Introduction
One
goal of the Regional Monitoring Program (RMP) is to determine seasonal
and annual trends of variability in the chemical and biological
water quality of the San Francisco Estuary. The United States Geological
Survey (USGS) maintains
a program of monthly water-quality measurements to supplement RMP
monitoring done three times each year. This element of the RMP is
designed to describe the changing spatial patterns of water-quality
variability from the lower Sacramento River to the southern limit
of the South Bay. Five water-quality parameters are measured as
descriptors of the chemical-biological status of the Estuary, and
as indicators of the key processes that control the concentration,
chemical form, or biological availability of toxic contaminants.
A second
objective of the RMP is to determine long-term trends in the concentrations
of trace elements and organic contaminants in the San Francisco
Estuary. This objective poses a difficult challenge because estuaries
have large natural variability that acts as noise around any signals
of water quality improvement or degradation over time. Progress
toward this second objective will require innovative approaches
for characterizing the natural variability of biological and chemical
conditions in the Estuary, and then separating these natural fluctuations
from any trends of change. In this chapter we summarize results
of the USGS measurement program for 1997, and use these results
to illustrate the general patterns of water-quality change caused
by natural processes of variability in the San Francisco Bay-Delta
ecosystem. Identification of these patterns, and their underlying
mechanisms, is an important step in the determination of trends
of change in trace contaminants.
The
Measurement Program
Design
This
element of the RMP characterizes water quality in the deep channel
of the Bay-Delta system. It includes measurements at a series of
fixed stations spaced every 36 km, from Rio Vista (lower Sacramento
River, Figure 3.46), through Suisun
Bay, Carquinez Strait, San Pablo Bay, the Central Bay, and the South
Bay to the mouth of Coyote Creek. Vertical profiles are taken at
each station, so this measurement program provides two-dimensional
(longitudinal-vertical) descriptions of spatial structure. Sampling
along the 145 km transect requires 1215 hours, so measurements
are taken at varying phases of the semidiurnal tide cycle. Although
it is logistically difficult to synchronize sampling to a constant
tidal phase, we minimized the effects of intratidal variability
by sampling near the periods of monthly minimum tidal energy when
possible. Therefore, this sampling program is biased toward neap
tide conditions, and it is confounded by intratidal variability
during the course of sampling. Sampling is confined to the central
channel, so it does not measure directly the transverse component
of water-quality variability across the broad shoals. However, sampling
along the axial transect does describe variability along the estuarine
salinity gradient, and it provides an integrative picture of all
the processes occurring upstream, in adjacent marshes and lateral
shoals, due to point source discharges, and within the local water
column (Jassby et al., 1997). Sampling was done once each month
along the entire North Bay-South Bay transect. More frequent sampling
was done in the South Bay to follow the dynamic water-quality changes
caused by the spring phytoplankton bloom (Cloern, 1996), such as
depletion of dissolved metals (Luoma et al., 1998). Sampling dates
for 1997 are listed in Table 3.3.
Water
Quality Parameters
This
element of the RMP measures five water-quality parameters, each
reflecting a different set of processes that cause estuarine variability.
Salinity measures the relative proportion of freshwater and seawater,
and the salinity distribution reflects the changing importance of
river flow as a source of dissolved materials carried into the Bay-Delta
from the Estuary's watersheds. Water temperature is an independent
indicator of mixing, and an important control on biological transformations
of reactive trace substances. The concentration of suspended particles
(as total suspended solids, TSS) changes in response to the alternating
tidal cycles of sediment deposition and resuspension, episodic wind-driven
resuspension, and riverine inputs of new sediments during periods
of high flow. These processes are relevant to the RMP because many
trace substances are reactive with particle surfaces, so the pathways
of transport, retention, and incorporation of these contaminants
into the food web are influenced by the transport of sediments.
