HESSDHydrology and Earth System Sciences DiscussionsHESSDHydrol. Earth Syst. Sci. Discuss.1812-2116Copernicus GmbHGöttingen, Germany10.5194/hessd-12-6505-2015Temperature and rainfall estimates for past 18 000 years in Owens Valley, California with a coupled catchment–lake modelYuZ.zhongbo.yu@unlv.eduhttps://orcid.org/0000-0002-3471-8577DongW.JiangP.State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing, ChinaDepartment of Geoscience, University of Nevada, Las Vegas, Las Vegas, Nevada, USADepartment of Conservation & Natural Resources, Nevada Division of Environmental Protection, Las Vegas, Nevada, USADivision of Hydrologic Sciences, Desert Research Institute, Las Vegas, Nevada, USAZ. Yu (zhongbo.yu@unlv.edu)03July20151276505653922April201509June2015This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://hess.copernicus.org/preprints/12/6505/2015/hessd-12-6505-2015.htmlThe full text article is available as a PDF file from https://hess.copernicus.org/preprints/12/6505/2015/hessd-12-6505-2015.pdf
Closed-basin lakes are intricately linked to the
hydrological systems and are very sensitive recorders of local
hydro-climatic fluctuations. Lake
records in closed-basins are usually used to
investigate the paleoclimate condition which is critical for understanding
the past and predicting the future. In this
study, a physically based
catchment–lake model was developed to extract
quantitative paleoclimate information including temperature and rainfall
over the past 18 000 years (ka) from lake records in
a hydrologically closed basin in the Owens River Valley,
California, US. The
initial model inputs were prepared based on current regional climate
data, boundary conditions from the General Circulation
Model, and fossil proxy data. The
inputs subsequently were systematically varied in order to produce the
observed lake levels. In this way,
a large number of possible paleoclimatic combinations can quickly narrow the
possible range of paleoclimatic combinations that could have produced the
paleolake level and extension.
Finally, a quantitative time-series
of paleoclimate information for those key times was
obtained.
Introduction
The information on past climate change has been extracted from
a variety of archives such as trees, ice, sediments, and corals
(Anchukaitis et al., 2013; Marcott et al., 2013; Steinman et al.,
2012; Viau et al., 2012). These archives can provide
a qualitative interpretation of the history of Earth's climate. In
past few decades, climate models emerge as an effective way to
quantitatively test the hypotheses of past climate change
(Kutzbach, 1987; COHMAP-Members, 1988; Street-Perrott and
Harrison, 1985). However, they are not reliable at regional scale
especially for variables such as precipitation (Kutzbach, 1987;
Groppelli et al., 2011; Jiang et al., 2013). To overcome these
shortcomings, an integrated utilization of numerical modeling and
paleocliamte archives is proposed to provide quantitative
paleoclimate information at region scale. Among these various
paleoclimate archives, lake levels in closed basins are the most
sensitive indicators of the water balance between precipitation
and evapotranspiration (Street-Perrott and Harrison, 1985).
In southern Great Basin, the regional distinctive tectonic
settings and geomorphic characteristics create many hydrologically
closed basins that were filled with water during pluvial lake
periods (Smith and Street-Perrott, 1983; Street-Perrott and
Harrison, 1985; Phillips, 1994; Bischoff et al., 1997; Menking
et al., 1997; Lowenstein et al., 1999; Bischoff and Cummins,
2001). Most of them shrunk or dried up under the current dry and
hot desert climate over this region. The evolution of lake size is
determined by various climate variables such as wind velocities,
relative humidity, temperature, and etc. Among them, precipitation
is the most important one (Smith, 1991). For instance, lakes in
Owens River system, California, are termed as “natural rain
gauges” (Smith and Bischoff, 1997) as they can track the changes
in precipitation within the catchments they are located
at. However, interpretations of climate changes based on the water
level changes is generally limited to identification of wetter or
drier conditions, and provide little information about the
specific nature of the climate change. The reason for this is that
lake level in a particular basin is a complicated function of
intra-basin and extra-basin climate and basin topography (Magny,
2004; Jones et al., 2001; Angel and Kunkel, 2010; Benson and
Thompson, 1987). In order to extract quantitative paleoclimatic
proxies from these lake records, one of the best approaches is
through numerical modeling.
A variety of models have been used to simulate the paleo-record of
closed basin lakes in arid and semiarid areas (Kutzbach, 1980;
Benson, 1981, 1986; Hostetler and Bartlein, 1990; Hostetler and
Benson, 1990; Hostetler et al., 1993, 1994; Ghile et al., 2014).
Physically-based lake models, which explicitly represent the
physical processes governing the energy and water balances of the
lake, offer a more robust way to predict climate induced changes
in water volume, level, and outflow of the lakes. A suitable lake
model for paleolake level studies should require a minimum of
site-specific parameters (Hostetler and Giorgi, 1993). In this
paper, we developed a coupled catchment–lake model and used it to
quantitatively estimate paleoclimate information, especially
annual mean precipitation and temperature in southwestern Great
Basin since the last glacial maximum (LGM). To test the hypothesis
that quantitative paleo-climate variables can be obtained through
numerical modeling in the closed basin area, we first derived the
past lake extent from field evidence, then we conducted an inverse
modeling with a physically based lake model for the simulation of
the derived lake extent at a specific time, and finally
reconstructed the climatic conditions. Compared to the
paleo-climate information derived from various archives or
simulated by the climate models, the proposed method can provide
quantitative estimate on temperature and precipitation at finer
regional resolution.
Owens River system
The Owens River system is located at the western margins of the
Great Basin. It is a hydrologically closed basin that consists of
a chain of lakes including Mono Lake, Owens Lake, China Lake,
Searles Lake, Panamint Lake, and Death Valley Lake (Fig. 1). The
floors of all these lakes except Mono Lake are now occupied by
playa lakes or salt flats. The valley is bound on the west by the
Sierra Nevada, on the northeast by the Inyo and White Moutains,
and on the southeast by the Coso Range. Presently, Owens River
drains an area of about 8550 km2. Due to the strong
rain shadow effect, most of the runoff is derived from about
16 % of the catchment area, which lies on the eastern slope of
the Sierra Nevada (Lee, 1912). Modern climate in the floor of the
Owens River system is semi-arid with about 15 cm of annual
precipitation. Thus, precipitation that fell directly on the
surface of the basins is an insignificant contribution to the lake
water budget, which could also be true for the lakes in the
paleo-Owens River system (Jannik et al., 1991). Street-Perrott and
Harrison (1985) termed such lakes as “amplifier” lakes, which
describe a simple relationship among basin runoff, lake
evaporation, and lake area (Smith and Bischoff, 1997).
