and at some times entirely replaced by coniferous pollen (mostly pine and TCT) in response to vertical migration of vegetation belts caused by climatic changes. Pollen data were reduced using a Q-mode factor analysis. Five factors were defined that account for more than 98 percent of the variance. Three of the factors summarize aspects of the behavior of the regional forest vegetation around Clear Lake, and two summarize the behavior of the aquatic and swamp vegetation in the lake itself. Zoning of the pollen diagrams was accomplished using an iterative program that minimized the total sums of squares of the factor loadings within zones. Twenty-one pollen zones are defined for core CL-73-4, four for core CL-73-7. The pollen zones of core CL-73-4 are used to propose a series of informal climatic units that include the time interval from the penultimate glaciation to the present. The major units proposed, from oldest to youngest, are: (1) Tsabal cryomer, (2) Konocti thermomer, (3) Pomo cryomer, and (4) Tuleyome thermomer (Holocene). The Pomo cryomer is divided into early, middle, and late phases. The early Pomo includes a series of five cold/warm oscillations that are designated the Tsiwi cryomers and the Boomli thermomers, numbered from Tsiwi 1 (oldest) to Boomli 5 (youngest). Middle Pomo time includes the Cigom 1 cryomer and the Halika thermomers, a series of three minor warm intervals. Late Pomo time includes the Cigom 2 cryomer and a transitional interval following it and preceding the Holocene. The climatic oscillations of the Tsiwi cryomers and Boomli thermomers were often quite abrupt; both sudden warmings and sudden coolings occurred. The most severe of these changes was the cooling that occurred at the end of the Konocti thermomer, when oak pollen frequencies dropped from more than 60 percent to about 25 percent within a stratigraphic interval of only 23 cm. These sudden changes were climatic catastrophes for the ecosystems that experienced them. The record in the sediments of algae with acid-resistant remains indicates that lake productivity was relatively high during warm intervals in the past, and that overall productivity increased as the lake became shallower and its thermal inertia decreased. The lake waters were probably transparent during the cooler parts of the last glacial cycle, but Clear Lake has probably not been as clear a lake during the Holocene.

Clear Lake core CL-73-4 records fluctuating abundances of oak pollen during the last glacial/interglacial cycle that correlate remarkably well with fluctuations in extensive pollen records from Grande Pile in France and Tenaghi Phillipon in Macedonia, as well as with the oxygen-isotope records from deep-sea cores. The record correlates less closely with other extensive records, including those for Lake Biwa, Japan, and Sabana de Bogotá, Colombia. Correlation of the record with the early Weichselian climatic sequence of northwestern Europe is excellent; both sequences show a series of five cryomer/thermomer fluctuations between the end of the last interglaciation (Eemian/Konocti, which is correlated here with the end of marine oxygen-isotope Stage 5e) and the onset of full continental glaciation at the end of Stage 5a. The fluctuations correlate both in their relative durations and in their relative amplitudes. The Clear Lake record also correlates with various North American sequences. The Sangamon interval of the mid-continent area correlates with the entire Konocti thermomer and early Pomo cryomer interval, and correlations with the glacial sequences of the Sierra Nevada and Rocky Mountains suggest that some Tahoe, Mono Basin, and Bull Lake moraines may be of Sangamon age. The proposed correlations of the Clear Lake record with other sequences have not been proved. The overall impression, however, is one of remarkable consistency, and it is likely that further work will provide more evidence in support of the sequence of five cryomer/thermomer cycles between the end of the last interglacial period and the onset of full glacial conditions about 70,000 years ago. This sequence is much more complicated than has been generally recognized, although parts of it have been known for many years. The sequence, which has now been found in several widely separated areas, should no longer be ignored.