For example, RMP data show strong correlations between suspended
solids concentration and the concentration of total mercury in San
Francisco Bay waters (Schoellhamer, 1997). This USGS measurement
program provides information about the large-scale changes in the
spatial distribution of TSS associated with river inputs. Variability
at shorter time scales is characterized by the continuous measurements
of TSS by moored instruments at fixed locations (Schoellhamer, 1996).
The
phytoplankton community represents the single largest component
of living biomass in San Francisco Bay, and we measure the distribution
of chlorophyll a as an index of this biomass. Unlike salinity and
TSS, chlorophyll a is a nonconservative quantity that changes in
response to processes of production and consumption, as well as
inputs and transports. The production of phytoplankton biomass involves
the uptake of inorganic forms of elements (including carbon, nitrogen,
phosphorus, and some trace metals) dissolved in the water, and then
transformation of these inorganic raw materials into new organic
matter packaged as algal cells. The partitioning of reactive elements
between dissolved and particulate forms can be highly influenced
by the phytoplankton community in San Francisco Bay (Cloern, 1996;
Luoma et al., 1998), and chlorophyll a concentration is a simple
indicator of the potential for these biotransformations.
We
measure dissolved oxygen (DO) concentration as an indicator of the
net trophic status of the Estuary. If the oxygen content of water
is undersaturated (less than that at equilibrium with atmospheric
oxygen), then oxygen is being consumed by the biota faster than
it is produced by photosynthesis (community respiration exceeds
primary production). Supersaturation of oxygen occurs when the photosynthetic
production of oxygen within the Estuary is faster than all the processes
of consumption. Therefore, DO concentration is an index of the balance
between production and oxygen consumption, a key descriptor of the
status of the ecosystem. Episodes of DO supersaturation occur during
periods of rapid phytoplankton primary production when the inorganic
forms of elements (carbon, nitrogen, phosphorus, silicon, cadmium,
etc.) are rapidly removed from solution and converted into particulate
form. Therefore, DO provides a useful indicator of the rate of phytoplankton-mediated
transformations of reactive elements and compounds in the water
column. Whereas chlorophyll a measures the abundance (or biomass)
of the phytoplankton, DO measures the activity level of the phytoplankton
community.
Methods
Data
for this RMP element were collected with an instrument package that
includes sensors for measuring: sampling depth, conductivity, temperature,
salinity (calculated from conductivity and temperature), TSS (optical
backscatter sensor), chlorophyll a (fluorometer), and DO (oxygen
electrode). The instrument package is lowered through the water
column, taking measurements about every 4 cm. Here, we report only
the measurements made in the upper meter of the water column, calculated
as the mean of all measurements made between 0.5 m and 1.5 m. The
complete data set, including measurements made at all depths, is
available as a data report (Baylosis et al., 1998) or over the Internet
at the USGS website that archives and displays results of the water-quality
program at http://sfbay.wr.usgs.gov/access/wqdata/.
The
conductivity and temperature sensors were calibrated by Sea-Bird
Electronics prior to the first sampling in January 1997. The optical
backscatter sensor, fluorometer, and oxygen electrodes were calibrated
each sampling date with analyses of water samples. Surface samples
were collected by pump, and bottom samples were collected with a
Niskin bottle. Aliquots were analyzed for: TSS (gravimetric method
of Hager, 1993); chlorophyll a (spectrophotometric method of Lorenzen,
1967, using the equations of Riemann, 1978); and dissolved oxygen
(automated Winkler titration, following Granéli and Granéli,
1991). Values reported here are calculated quantities based on daily
calibrations of the optical backscatter, fluorescence, and oxygen
sensors from linear regressions of measured concentrations versus
voltage output of each instrument.
1997
Results
The
Year of the Big Storm
Residents
of northern California will remember 1997 as the year of the Big
Storm, a three-day period of record precipitation and river flow
centered on New Year's Day. The New Year's storm was preceded by
high precipitation in December 1996, when runoff was about three
times the December mean. Roos (1997) describes the event:
Record streamflow, especially the 3-day flood volumes, were produced
in many of the major rivers. The sheer volume of runoff exceeded
the flood control capacity of Don Pedro and Millerton reservoirs
in the central Sierra foothills, sending large amounts of excess
water down the Tuolumne and San Joaquin rivers. Most other foothill
reservoirs made releases that brought rivers downstream up to maximum
flood design capacity. Major flooding occurred along the uncontrolled
Cosumnes River southeast of Sacramento, on the Tuolumne River near
Modesto, and the San Joaquin River near Fresno.