Owens Lake, at the base of high mountains, responds firstly to the
increasing regional precipitation. The Owens Lake is the terminal
of the Owens River, it was about 10 m deep and
290 km2 in area before agricultural irrigation became
extensive in 1912. All of the river's water was diverted to Los
Angeles in 1912, and the lake desiccated. Searles Lake was the
third in a chain of five permanent lakes receiving water from the
Owens River during the late Pleistocene, and Mono Lake was
separated from the Owens River system to the south by
a high-altitude sill in the late Wisconsin (Benson and Thompson,
1987). During the LGM, Owens Lake, China Lake, and Searles Lake
overflowed and the Panamint Valley is a terminal of the Owens
River hydrological system. Lake stages at Searles Lake are
sensitive to climatic changes because of the storage capacities of
the upper lakes in the series. Inflow to Searles Lake depends on
overflow from the other lakes and therefore is first affected by
a decreasing inflow in their lake system.
Studies on lacustrine outcrops, cores, and landforms have allowed
the reconstruction of the past histories of lakes in the Owens
River system and its downstream basins (Smith and Street-Perrott,
1983; Smith and Bischoff, 1997). The paleo lake levels are
recorded by the geomorphic and sedimentary evidence including
staircases of abandoned shorelines and abrupt changes of facies in
sediments. Smith and Street-Perrott (1983) provide a chronology
of Late Wisconsin to present lake level fluctuations for Searles
Lake (Fig. 2).
Benson et al. (1997) identified two hiatuses at 2.25 and
9.2 m based on the 14C data of Core OL84B from
Owens Lake. These two hiatuses represent two desiccation events
that occurred ∼15.3–13.5 and ∼6.1–4.3 ka (all ages
used in the text of this paper are on 14C time scale
before present (BP)) in Owens Lake. The δ18O data
of sediments between the two hiatuses show four abrupt dry/wet
oscillations that have their correspondents in the North Atlantic
region. Relatively wet intervals precede each of the dry events.
An extreme overflow occurred at about 12 ka, which resulted in
the lowest δ18O (-13 ‰) of lake
carbonate (Benson, 1999). Based on the ostracode assemblage from
the Owens Lake, Forester (2000) derived more details on lake level
changes of the Owens Lake from 25 to 4 ka; Li et al. (2000)
provided detailed information on climate for the past
1000 years; and Benson and Paillet (2002) plotted
δ18O with age for past 18 ka (Fig. 2).
Bacon et al. (2006) suggested that pluvial Owens Lake had dropped
45 m from its latest Pleistocene highstands of
1145 m by 11 600 yr BP. This lowstand was
followed by an early Holocene transgression that attained
a highstand near 1135 m before dropping to 1120 m
at 7860–7650 yr. The lake then lowered another
30 m to shallow and near desiccation levels between 6850
and 4300 yr BP and minor lake-level rise after
4300 yr BP, followed by alkaline and shallow conditions
during the latest Holocene.
In summary, the detailed paleolake records in the Owens River
system offer a good opportunity to extract quantitative
paleoclimate information in the western Great Basin.
Description of model and modelling strategies
The surface area of a closed-basin lake under natural conditions is
strictly dependent on the dynamic equilibrium between precipitation
and evapotranspiration over its entire catchment (Halley,
1714). Any changes in this equilibrium result in a change in
terminal lake depth, which directly influences its area, and the
cumulative lake area in the drainage basin (Benson and Paillet,
1989). The mean annual water balance of a lake is governed by the
equation (Street-Perrott and Harrison, 1985):
ΔV=AL(PL-EL)+(R-D)+(GI-GO)
where ΔV is the net change in volume of the lake,
PL is precipitation on the lake, EL is
evaporation from the lake, AL is the area of the
lake, R and D are runoff from the catchment and the surface
discharge from the lake, respectively, and GI and
GO are groundwater flows into and out of the lake
respectively. For a closed basin lake, GI and
GO can be assumed negligible, and D is zero, so
Eq. () reduces to the following form for equilibrium
conditions:
R=AL(EL-PL).
If the runoff from the drainage basin can be represented by
R=AB(PB-EB)
where AB represents area of the catchment,
PB is the precipitation over the catchment, and
EB is the evapotranspiration over the catchment, then
we have
AB(PB-EB)=AL(EL-PL).
This simple expression shows that the equilibrium area of a closed
lake under natural conditions is strictly dependent on the
precipitation and evaporation over its catchment and water
surface. In the Owens River system, based on paleolake records,
values for AL and AB for those
paleolakes can be measured quite accurately using a digital
elevation model (DEM). Remaining components in Eq. () are
precipitation and evaporation over the drainage basin. The
evaporation mainly depends on temperature, thus the purpose for
this paper is to develop a coupled catchment–lake model to resolve
these two unknown variables in the Eq. ().
The evaporation value depends on many climatic factors including
solar radiation, temperature, wind speed, and cloud cover. With
the exception of temperature, other relevant factors are difficult
to reconstruct from geologic data. Many studies assume that
paleo-values for the evaporation can be satisfactorily estimated
from empirical relationships between modern data on evaporation
and air temperature. However, a change in evaporation rates could
result from higher wind velocities, higher relative humidities and
lower solar radiation values, and greater amounts of precipitation
on the lake surface (Smith and Street-Perrott, 1983). Therefore,
it is desirable to have a model that considers all of these
factors. Hostetler and Bartlein (1990) developed one-dimensional
surface energy-balance lake model, where the vertical heat
transfer was simulated by eddy diffusion and convective
mixing. Several studies using this model have successfully
simulated the modern and paleolake level change both in humid and
arid regions (Vassiljev, 1997; Hostetler and Benson, 1990;
Hostetler et al., 1994). Orndorff (1994) developed a surface
hydrologic model (OSHM) that has been successfully applied to the
Owens River system to test the proxy estimates of the LGM against
the paleolake records. The OSHM has three modules: the snow module
that computes mean monthly snowfall, snowmelt, snowpack, ice
accumulation, ice transport, and icemelt for each grid cell based
on the input of temperature, precipitation, and elevation of that
cell, the runoff module that calculates mean annual runoff from
available water (rain, snowmelt, and icemelt), and the lake module
computes lake extent from the results of basin-wide mean annual
runoff calculated by the runoff module and lake evaporation. In
the OSHM, empirical relationships between modern data on
evaporation and air temperature were used to calculate the
evaporation for the pluvial lakes in the Owens River system during
the LGM, which may not represent the actual situation as discussed
above. In this study, the lake module in the OSHM was modified
with the addition of Hostetler's lake model and used to simulate
evaporation over the lake surface.
The paleoclimatic evaluation strategy is first to model lake
extent under current climatic conditions with the coupled
catchment–lake model developed in this study. The input parameters
under current climatic conditions are then systematically varied
in order to reconstruct lake extent based on lake records under
paleoclimatic conditions. The modeling strategy is essentially an
inverse approach to inferring paleoclimatic conditions based on
past lake extent. The advantage of this modeling strategy is that
a large number of possible paleoclimatic combinations can be
quickly narrowed to a possible range of temperature/precipitation
combinations that could have produced a particular paleolake
extent.