Fossil diatoms from a 177-m core (CL-80-1) taken near the center of the main basin (Upper Arm) of Clear Lake, California, provide evidence about the stratigraphie relationships, age, and environmental history of these lacustrine deposits. In general, diatom assemblages from the core are dominated by planktonic genera such as Stephanodiscus , Cyclotella , and Melosira . Shallow-water species of Fragilaria and Amphora are common and sometimes abundant. Several planktonic diatoms from the core are also found in the Kelseyville Formation, which is exposed on the southern margin of Clear Lake and is inferred to underlie the modern lacustrine deposits. The presence of these taxa in the same stratigraphie order in both the Clear Lake core and the Kelseyville Formation suggests partial correlation between the two and implies a relationship between the Kelseyville Formation and the lacustrine sediments beneath Clear Lake. In the upper 50 m of core CL-80-1, diatom assemblages apparently reflect late Pleistocene and Holocene paleoenvironmental changes, although their environmental significance may be obscured by reworking of diatoms from older sediments, by tectonically caused changes in patterns and rates of sedimentation, and by the impact of volcanism. Nevertheless, the diatoms indicate that lacustrine environments have been characterized by fresh, moderately deep, nutrient-rich water throughout much of their sedimentary history. Cooler climatic and lacustrine environments of the late Pleistocene were characterized by a codominance of Stephanodiscus and Melosira species, implying a mesotrophic to eutrophic, stratified lake. After the change from Pleistocene to Holocene climates, Clear Lake became yet more eutrophic and turbid. Stratification was short-term and irregular, and warm-water conditions extended throughout a greater portion of the growing season although there is evidence for a middle Holocene return to cooler and moister conditions. The modern limnology of Clear Lake, which is characterized by massive blooms of blue-green algae and by the abundance of Melosira granulata , apparently began about 15,000 years ago.

Modern-day Clear Lake is a turbulent, turbid, permanent, polymictic lake. It is also a highly productive fresh-water lake whose dominant solutes are Mg 2+ + Ca 2+ – HCO 3 − . The lake has a diverse and abundant limnetic community, yet has a depauperate benthic community. The benthic community structure appears to be ecologically simple as the result of turbulence-induced substrate instability coupled with unpredictable periods of anoxia induced by the oxygen-consuming organic matter. Ostracodes, which are ubiquitous, largely benthic, environmentally sensitive, and diverse organisms, are represented in Clear Lake only by the nektic species Cypria ophtalmica . The substrate conditions provide adequate reason for the absence of most ostracodes from the modern lake, but their absence in the fossil record suggests that the modern lacustrine environment existed in the past despite known climate changes. This seeming paradox can be explained by considering the influence of various types of climatic change on the lacustrine environment; certain types of climate-environmental changes would maintain a lacustrine environment unsuited to ostracodes. These hypothesized climate-lacustrine environmental changes would favor the modern (Holocene) and other oak-dominated periods to be the warmest and driest in Clear Lake history, whereas the pine-TCT (Taxodiaceae, Cupressaceae, and Taxaceae) periods would be cooler and wetter than today. The largely barren ostracode record, coupled with rare ostracode occurrences, would support a glacial-interglacial Clear Lake climatic history characterized primarily by changes in the annual precipitation-evaporation budget.

We have identified ash beds in sediment cores of Clear Lake, California, by the chemistry of their volcanic glasses and petrography. These identifications enable us to correlate between cores, and to correlate three ash beds to several localities outside the Clear Lake basin where they have been isotopically dated or their ages estimated by stratigraphically bracketing dates. The three dated ash beds are ash bed 1 (Olema ash bed), estimated to be between 55 and 75 ka, in two deep cores CL-80-1 and CL-73-4, and two ash beds in core CL-80-1, ash bed 6 (Loleta ash bed), estimated to be between 0.30 and 0.39 Ma, and ash bed 7, estimated to be about 0.4 Ma. Available age control from extrapolation of radiocarbon ages downward in the two cores, age constraints from correlations of ash beds, and etching of mafic minerals in ash beds at depths below about 118 m in core CL-80-1 suggest the following depositional histories for the two cores: in core CL-73-4, sedimentation appears to have been rapid (about 1 mm/yr) and continuous from about 120 ka to the present, corresponding to a depth interval from about 115 m to the present lake bottom. In the deeper core CL-80-1, sedimentation took place at a relatively moderate rate (0.4 mm/yr) from about 460 ka until sometime between about 300 and 140 ka, corresponding to a depth interval from about 168 to 118 m. Slow deposition or erosion took place sometime during the interval from about 300 to 140 ka, corresponding to an inferred hiatus at a depth of about 118 m. From 140 ka to the present, rapid sedimentation took place at about the same rate (about 0.8 mm/yr) as in core CL-73-4, corresponding to a depth interval from about 118 m to the present lake bottom. The age of sediments in Clear Lake is not well constrained within the depth interval of about 70 to 130 m in the two deep cores, and the duration of the putative hiatus at about 118 m in the core CL-80-1 may be shorter than we propose. The presence of a hiatus at about 118 m depth in this core, however, is suggested by etching of mafic minerals in tephra layers below this level but not above, indicating that a period of subaerial exposure, or exposure above the groundwater table, had occurred for sediments below this level.