Roos
estimates that runoff during January 1997 was about 390 percent
of the average, and probably a record for the month. He ends his
description:
The New Year's Day storm is one for the record books. December 1996
and January 1997 are the two wettest consecutive months on record
for the northern Sierra 8-station average, with a combined total
of 47.6 inches of precipitation.
January
1997 was followed by very dry periods in February and March, so
the 1997 record of Delta outflow into the San Francisco Estuary
was characterized by exceptional outflows in January, receding outflows
in February and March, and then persistently low outflows the remainder
of the year (Figure 3.47). This
simple, high-amplitude fluctuation in Delta outflow provides an
excellent example for illustrating the principle that the water
quality and biological communities of estuaries respond quickly
and strongly to changes in river flow. Some of these changes have
direct relevance to the RMP and its objectives of determining trends
of change.
Effects
of the Big Storm on Water Quality of the Estuary
Water
quality of San Francisco Bay, like all urbanized estuaries, is influenced
by a combination of both natural forces and human activities. Estuaries
are ecosystems where freshwater and seawater mix, and the proportions
of fresh and seawater change in response to fluctuations in river
flow. The freshwater-seawater mix within the Estuary is measured
as the salinity distribution, and the changing salinity distribution
reflects large changes in water quality and biological communities
caused by change in the relative proportions of fresh and seawater.
The New Year's Flood of 1997 provides a large natural experiment
to show how water quality and biological communities and processes
of the San Francisco Estuary change in response to an extreme event
of high river flow.
Results
from the USGS salinity measurements are depicted in Figure
3.47, which shows the changing spatial patterns of salinity
as gray-scale shadings. The upper panel shows the daily record of
the Delta Outflow Index (DOI). The bottom panel shows the patterns
of salinity variability as shaded contour images, where shading
intensity is proportional to salt content of the surface waters.
The vertical axis represents the longitudinal transect from the
lower Sacramento River (top of image, at kilometer 92), to the Central
Bay at Angel Island (kilometer 0), and then to the lower South Bay
at the mouth of Coyote Creek (kilometer -52.7). The horizontal axis
represents monthly variability during 1997. This, and following
images, are based on interpolations of the 533 surface measurements
made during the twenty USGS sampling cruises in 1997. Here, dark
shading indicates high salinity and light shading indicates low
salinity. The thick solid line shows the changing position of the
surface salinity of 2 psuan index of the location of the interface
between freshwater and brackish water in the Estuary. This image
shows the strong response of the salinity distribution to the New
Year's Flood. The first USGS sampling was done on January 13, and
sampling was stopped in Carquinez Strait because flows were so great
that the research ship could not progress further upstream against
the strong currents. The next sampling on January 28 was completed
up to Rio Vista, and it showed a remarkable salinity distribution
with freshwater in the surface layer as far downstream as San Pablo
Bay. A nearly-fresh (salinity of 1 psu) surface layer was found
in the Central Bay near Angel Island on January 28.
The
solid diamonds on Figure 3.47 show
the times and locations of USGS measurements during the three periods
of RMP water sampling. The first RMP sampling of 1997 occurred at
the time of the most extreme change in the salinity distribution,
about three weeks after the January5 peak in Delta outflow. The
mean DOI during the first RMP water sampling was 6,420 m3/s,
nearly double the previous high DOI during an RMP sampling (Table
3.4). The mean surface salinity along the entire USGS sampling
transect was only 3.4 psu on January 28; during this period, the
surface waters of the Estuary were, on average, about 90% freshwater.