A simulation on lake extent begins with dry closed basins In order
to prevent water from overflowing into previously considered
basins, basins are taken into account in order from the highest to
lowest active outlet elevation, which means that the Monon Lake is
first full and overflows to Owens Lake, then the Owens Lake is
full and overflows to China Lake, then China Lake is full and
overflows to Sears Lake, the Sears Lake is full and overflows to
Panamant Lake, finally the Panamant Lake is full and overflows to
Death Valley Lake. Each basin's runoff volume from its catchment
is added to the current lake volume at each time step. The lake
volume is compared to the basin threshold volume, which
corresponds to a lake level equal to the controlling elevation of
the lowest basin outlet. Overflow occurs into the basin on the
other side of the outlet when the lake volume exceeds the
threshold volume. The model also checks for lake merging during
overflow, which occurs when two lakes with a common active outlet
overflow, thus inundating the active outlet. If two lakes merge,
the downstream basin becomes a part of the upstream basin, and the
remaining outlets of both basins are sorted to determine the new
active outlet for the complex basin (Orndorff, 1994). Lake
evaporation is then calculated using the eddy diffusion and
convective mixing (Hostetler and Bartlein, 1990), and the lake
level is adjusted accordingly. The simulation runs in one year
time step until the cumulative lake volume equilibrates or the run
time exceeds a specified end time. Benson and Paillet (1989) state
that “the proper gage of lake response to change in the
hydrologic balance is neither lake depth (level) nor lake volume
but instead lake surface area”, thus this study focused on lake
surface area for a comparison of simulated lake extent and derived
lake extent based on geologic evidence.
Calibration of catchment–lake model
The catchment–lake model used in this study was developed by
coupling a distribution hydrology model (Orndorff, 1994) and an
energy-balance lake model (Hostetler and Bartlein, 1990). Both of
these two models were independently calibrated with observed data
(Orndorff, 1994; Hostetler and Bartlein, 1990; Hostetler, 1991),
thus they are valid when use them independently. However, the
catchment–lake model developed for this study has to be calibrated
before applying it to simulate paleolake levels. The Mono Lake is
presently only lake with standing water in Owens Valley. The three
major streams (Rush, Lee Vining, and Mill Creeks) that delivery
water to Mono Lake originate in the high Sierra Nevada (Benson
et al., 1990), so the hydrological characteristics of Mono Lake and
Owens Lake is similar. Observed data including climate data,
hydrological data and lake level data are available for the Mono
Lake drainage basin since 1857
(Mono-Basin-Environmental-Impact-Report, 1993). The data on
measured temperature profiles and lake evaporation are also
available for some periods of time (MacIntyre et al., 1999). The
calibration was done with input data including modern precipitation
and temperature matrix data for the Mono Lake drainage basin from
the LCM (Stamm, 1992), solar radiation data from the Desert Rock,
vapor pressure, wind speed and cloudiness data from stations close
to the Mono Lake. The simulated monthly runoff for the Mono Lake
drainage basin is compatible with the observed (Fig. 3) and the
annual runoff is about 1 % less than the observed. The
simulated lake surface area is 224 km2 that is 1.8 %
less than the average of the observed lake surface area from 1940
to 1989 (Mono-Basin-Environmental-Impact-Report, 1993). The lake
temperature profile simulated by the model agreed very well with
measured temperature profile (Fig. 4a). Furthermore, the simulated
evaporation also compares well with the observed evaporation data
through the grant pan, but slight higher than the evaporation
estimated from water budget method (Fig. 4b). These comparisons
indicated that the overall ability of the coupled catchment–lake
model developed here to reproduce observed basin-wide mean annual
runoff, mean lake surface area, temperature profile and evaporation
of lake water in modern Mono Lake drainage basin.
Input parameters
A number of input parameters are required for the coupled
catchment–lake model. Coarse grid cell (5km×5km) used in the OSHM missed some small snow cover and
stream networks, and did a poor job representing some basin shapes
(Orndorff, 1994). In this study, the fine resolution
(1km×1km) data was used to obtain
better results. The topographic data used in this model is from
the global 30 arcsec elevation data set (GTOPO30)
(https://lta.cr.usgs.gov/GTOPO30). The
GTOPO30 has a horizontal grid spacing of 30 arcsec
(approximately 1 km). Observed solar radiation, cloud
cover, wind speed, atmospheric pressure from near weather
stations, and modern monthly temperature and precipitation matrix
from local climate models (Stamm, 1992) that are based on boundary
conditions including terrain, wind field, and radiation balance
were used to drive the newly developed catchment–lake model and to
reproduce the historical lake level of the Owens
Lake. Precipitation and temperature matrix data at 18, 15, 12, 9,
and 6 ka were prepared based on the proxy data in Table 1 by
applying the appropriate perturbation (simple additive change for
temperature, and multiplicative change for precipitation) to the
modern monthly precipitation and temperature matrix data from the
LCM. For example, climate at the LGM might be hypothesized to be
5∘ colder and 50 % wetter than the present based on
proxy data in Table 1. The input climate matrix at the LGM could
be prepared for temperature by subtracting 5 ∘C from the
modern temperature matrix, and for precipitation by multipling 1.5
to modern precipitation matrix. Other climate parameters including
cloud cover (Fig. 5a), solar radiation (Fig. 5b and c), and wind
speed (Fig. 6) were from historic records for modern conditions,
and from the Community Climate Model (CCM0) (Kutzbach and Guetter,
1986) and the results of paleoclimate simulation of North America
(Bartlein et al., 1998) for paleoclimatic simulation. However, the
single monthly value for these parameters was used for whole
area. The reason for this is: first, there is no such final
resolution data available for these parameters in the study area;
second, previous study indicated that precipitation and
temperature are two primary factors controlling the glacial extent
(Plummer and Phillips, 2003).
As most observation data are available only for Owens Lake and
Searles Lake, the comparison between the simulated results and the
observation data in this study focused on these two lakes.
ResultsSimulation on modern lakes
Mean annual runoff, computed for modern climate from the runoff
module is input to the lake model to simulate modern lake extent in
the Owens Valley. The resulting lakes along with hillshed and
stream network that were derived from the DEM are shown in
Fig. 7a. The lake system converges in 80 years. There is no
lake mergence occurring. The results from modern simulation
accurately portray Mono Lake, Lake Crowley (in the Long Valley
basin), Black Lake (in Adobe Basin), and Owens Lake. Simulated Mono
Lake has a surface area of 224 km2 that is about 2 %
less as compared to the average 228 km2 of the observed
lake surface area from 1940 to 1989
(Mono-Basin-Environmental-Impact-Report, 1993). Simulated Owens
Lake has a surface area of 302 km2 that is about 4 %
more as compared to an observed pre-diversion surface area of
290 km2 (Smith and Street-Perrott, 1983).