Radiocarbon dating of disseminated organic matter from 10 horizons in Clear Lake core CL-73-4 produced apparent ages ranging from 4,230 to 32,650 B.P. Old carbon from lake sediments and springs beneath the lake adds about 4,200 years to the apparent age of each Holocene sample. A significant component of younger carbon—which cannot be completely removed by cleaning in sodium hydroxide solution—makes the dates older than 20,000 yr unacceptable. The younger dates are corrected for the old carbon effect, calibrated to the dendrochronologic time scale, and then used to derive a sedimentation rate for the Holocene part of the core. Sediment accumulation is expressed as the mass in kilograms per square centimeter of noncombustible overburden above a given level in the core in order to compensate for variations in degree of sediment compaction and organic content. The Holocene sedimentation rate, when applied to the entire core, yields an estimated core-bottom age of 133 ka. This independent evidence is consistent with the correlation of the high oak-pollen zone just above the base of the core with the last interglaciation. When the oak pollen maxima at the top and bottom of the core are equated with the Holocene and the last interglacial, the larger intervening fluctuations in the oak curve show a marked similarity to the climatic record preserved in deep ocean sediments. We correlate the major fluctuations of the oak pollen curve with their counterparts in the deep-sea record, and further refine the Clear Lake time scale by adjusting the age of the apparent Stage 5/4 boundary to 73 ka. The revised time scale indicates that sedimentation rates during the last glacial and interglacial were slightly higher and lower, respectively, than during the Holocene. According to the revised time scale, interstadial events in the Clear Lake pollen record appear synchronous with prominent radiocarbon-dated interstadials in other areas, as well as with high sea stands dated by uranium-series disequilibrium methods.

The diagenesis of amino acids in sediments from Clear Lake core CL-80-1 is indicated by changes in amino acid concentrations, compositions, and stereochemistry. Concentrations of total amino acids decrease with depth, but the decrease is not systematic, possibly reflecting a nonuniformity in sedimentary and postdepositional processes affecting the amino acids. Ratios of neutral/acidic amino acids may indicate that the pH of interstitial water is slightly alkaline to slightly acidic and that the organic matter is well humified. Ratios of nonprotein/protein amino acids suggest that some changes in amino acids with depth result from microbial degradations. The extent of racemization of alanine increases with depth; the trends of these data may be explained, in part, by rapid sedimentation within the lake. Agreement between extents of alanine racemization for sediments from equivalent depths in two cores from the lake suggests that diagenetic temperatures are uniform within the sediments of the northern basin of Clear Lake.

A deep-sea core collected on the continental slope off northern California contains a pollen stratigraphy for the past 20,000 yr that can be correlated to the pollen stratigraphy from the upper section of Clear Lake core CL-73-4. The occurrence in one sequence of pollen, reflecting the local continental paleoclimates, and marine microfossils reflecting the local paleoceanography, allows a comparison of concurrent responses of the local ocean and adjacent continental area to global climate changes. The interpretation of the two data sets gives a complex progression of changes that are probably interrelated, such as upwelling that produced coastal fogs. The changes in climatic and oceanographic environmental conditions that occurred in response to the switch from global glacial to interglacial conditions was not a smooth progression of increasingly moderate regimes; rather, the changes appear to be a complicated series of states that suggests a disequilibrium mode lasting from about 15,000 to 5,000 yr ago.