This period when the freshwater fraction in the Estuary was very
high should be reflected in the distributions of trace contaminants
measured by other RMP elements. For example, the pattern in Figure
3.47 suggests that contaminants with local sources were flushed
from the Estuary during this period of unusually high river flow
and low salinity.
The
patterns in Figure 3.47 show only
the results of the surface measurements, and these do not reflect
the salinity of deep waters in the Estuary. Freshwater has lower
density than seawater, so it has a tendency to float on top of seawater
unless turbulent mixing from wind and tides is strong enough to
vertically mix the surface freshwater and deep seawater. During
periods of high river flow, such as January and February 1997, the
water in the Estuary channels becomes layered, or stratified, with
low-salinity water carried seaward on top of high-salinity deep
water. For example, on January 28 when the surface salinity at Angel
Island (USGS station 17) was only 2 psu (nearly freshwater), the
bottom waters at that location had salinity of 21.1 psu (mostly
seawater). Therefore, events of high river flow change both the
horizontal and the vertical distributions of salinity and other
water-quality constituents.
The
exceptional peak of freshwater inflow in early January (DOI = 14,840
m3/s) was followed by a second peak (DOI =7,514 m3/s)
in late January, and then a spring of receding inflow and a summer/autumn
of persistent low inflow. The gray-scale shadings of Figure
3.47 show how the surface salinity of the Estuary responded
to the periods of receding river flow, with the entire Estuary becoming
progressively saltier after the January storms ended. This progressive
increase in the seawater fraction was reflected in the changing
position of the line where surface salinity = 2 psu (Figure
3.47). By the time of the second RMP water sampling in April
1997, this position was displaced about 75 km upstream and positioned
in Suisun Bay. Between the RMP first and second samplings, the mean
surface salinity of the Estuary increased from 3.4 to 16.7 psu (Table
3.4), and we expect large changes in the concentrations of some
trace substances between these two sampling periods. The third RMP
water sampling, in late July/early August, occurred after three
months of sustained low flow, when the mean surface salinity along
the USGS transect was 18.2 psu (Table
3.4).
The
salinity distribution of Figure 3.47
shows how dissolved components of water quality can change in response
to seasonal fluctuations in river flow. Other aspects of water quality
change with river flow, including the distribution and concentration
of suspended matter. In Figure 3.48
we show the changing patterns of total suspended solids (TSS) concentration
as a measure of the abundance of particles suspended in surface
waters of the Estuary. Most suspended particles in the San Francisco
Estuary are mineral (clay) particles, and the gray-scale shadings
in Figure 3.48 show the large pulse
input of suspended sediments to the northern Estuary following the
1997 New Year's Flood. The highest TSS concentration measured in
surface waters by the USGS monitoring was 227 mg/L. During the first
RMP water sampling, TSS concentrations were uniformly high in the
northern Estuary, and a plume of low salinity-high turbidity water
was observed into the Central Bay. The mean TSS concentration in
surface waters along the USGS transect was 91mg/L, the highest mean
concentration in the five years of RMP sampling (Table
3.4).
As
the river flow receded in spring, the riverine sources of sediments
to the Estuary were reduced, and the mean TSS concentration fell
to 38mg/L (Table 3.4). Since many
trace contaminants are reactive with particle surfaces, we expect
large changes in the concentrations and distributions of the particle-reactive
substances between the first and second RMP samplings of 1997. We
also expect that these seasonal trends continued into the time of
the July/August sampling when TSS concentrations were low throughout
the Estuary (Figure 3.48). This
image also shows localized regions of high TSS concentration in
the lower South Bay, especially below the Dumbarton Bridge, reflecting
inputs of sediments from the urban watershed of the South Bay during
the January storms.
The
Spring Phytoplankton Bloom
As
the RMP evolves and matures, we continue to learn new lessons about
the key processes which control the transport, biogeochemical cycling,
and ecosystem effects of trace contaminants in the San Francisco
Estuary. In recent years we have learned that phytoplankton uptake
and assimilation of trace metals (such as cadmium, nickel, zinc,
and selenium) is a key biological process which transforms these
elements from dissolved to particulate phases (Luoma et al., 1998).