Simulations on lakes at Last Glacial Maximum
Simulations on lakes at the LGM in the Owens Valley were done with
modern temperature and precipitation matrices perturbed based on
proxy-based LGM temperature and precipitation departures (Table 1),
and other climate parameters including solar radiation, cloud
cover, and wind speed from CCM0 (Kutzbach and Guetter, 1986) and
the results of paleoclimate simulation of North America (Bartlein
et al., 1998) that were fixed. By varying combinations of
temperature and precipitation with appropriate perturbation until
the derived lake extent at the LGM from field evidence was
reproduced, the final combination with temperature 5.5 ∘C
cooler and 1.25 times precipitation of modern climate conditions
was obtained. Simulated final lake extents with hillshed and stream
network are shown in Fig. 7f. Owens Lake overflows, and has a lake
surface area of 692 km2. China Lake and Searles Lake
coalesce and have an area of 949 km2. Searles Lake also
overflows and a small lake with an area of 144 km2 was
formed in Panamint Valley. These results are very compatible with
observed lake extent in Owens Valley at the LGM reported by Smith
and Street-Perrott (1983).
Simulation on lakes at 15 ka
The techniques used to prepare input data for simulations on lakes
in Owens Valley at 15 ka are same as them for the LGM. Based on
14C and sedimentary features, Benson et al. (1997)
reported a desiccation event would occur for Owens Lake at 15 ka.
Because Owens Lake is an upstream lake of Searles Lake, and
Searles Lake received most of its inflow from the overflow of
Owens Lake, Searles Lake could also desiccate at 15 ka. This is
supported by the field evidence that the Searles Lake was at its
low water level with an elevation at 510 m. Therefore, the
simulation on lake extent in Owens Valley at 15 ka is to find
a combination of precipitation and temperature that can create
a dry Owens Lake and Searles Lake. After multiple runs,
a combination with temperature 1.8 ∘C cooler and
precipitation 20 % less than modern climate condition could
produce a dry Owens Lake and Searles Lake (Fig. 7e). The results
from this simulation also indicated that a significant decline in
the level of Mono Lake. This is consistent with possible hiatuses
in cores from the Mono Lake basin (Newton, 1991) and major
declines in the levels of Mono Lake and Lake Lahontan (Benson
et al., 1998, 1996).
Simulation on lakes at 12 ka
The δ18O data from Core OL84B drilled in Owens Lake
indicated the lowest values of δ18O at 12 ka for
the last 15 ka (Fig. 2) (Benson et al., 1997). This represents
highest ratio of overflow to inflow into Owens Lake, which implied
that Searles Lake probably also received its highest inflow for
the last 15 ka. The field evidence indicated that the lake level
of Searles Lake started to increase at 12 ka and reach its
highest level at 11 ka (Fig. 2) (Smith and Street-Perrott,
1983). It can be expected that the overflow from Searles Lake
might finally reach its highest, and the largest lake might be
formed in Panamint Valley in the last 18 ka. Based on multiple
runs, a combination of temperature 4.5 ∘C cooler and 1.8
times precipitation of modern climate conditions could reproduce
a lake system with the highest lake level in Panamint Valley
(Fig. 7d) since the LGM. However, the lake level of Panamint Lake
was still not high enough for overflow.
Simulation on lakes at 9 ka
δ18O values of Owens Lake at 9 ka are around
27 ‰ indicating Owens Lake at its hydrological closure
(Benson et al., 1996). A dry event was recognized based on the
presence of prismatic cracking that suggests the existence of
a soil formed during subaerial exposure of lake sediments at about
9 ka (Benson et al., 1997). In meantime, Searles Lake was at its
lowest water level since the LGM (Smith and Street-Perrott,
1983). A combination of temperature 0.5 ∘C warmer and 1.2
times precipitation of modern climate conditions could reproduce
a lake system in Owens Valley at 9 ka (Fig. 7c).
Simulation on lakes at 6 ka
Owens Lake was at its second desiccation event at 6.1–4.3 ka
(Benson et al., 1997), and Searles Lake was at its water level
about same as today (Smith and Street-Perrott, 1983), thus the
lakes in Owens Valley at 6 ka might have shallower water stand
than them under modern climatic conditions. Based on this
evidence, a combination of temperature 1.2 ∘C warmer and
0.9 times precipitation of modern climate conditions could produce
a lake system in Owens Valley at 6 ka (Fig. 7b).
Sensitivity analysis
Sensitivity analysis was only performed for the simulation of lakes
in Owens Valley at the LGM (18 ka). Based on the proxy data at the
LGM in Table 1, a combination of the lowest temperature
(7.5 ∘C cooler than modern temperature) and the maximum
precipitation (2.40 times modern precipitation) was used to prepare
the input data for simulation on lake extent. The results from
simulation based on this combination indicate that all basins
including Death Valley in Owens River system are full of water,
which is not case in the last 18 ka. Another combination of the
highest temperature (3.0 ∘C cooler than modern temperature)
and the lowest precipitation (1.2 times modern precipitation) was
used to simulate lake extent in Owens Valley. The results from this
simulation indicate that Searles Lake is not full and no lake was
formed in Panamint Valley. The results from these two extreme cases
of the combination are against to geological evidences. First,
Panamint Lake did not overflow, and there is no full lake formed in
Death Valley in the last 18 ka (Smith and Street-Perrott, 1983);
second, Mono Lake was separated from the Owens River system to the
south by a high-altitude sill in the late Wisconsin (Benson and
Thompson, 1987); third, Searles Lake was full and overflowed into
Panamint Valley where a small lake was formed at the LGM (Smith and
Street-Perrott, 1983). Therefore, the coupled catchment–lake model
developed in this study is very sensitive to a change of both
temperature and precipitation, and it can be easily used to infer
paleoclimatic conditions based on past lake extent.