Clear Lake in Lake County, California, has an endemic fish fauna composed of five lake-adapted forms derived from lowland stream-adapted forms present in surrounding drainage basins. Two of the five endemic forms are extinct. The three remaining endemics maintain themselves despite the destruction of sloughs and tule beds surrounding Clear Lake that are used for spawning and nursery areas. Trophic specializations of the endemic fishes indicate past selection for feeding on small benthic and pelagic invertebrates. The presence of fine particles in the substrate and the reduced activity of tributary streams for at least the past 10,000 yr are major hydrographic features contributing to the evolution of these trophic adaptations. Subfossil scales of the endemic Clear Lake tuleperch, ( Hysterocarpus traskii lagunae ) present in three U.S. Geological Survey cores (CL-73-7, -6, and -8), removed from the bottom of Clear Lake in 1973 were analyzed by Casteel and others (1975, 1977a, b, 1979) for age and growth rate. Periods of increased scale growth were inferred to represent warming of the lake. Comparison of the Casteel data with pollen data (Adam and others, 1981) indicate that maximum scale growth (core CL-75-8) occurred at about 19 ka (=15 ka, according to Robinson and others, this volume) during a cold interval. Fluctuations in scale density in cores CL-73-4 and CL-73-7, however, seem to follow fluctuations in oak pollen. It is therefore concluded that maximum-scale growth represents cool periods, whereas maximum-scale density represents warm periods in the history of the lake. During the period that maximum-scale growth occurred, Clear Lake basin may have also been closed off from surrounding basins and the lake enriched with nutrients.

Earthquake locations for the time period March 1975 through March 1983 indicate diffuse seismicity in the Clear Lake area, with the area to the southeast of the lake having the highest level of activity. Swarms lasting 2 days, with events of magnitude ≥3.5, have been observed in the Konocti Bay fault zone. In contrast, almost no microearthquake activity is associated with the Collayomi fault zone. A 10-km-wide northeast-southwest-trending zone of seismicity is apparent although there is no known corresponding geologic feature. Earthquake depths in the Clear Lake area are shallower than in the surrounding major right-lateral fault zones. Focal mechanism solutions show predominant strike-slip with significant normal dip-slip movement. The maximum and minimum compressive stress orientations, north-northeast and east-southeast, respectively, are consistent with the San Andreas right-lateral transform boundary region. However, the component of extensional stress illustrated by normal fault-plane solutions and the shallowness of seismicity may be related to an inferred nearby crustal partial melt body.

During 1973 and 1974, precision temperatures (±0.02°C absolute) were measured in four drill holes, ranging in depth from 42 to 122 m, in Clear Lake, Lake County, California. The departure of the measured temperatures from their predrilling values was found to vary considerably among the holes. It is a function of the length of time after drilling that the buoy system supporting the plastic casing survived the storms on the lake, as well as the duration of the drilling disturbance. With one exception, CL-73-7, most of the holes were lost within a month after drilling. Thermal-conductivity measurements on the sediment cores from the holes were measured using the needle-probe technique. These measurements indicate that the sapropelic muds that underlie most of the main body of the lake and the peat of the southeast, or Highlands Arm, of the lake have thermal conductivities only slightly greater than that of water. Conductivities of the coarse-grained sediments associated with the deltaic deposits of Kelsey Creek are two to three times that of water and are not grouped as tightly as the fine-grained muds and peats. Heat flows calculated from the above measurements are 1.5 to 1.6 heat-flow units (HFU) in the main basin of Clear Lake and about 2.4 HFU in the Highlands Arm of the lake. These values are considerably lower than expected, based on heat-flow measurements in The Geysers 25 km south of Clear Lake. A correction for an average sedimentation rate in the lake of 0.68 mm/yr would raise the observed heat flows about 13 percent. Although the exact thickness of sediment is unknown, a correction for the refraction effect caused by the thermal conductivity contrast between the low-conductivity lake sediments and the higher conductivity surrounding rock would tend to increase the heat flow at depth in the main basin only slightly for any reasonable thickness of sediment. In the Highlands Arm of the lake, the geometry is different, and the correction could raise the observed heat flux at depth as much as three times that observed, again for reasonable sediment thicknesses. However interesting these numbers are, it should be cautioned that sediment composition and thickness below about 200 m in the lake are unknown, and large differences in conductivity could drastically change these corrections. The possibility of such high heat flows, though, is encouraging enough that future measurements at greater depths should be contemplated. Aside from the refraction effects there are other possible causes of the low heat flows: (1) Clear Lake is bounded by faults and at least one fault is inferred to pass beneath the main body of the lake—downward cold water movement along such faults could absorb heat and decrease the heat flux; and (2) water may be moving down through the sediments, although generally the permeability of lake sediments is rather low. Depending on the velocity of the movement, a substantial reduction in heat flow could result. The downward movement of water over such a large area (114 km 2 ) could be a source of recharge for The Geysers-Clear Lake geothermal system.