These biological transformations are a key step in the trophic transfer
of trace contaminants leading to bioaccumulation within the tissues
of upper-trophic-level consumer animals. Phytoplankton, as the dominant
primary producers in the Estuary, probably also play a key role
in the uptake and metabolism of some trace organic contaminants,
such as PCBs. The USGS water-quality element of RMP follows phytoplankton
dynamics in San Francisco Bay by measuring chlorophyll concentration
(an index of the biomass, or abundance, of the phytoplankton community)
and dissolved oxygen concentration (as an index of the rate of photosynthesis,
or productivity, of the phytoplankton community).
Chlorophyll
distributions show that phytoplankton abundance in the San Francisco
Estuary is usually low, except for periods of very high biomass
(algal blooms) which can develop in the South Bay during the months
February through April (Cloern, 1996). During 1997, the South Bay
spring bloom was characterized by four distinct episodes of enhanced
chlorophyll biomass, beginning with a small event north of the San
Mateo Bridge in late January (Figure
3.49). This was followed by three larger events of rapid chlorophyll
increase, and then decline, in late February, late March, and late
April (at the time of the second RMP water sampling). These events
of enhanced phytoplankton biomass were reflected in the dissolved
oxygen (DO) content of surface waters, and we observed simultaneous
events of oxygen supersaturation during the March and April algal
bloom events (Figure 3.49). The
patterns of DO variability (Figure
3.50) suggest that phytoplankton primary productivity was very
high in the South Bay during late March and late April, so we expect
large changes in the form or concentrations of reactive trace substances
between the RMP water samplings in January (low phytoplankton activity)
and April (high phytoplankton activity). The third RMP water sampling
occurred in July/August, after three months of persistent low phytoplankton
biomass and productivity, suggesting that the biological effect
of phytoplankton on trace substances was small then.
Results
of USGS sampling suggest that the RMP sampling in April 1997 should
provide an excellent opportunity to measure the effects of phytoplankton
uptake/metabolism on trace contaminants, because this sampling was
done during the period of highest chlorophyll concentration (maximum
34.9 mg/m3) and DO concentration (maximum 133% saturation)
among all the RMP water samplings done since 1993 (Table
3.4).
The
1997 El Niño Event
One
of the strongest El Niño events of the past century developed
in 1997, when temperatures in the eastern Pacific Ocean warmed to
over five degrees Celsius (°C) above normal. This global-scale
climatological phenomenon propagated along the west coast of South
and North America, and its effects remind us that the San Francisco
Estuary is connected to, and influenced by, changes in the adjacent
coastal ocean. The surface water temperatures of the southern and
northern regions of the Estuary (Figure
3.51) were not unusual in 1997, but the water temperatures in
the Central Bay (closest to the Pacific Ocean) were the highest
recorded since the RMP sampling began. Effects of oceanic warming,
which reflect large-scale changes in coastal oceanic circulation,
were most pronounced during the RMP water sampling of July/August
(Figure 3.51). For example, bottom
temperature at USGS station 18 in Central Bay was 17.67°C on
August 5, compared to bottom temperature of only 15.44°C in
August 1996. This temperature anomaly suggests that the character
of the Central Bay region, including its water quality and biological
communities, could have been affected by coastal changes influenced
by the 1997 El Niño event.
Although
we are far from a complete understanding of how coastal oceanographic
processes influence water quality in the San Francisco Estuary,
RMP results from recent years show that water quality does change
in response to coastal upwelling (Cloern et al., 1997) and El Niño
warming. These coastal forcings probably induce a cascade of chemical
and biological changes within the Estuary, through linkages which
are not yet fully understood. Observations during the 1997 El Niño
suggest one mechanism through which the coastal ocean might influence
estuarine water quality and toxicity. The USGS and SFEI conducted
a special study during August 1997 in response to reports of animal
mortalities and visible red tides in regions of the Central Bay
(Cole and Cohen, 1998). This study showed high abundances of the
dinoflagellate Gymnodinium sanguineum, a toxin-producing species
of phytoplankton usually found in tropical or subtropical waters.