Discussion
The coupled catchment–lake model developed in this study has
several advantages for paleoclimatic interpretation of paleolake
extent. By using physically based models to find the climatic
conditions that could produce a particular lake extent at specific
time, a quantitative estimate on temperature and precipitation
against field evidences was obtained. This approach allows direct
consideration of the effects of changes in both precipitation and
temperature, as well as numerous climate variables including cloud
cover, solar radiation, and wind speed. Simulated lake surface
areas and elevations of lake surface for major lakes in Owens
Valley in the last 18 ka are listed in Table 2. The combinations
of temperature (Fig. 8a) and precipitation (Fig. 8b) that could
produce observed lake extent in Owens Valley in the last 18 ka
were obtained from the simulations above. We also plotted
reconstructed temperature and precipitation based on pollen in
Owens Lake core OL84B in Fig. 8. Simulated combination of
temperature and precipitation at the LGM (18 ka) is
5.5 ∘C cooler and 1.25 times of modern climate
conditions. This result is close to a proxy data (Merrill and
Péwé, 1977) in Table 1, but different from other proxy
data in Table 1. This could be a result that the proxy-based
estimates of LGM climate are representative of the area in which
the fossil assemblage or glacial feature was studied. A mixed
conifer forest from Kings Canyon that includes red fir, western
juniper, incense cedar, sugar pine, ponderosa pine, California
nutmeg, and single-needle pinon pine (Cole, 1983) indicated
a colder than today with near-modern precipitation levels in the
southern Sierra Nevada at 18 ka. The 9.2 m hiatus found in
Owens Lake core OL84B was dated at 15.5 to 13.5 ka (Benson
et al., 1997). Our simulation indicated lake extent at 15 ka is
the smallest in the last 18 ka. The combination of precipitation
and temperature produced this smallest lake extent is the lowest
precipitation in the last 18 ka, and about 1.8 ∘C cooler
than modern temperature. The lake extent at 12 ka in Owens Valley
is largest in the last 18 ka. Simulated combination of
precipitation and temperature at 12 ka is 4.5 ∘C cooler
and 1.8 times of modern climate conditions. This result is
consistent with the pollen data from the core OL84B
(Fig. 8a, b) (Mensing, 2001). The pollen data indicated that
a mean annual precipitation of 308–370 mm and a >80 % increase in effective moisture, and 4 to 5 ∘C
cooler than the present mean temperature in Owens Valley start
from 13.5 ka. A very wet climate in the western Great Basin at
this time is also supported by the δ18O data from
Owens Lake (Benson et al., 1996) and Mono Lake (Benson et al.,
1998), and ages of tufa from Searles Lake (Lin et al., 1998) and
Lake Lahontan (Benson, 1993) that indicate high lake stands from
14 to 13.5 ka. However, the climate in Sierra Nevada shifted
from cool, wet conditions to possibly a more seasonal climate with
cool, wet winters and warmer summers (Mensing, 2001; Smith and
Anderson, 1992), but still wetter and cooler than today
(Spaulding, 1983) after 12 ka. Simulated lake extent in Owens
Valley at 9 ka indicated a significantly decrease in lake extent
from 12 ka. The last deglaciation was interrupted by a worldwide
cooling event, the Younger Dryas (YD) from 11 to 10 ka. Studies
from western North America have identified late-glacial climatic
oscillations roughly synchronous with the YD interval (Stuiver
et al., 1995). The pollen data from the core OL84B indicates
a series of abrupt climatic oscillations between 10.8 to 9.5 ka,
but is not sufficient to clearly define the YD for direct
comparison with other sites (Mensing, 2001). Summer insolation
reached a maximum between 9 to 8 ka, resulting in higher summer
temperatures and probably increased seasonality (Grigg and
Whitlock, 1998). Low lake levels and the increased dominance of
desert shrubs indicate the beginning of warm, dry Holocene
conditions. The results from our simulations indicate
0.5 ∘C warmer and 1.2 times of modern climate conditions
could produce observed lake extent at 9 ka, which is generally
agreement with high isolation and increased desert shrubs. Second
hiatus found in the core OL84B indicates that Owens Lake was
probably dry at 6 ka (Benson et al., 1997), and quantitative
analysis of the pollen record from Sierra Nevada suggests
temperatures 1.4–2.1 ∘C warmer than today (Adam and
West, 1983). The lake level of Searles Lake was also low at 6 ka
(Smith and Street-Perrott, 1983). Simulated lake extent with
a combination of temperature 1.2 ∘C warmer and 0.9 times
precipitation of modern climate conditions is consistent with
geologic evidences.
The climate in Owens Valley after 6 ka is probably similar to
modern climate conditions. However, a slightly increase in
precipitation and decrease in temperature could happen, because
the historic lake level of Owens Lake is higher than it at
6 ka. An increased frequency of modern extreme storm events in
Mojave River watershed in late Holocene was concluded based on
lake deposits in the Silver Lake playa, CA (Enzel et al.,
1989). The relatively high lake level of Searles Lake from 5 to
3 ka could be a result of an increased frequency of modern
extreme storm events and summer monsoon circulation (Bush, 2001).
Conclusions and future work
The coupled catchment–lake model developed in this study is capable
of accurately simulating lake extent as a function of modern
climate and paleoclimate. The purpose for this model is used to
quantitatively estimate paleoclimate, especially annual
precipitation and temperature against field evidence in
a catchement-lake watershed hydrologic system. The simulations on
lake extent at 18, 15, 12, 9, 6 ka, and modern climate conditions
are very compatible with observed or derived data. The annual
precipitation and temperature of Owens Valley for these times are
generally in agreement with the proxy data that are derived from
Owens Valley and the places near Owens Valley. The accuracy of our
quantitative estimates in paleotemperature and paleoprecipitation
in Owens Valley is completely dependent on the accuracy of the
field observations, especially the elevation of paleo-shorelines
and their chronology for the lakes. Therefore, numeric values for
the temperature and precipitation at 18, 15, 12, 9, and 6 ka are
only effective for the geologic evidences used this study.
However, these numeric values of paleotemperature and
paleoprecipitation can be adjusted based on new geologic
evidence. The two advantages of the coupled catchment–lake model
are: (1) based on the proxy data, the possible range of
temperature/precipitation combinations that could produce
a particular paleolake extent can be easily obtained by narrowing
a large number of possible paleoclimatic combinations, (2) the
model developed in this study is a physically based model that
requires a minimum of site-specific parameters (Hostetler and
Giorgi, 1993), thus it can be applied in any lakes if input
parameters are available.
Simulations performed in this study did not consider that the
seasonal distribution of precipitation and temperature in the last
18 ka might be different from modern climatic conditions. The
reason for this is that we seldom have information on the seasonal
distribution of precipitation and temperature in the past.
However, the lake levels of the Mediterranean region were
significantly affected by the seasonal distribution of temperature
(Prentice et al., 1992). Therefore, it is very important to
consider the seasonal distribution of paleoclimate into the
simulation on the paleolake extent in Owens Valley with
accumulation of more data on seasonal distribution of
paleoprecipitation and paleotemperature. Besides, an initially dry
lake was assumed for the simulation, which is not true for most
situations. Thus, the simulation could be improved by running the
model continuously a high lake level at 18 ka to a historic lake
level. A time-series of temperature and precipitation from the
continuous simulation is more useful than the discrete results in
this study.
Acknowledgements
This project was funded by the Department of Energy under Sponsored
Project (DE-RP08-OONV13813). The authors wish to acknowledge S. W. Hostetler for providing us the source code of his lake
model. Many thanks to K. M. Menking for her assistance on running the
Hostetler's lake model. The third author was funded by the support
of the Sulo and Aileen Maki Endowment. The authors acknowledge that
the views herein are that of the authors only and does not necessary
reflect the review of the funding agencies and the organizations
they are affiliated to.
References
Adam, D. P. and West, G. J.: Temperature and precipitation estimates through
the last glacial cycle from Clear Lake, California, pollen data, Science,
219, 168–170, 1983.
Anchukaitis, K. J., D'Arrigo, R. D., Andreu-Hayles, L., Frank, D.,
Verstege, A., Curtis, A., Buckley, B. M., Jacoby, G. C., and Cook, E. R.:
Tree-ring-reconstructed summer temperatures from Northwestern North America
during the last nine centuries, J. Climate, 26, 3001–3012, 2013.
Angel, J. R. and Kunkel, K. E.: The response of Great Lakes water levels to
future climate scenarios with an emphasis on Lake Michigan-Huron, J. Great
Lakes Res., 36, 51–58, 2010.
Bacon, S. N., Burke, R. M., Pezzopane, S. K., and Jayko, A. S.: Last glacial
maximum and Holocene lake levels of Owens Lake, eastern California, USA,
Quaternary Sci. Rev., 25, 1264–1282, 2006.
Bartlein, P. J., Anderson, K. H., Anderson, P., Edwards, M., Mock, C.,
Thompson, R. S., Webb, R. S., Webb III, T., and Whitlock, C.: Paleoclimate
simulations for North America over the past 21 000 years: features
of the simulated climate and comparisons with paleoenvironmental data,
Quaternary Sci. Rev., 17, 549–585, 1998.
Benson, L. V.: Paleoclimatic significance of lake-level fluctuations in the
Lahontan Basin, Quaternary Res., 16, 390–403, 1981.
Benson, L. V.: Factors affecting 14C ages of lacustrine carbonates: timing and
duration of the last highstand lake in the Lahontan Basin, Quaternary Res.,
39, 163–174, 1993.
Benson, L. V. and Paillet, F.: The use of total lake-surface area as an
indicator of climatic change: examples from the Lahontan Basin, Quaternary
Res., 32, 262–275, 1989.
Benson, L. V. and Paillet, F.: HIBAL: a hydrologic-isotopic-balance model for
application to paleolake systems, Quaternary Sci. Rev., 21, 1521–1539, 2002.
Benson, L. V. and Thompson, R.: Lake-level variation in the Lahontan Basin for
the past 50 000 years, Quaternary Res., 28, 69–85, 1987.
Benson, L. V., Burdett, J. W., Kashgarian, M., Lund, S., Phillips, F., and
Rye, R. O.: Climatic and hydrologic oscillations in the Owens Lake Basin and
adjacent Sierra Nevada, California, Science, 746–749, 1996.
Benson, L. V., Burdett, J., Lund, S., Kashgarian, M., and Mensing, S.: Nearly
synchronous climate change in the Northern Hemisphere during the last glacial
termination, Nature, 388, 263–265, 1997.
Benson, L. V.: The sensitivity of evaporation rate to climate change: results
of an energy-balance approach, US Department of the Interior, Geological
Survey, Denver, Colorado, USA, 1986.
Benson, L. V., Currey, D. R., Dorn, R. I., Lajoie, K. R., Oviatt, C. G.,
Robinson, S. W., Smith, G. I., and Stine, S.: Chronology of expansion and
contraction of four Great Basin lake systems during the past
35 000 years, Palaeogeogr. Palaeocl., 78, 241–286, 1990.
Benson, L. V., Lund, S. P., Burdett, J. W., Kashgarian, M., Rose, T. P.,
Smoot, J. P., and Schwartz, M.: Correlation of late-Pleistocene lake-level
oscillations in Mono Lake, California, with North Atlantic climate events,
Quaternary Res., 49, 1–10, 1998.
Betancourt, J. L.: Late quaternary biogeography of the Colorado Plateau, in:
Packrat Middens: The Last 40000 years of Biotic Change, edited by:
Betancourt, J. L., Van Devender, T. R., and Martin, P. S., The University of
Arizona Press, Tucson, 259–292, 1990.
Bischoff, J. L. and Cummins, K.: Wisconsin glaciation of the Sierra Nevada
(79 000–15 000 yr BP) as recorded by rock flour in sediments of
Owens Lake, California, Quaternary Res., 55, 14–24, 2001.
Bischoff, J. L., Menking, K. M., Fitts, J. P., and Fitzpatrick, J. A.:
Climatic oscillations 10 000–155 000 yr BP at Owens Lake,
California reflected in glacial rock flour abundance and lake salinity in
core OL-92, Quaternary Res., 48, 313–325, 1997.
Bush, A. B. G.: Pacific sea surface temperature forcing dominates orbital
forcing of the early Holocene monsoon, Quaternary Res., 55, 25–32, 2001.
COHMAP-Members: Climatic changes of the last 18000 years:
observations and model simulations, Science, 241, 1043–1052, 1988.
Cole, K. L.: Late Pleistocene vegetation of Kings Canyon, Sierra Nevada,
California, Quaternary Res., 19, 117–129, 1983.
Cole, K. L.: Reconstruction of past desert vegetation along the Colorado
River using packrat middens, Palaeogeogr. Palaeocl., 76, 349–366, 1990.
Dohrenwend, J. C.: Nivation landforms in the western Great Basin and their
paleoclimatic significance, Quaternary Res., 22, 275–288, 1984.
Enzel, Y., Cayan, D. R., Anderson, R. Y., and Wells, S. G.: Atmospheric
circulation during Holocene lake stands in the Mojave Desert: evidence of
regional climate change, Nature, 341, 44–47, 1989.
Forester, R. M.: An ostracode record of Holocene climate change from Owens Lake,
California, in: Impacts of Climate Change on Landscapes of the Eastern Sierra
Nevada and Western Great Basin, edited by: Jayko, A. S. and Miller, C. I.,
US Geol. Surv. Open-File Rept 01-202, US Geological Survey, 7–8, 2000
Ghile, Y., Moody, P., and Brown, C.: Paleo-reconstructed net basin supply
scenarios and their effect on lake levels in the upper great lakes, Climatic
Change, 127, 305–319, 2014.
Grigg, L. D. and Whitlock, C.: Late-glacial vegetation and climate change in
western Oregon, Quaternary Res., 49, 287–298, 1998.Groppelli, B., Bocchiola, D., and Rosso, R.: Spatial downscaling of
precipitation from GCMs for climate change projections using random cascades:
a case study in Italy, Water Resour. Res., 47, W03519, 10.1029/2010WR009437, 2011.
Halley, E.: A Short Account of the Cause of the Saltness of the Ocean, and of
the Several Lakes That Emit no Rivers; With a Proposal, by Help Thereof,
to Discover the Age of the World, Produced before the Royal-Society by Edmund
Halley, RS Secr, Philos. Trans., 29, 296–300, 1714.
Hostetler, S.: Modeling the late-Pleistocene paleohydrology of the Great
Basin, in: Proceedings of the US Geological Survey global change research
forum, edited by: Kelmelis, J. and Snow, K., US Geological Survey
Circular, Herndon, Virginia, 1–90, 1991.
Hostetler, S. and Bartlein, P.: Simulation of lake evaporation with
application to modeling lake level variations of Harney-Malheur Lake, Oregon,
Water Resour. Res., 26, 2603–2612, 1990.
Hostetler, S. and Benson, L. V.: Paleoclimatic implications of the high stand
of Lake Lahontan derived from models of evaporation and lake level, Clim.
Dynam., 4, 207–217, 1990.
Hostetler, S. and Giorgi, F.: Use of output from high-resolution atmospheric
models in landscape-scale hydrologic models: an assessment, Water Resour.
Res., 29, 1685–1695, 1993.
Hostetler, S., Bates, G., and Giorgi, F.: Interactive coupling of a lake
thermal model with a regional climate model, J. Geophys. Res., 98, 5045–5057, 1993.
Hostetler, S., Giorgi, F., Bates, G., and Bartlein, P.: Lake-atmosphere
feedbacks associated with paleolakes Bonneville and Lahontan, Science, 263, 665–668, 1994.
Jannik, N. O., Phillips, F. M., Smith, G. I., and Elmore, D.: A 36Cl
chronology of lacustrine sedimentation in the Pleistocene Owens River system,
Geol. Soc. Am. Bull., 103, 1146–1159, 1991.
Jiang, P., Gautam, M. R., Zhu, J., and Yu, Z.: How well do the GCMs/RCMs
capture the multi-scale temporal variability of precipitation in the
Southwestern United States?, J. Hydrol., 479, 75–85, 2013.
Jones, R., McMahon, T., and Bowler, J.: Modelling historical lake levels and
recent climate change at three closed lakes, Western Victoria, Australia
(c. 1840–1990), J. Hydrol., 246, 159–180, 2001.
Kutzbach, J. E.: Estimates of past climate at Paleolake Chad, North Africa,
based on a hydrological and energy-balance model, Quaternary Res., 14, 210–223, 1980.
Kutzbach, J. E.: Model simulations of the climatic patterns during the
deglaciation of North America, in: North America and Adjacent Oceans During
the Last Deglaciation, edited by: Ruddiman, W. F. and Wright, J. H. E.,
Geological Society of America, Boulder, Colorado, 425–446, 1987.
Kutzbach, J. E. and Guetter, P. J.: The influence of changing orbital
parameters and surface boundary conditions on climate simulations for the
past 18 000 years, J. Atmos. Sci., 43, 1726–1759, 1986.
Leffler, P. M. and Cochran, G. F.: Utilizing the historical climate record to
construct an analogue of past (wetter) climatic regimes in the southwest
Great Basin, in: Proceedings of the Topical Meeting on Nuclear Waste
Isolation in the Unsaturated Zone, American Nuclear Society, Inc., La Grange
Park, Illinois, 73–77, 1989.
Li, H.-C., Bischoff, J. L., Ku, T.-L., Lund, S. P., and Stott, L. D.: Climate
variability in East-Central California during the past 1000 years
reflected by high-resolution geochemical and isotopic records from Owens Lake
sediments, Quaternary Res., 54, 189–197, 2000.
Lin, J. C., Broecker, W. S., Hemming, S. R., Hajdas, I., Anderson, R. F.,
Smith, G. I., Kelley, M., and Bonani, G.: A reassessment of U-Th and 14C
ages for late-glacial high-frequency hydrological events at Searles Lake,
California, Quaternary Res., 49, 11–23, 1998.
Lowenstein, T. K., Li, J., Brown, C., Roberts, S. M., Ku, T.-L., Luo, S., and
Yang, W.: 200 ky paleoclimate record from Death Valley salt core, Geology,
27, 3–6, 1999.
MacIntyre, S., Flynn, K. M., Jellison, R., and Romero, J. R.: Boundary mixing
and nutrient fluxes in Mono Lake, California, Limnol. Oceanogr., 44, 512–529, 1999.
Magny, M.: Holocene climate variability as reflected by mid-European
lake-level fluctuations and its probable impact on prehistoric human
settlements, Quatern. Int., 113, 65–79, 2004.
Marcott, S. A., Shakun, J. D., Clark, P. U., and Mix, A. C.: A reconstruction
of regional and global temperature for the past 11 300 years,
Science, 339, 1198–1201, 2013.
Menking, K. M., Bischoff, J. L., Fitzpatrick, J. A., Burdette, J. W., and
Rye, R. O.: Climatic/hydrologic oscillations since 155,000 yr BP at
Owens Lake, California, reflected in abundance and stable isotope composition
of sediment carbonate, Quaternary Res., 48, 58–68, 1997.
Mensing, S. A.: Late-glacial and early Holocene vegetation and climate change
near Owens Lake, eastern California, Quaternary Res., 55, 57–65, 2001.
Merrill, R. K. and Péwé, T. L.: Late Cenozoic Geology of the White
Mountains, Arizona, 1, Ariz. Bur. Geol. and Miner. Technol., University of Arizona, USA, 1977.
Mifflin, M. D. and Wheat, M. M.: Pluvial Lakes and Estimated Pluvial Climates
of Nevada, Mackay School of Mines, University of Nevada, Reno, USA, 1979.
Mono-Basin-Environmental-Impact-Report: Final environmental impact report for
the review of Mono Basin water rights of the city of Los Angeles, California
State Water Resources Control Board, Sacramento, 1993.
Newton, M.: Holocene stratigraphy and magnetostratigraphy of Owens and Mono
Lakes, eastern California, PhD thesis, University of Southern California,
University of Southern California, USA, 1991.
Orndorff, R. L.: Development of a surface hydrologic model and its
application to modern and last glacial conditions within the Great Basin,
Kent State University, Ohio, USA, 1994.
Phillips, F. M.: Environmental tracers for water movement in desert soils of
the American Southwest, Soil. Sci. Soc. Am. J., 58, 15–24, 1994.
Plummer, M. A. and Phillips, F. M.: A 2-D numerical model of snow/ice energy
balance and ice flow for paleoclimatic interpretation of glacial geomorphic
features, Quaternary Sci. Rev., 22, 1389–1406, 2003.
Prentice, I. C., Guiot, J., and Harrison, S. P.: Mediterranean vegetation,
lake levels and palaeoclimate at the Last Glacial Maximum, Nature, 360, 658–660, 1992.
Smith, G. I.: Continental paleoclimatic records and their significance, in:
Quaternary Nonglacial Geology: Conterminous US, edited by: Morrison, R. B.,
The Geological Society of America, Boulder, Colorado, USA, 35–41, 1991.
Smith, G. I. and Bischoff, J. L.: Core OL-92 from Owens Lake: project
rationale, geologic setting, drilling procedures, and summary, Geol. S. Am.
S., 317, 1–8, 1997.
Smith, G. I. and Street-Perrott, F. A.: Pluvial lakes of the western United
States, in: Late-quaternary environments of the United States, Univeristy of
Minnesota Press, Minneapolis, Minesota, 190–212, 1983.
Smith, S. J. and Anderson, R. S.: Late Wisconsin paleoecologic record from
Swamp Lake, Yosemite National Park, California, Quaternary Res., 38, 91–102, 1992.
Spaulding, W. G.: Vegetation and climates of the last 45,000 years in
the vicinity of the Nevada Test Site, south-central Nevada, US Geological
Survey, Washington, D.C., 1983.
Spaulding, W. G. and Graumlich, L. J.: The last pluvial climatic episodes in
the deserts of southwestern North America, Nature, 320, 441–444, 1986.
Stamm, J. F.: Modeling local paleoclimates and validation in the southwest
United States, Oak Ridge Inst. for Science and Education, Kent State University, TN, USA, 1992.
Steinman, B. A., Abbott, M. B., Mann, M. E., Stansell, N. D., and
Finney, B. P.: 1500 year quantitative reconstruction of winter
precipitation in the Pacific Northwest, P. Natl. Acad. Sci. USA, 109, 11619–11623, 2012.
Street-Perrott, F. A. and Harrison, S. P.: Lake levels and climate
reconstruction, in: Paleoclimate Analysis and Modeling, edited by:
Hecht, A. D., John Wiley and Sons, New York, 291–340, 1985.Stuiver, M., Grootes, P. M., and Braziunas, T. F.: The GISP2 δ18O
climate record of the past 16,500 years and the role of the sun,
ocean, and volcanoes, Quaternary Res., 44, 341–354, 1995.
Thompson, R. S., Whitlock, C., Bartlein, P., Harrison, S., and Spaulding, W.:
Climatic changes in the western United States since 18,000 yr BP, in:
Global Climates Since the Last Glacial Maximum, edited by:
Wright, H. E., Kutzbach, J. E., Webb III, T.,
Ruddiman, W. F., Street-Perrott, F. A., and Bartlein, P. J., Univerisity of
Minnesota Press, Minneapolis, USA, 468–513, 1993.
Thompson, R. S., Anderson, K. H., and Bartlein, P. J.: Quantitative
paleoclimatic reconstructions from late Pleistocene plant macrofossils of the
Yucca Mountain region, US Geological Survey Open-File Report 99-338,
US Geological Survey, Denver, Colorado, USA, 1999.
Vassiljev, J.: Simulating the Palaeorecord of Northern European Lakes Using
a Coupled Lake-Catchment Model, Lund University, Lund, Sweden, 1997.
Viau, A., Ladd, M., and Gajewski, K.: The climate of North America during the
past 2000 years reconstructed from pollen data, Global Planet.
Change, 84, 75–83, 2012.
Proxy data in the southwest US in
the last 18 ka.
TimeTemperature (∘C)Precipitation (%)SourceLate Wisconsin-6.25 annual -3.0 annual -5.5 annual -6.5 annual -6.0 summer -6.7 annual +1.0 winter, -1.0 summer -3.0 Jan, -3.0 Jul -5.5 to –4.9 annual+37cm winter +68% annual +27.5 % annual +65% winter, -45 % summer Summer precipitation 10 % of annual +32% annual +57% winter, +56 % summer +19mm Jan, -31 mm Jul 1.4 to 1.7× model annualDohrenwend (1984) Mifflin and Wheat (1979) Merrill and Pewe (1977) Spaulding (1985) Betancourt (1990) Cole (1990) Leffler and Cochran (1989) Spaulding and Graumlich (1986) Thompson et al. (1999)18 ka 20.5 to 18 ka14 to 11.5 ka-3.29 annual -3.17 Jan, -3.01 Jul -7.5 annual -6.7 annual-0.29 mmday-1 annual +0.25 mmday-1 Jan, -0.84 mmday-1 Jul 2.40×2.58×Thompson et al. (1993) Thompson et al. (1999) Thompson et al. (1999)12 ka-2.52 annual -3.01 Jan, -0.63 Jul-0.18 mmday-1 annual -0.27 mmday-1 Jan, -0.15 mmday-1 JulThompson et al. (1993)9 ka+0.43 annual -0.09 Jan, +2.15 Jul+0.30 mmday-1 annual +0.80 mmday-1 Jan, -0.27 mmday-1 JulThompson et al. (1993)6 ka+0.69 annual +0.30 Jan, +0.68 Jul-0.03 mmday-1 annual -0.16 mmday-1 Jan, +0.07 mmday-1 JulThompson et al. (1993)
Simulated lake extent and elevation of lake
levels in Owens Valley in the last 18 ka.
AgeMono Lake Owens Lake Searles Lake Panamint Lake (ka)Elevation (m)Area (km2)Elevation (m)Area (km2)Elevation (m)Area (km2)Elevation (m)Area (km2)1820404611145692688949340144151952258107030475231031221206891145692688949350349919782701100302525252325946194917210759649043203001958227109730251522532030
Location map of Owens River system
(modified from Smith and Bischoff,
1997).
δ18O for core OL84B
(solid line with dot) (Benson et al., 1997),
δ18O from GISP2 (solid line) (Stuiver et al., 1995), and
elevation of lake surface for Searles Lake (Smith and
Street-Perrott, 1983) in the last
20 ka.
A comparison of observed runoff and simulated
runoff in Mono Lake drainage basin.
A plot of simulated and measured temperature
profile (MacIntyre et al., 1999) in
Mono Lake (upper panel); simulated evaporation and
observed evaporation for Mono Lake (lower panel).
Inputs of cloud cover and solar radiation for
the simulations at 18, 15, 12, 9, and 6 ka (Bartlein et al., 1998).
Inputs of wind speed for the simulations at at
18, 15, 12, 9, and 6 ka (Kutzbach and
Guetter, 1986).
Simulated lake extents (deep blue) in the last
18 ka, and hillshade (gray scale) and stream network
(light blue) derived from DEM data.
Simulated temperature (solid line with dots)
and estimated temperature based on pollen (solid line)
(Mensing, 2001) in Owens Lake (left panel) and simulated
precipitation (solid line with dots) and estimated precipitation based on
pollen (solid line) (Mensing, 2001) in Owens Lake
(right panel).