During this red tide episode Cole and Cohen (1998) observed mortalities
and poor condition of attached invertebrate animals, such as mussels
and tunicates, in marinas around Central Bay. Numerous dead fish
(including adult striped bass and halibut) were observed in Aquatic
Park lagoon.
The
simultaneous occurrence of temperature anomalies (warm water) and
toxic red tides in Central Bay shows how events of change in the
coastal ocean can propagate into the San Francisco Estuary and cause
changes in water quality and biological communities. The unusual
events of summer 1997 remind us that reports of fish and shellfish
mortality are increasing in response to nutrient enrichment and
stimulation of harmful algal blooms in estuaries around the world.
They also remind us that San Francisco Bay harbors at least twenty
known species of toxin-producing phytoplankton (Rodgers et al.,
1996). Therefore, sources of toxicity can be produced biologically
within the Estuary, so the RMP goals of determining trends of water-quality
change should include consideration of the potential impact of toxic
blooms and the prospect that events of algal-derived toxicity could
become more frequent in San Francisco Bay, as they have in other
nutrient-rich coastal ecosystems.
Summary
In
this chapter, we use results from twenty USGS sampling cruises to
describe some key features of water-quality variability in San Francisco
Bay during 1997. The patterns of variability are displayed as shaded
images showing the annual cycle and the spatial gradients of water
quality, from the Sacramento River to the southern South Bay. The
five water-quality parameters described here were chosen as indicators
of different processes of estuarine variability, so results from
this program element can be used as a starting place for interpreting
the more complex patterns of variability in trace contaminants and
their effects. We use results from 1997 to illustrate some general
lessons of estuarine variability that are clearly evident in the
easily-measured quantities: salinity, temperature, TSS, chlorophyll,
and DO. These same lessons apply to trace substances, and we hope
these lessons will be useful guides for identifying the patterns
and causes of variability in trace substances, which are also influenced
by the large events of 1997: the New Year's Flood of 1997; sustained
periods of high phytoplankton production in the South Bay during
spring; and the changes induced by the 1997 El Niño event.
Acknowledgments
This
program of water-quality measurement in San Francisco Bay is a partnership
supported by the United States Geological Survey (Water Resources
Division National Research Program, Toxic Substances Hydrology Program,
San Francisco Bay Ecosystem Program) and the Regional Monitoring
Program for Trace Substances. We thank the coordinators of these
programs and the members of the San Francisco Estuary Institute
and the San Francisco Regional Water Quality Control Board for their
continuing support of this partnership.
References
Baylosis,
J.I., B.E. Cole, and J.E. Cloern. 1998. Studies of the San Francisco
Bay, California, Estuarine Ecosystem. Regional Monitoring Program
Results. 1997. U.S. Geological Survey Open-File Report 98-168, 194p.
Cloern,
J.E. 1996. Phytoplankton bloom dynamics in coastal ecosystems: A
review with some general lessons from sustained investigation of
San Francisco Bay, California. Reviews of Geophysics 34:127168.
Cloern,
J.E., B.E. Cole, J.L. Edmunds, and J.I. Baylosis. 1997. Water quality
variability in San Francisco Bay, some general lessons from 1996
sampling. In 1996 Annual Report: San Francisco Estuary Regional
Monitoring Program for Trace Substances, San Francisco Estuary Institute,
Oakland, CA. pp. 4660.
Cole,
B.E. and A. Cohen. 1998. Red tide in Berkeley Marina raises concern
for toxic blooms in Central Bay. Newsletter, Interagency Ecological
Program for the Sacramento-San Joaquin Estuary, v. 11(1):1113.
Granéli,
W. and E. Granéli. 1991. Automated potentiometric determination
of dissolved oxygen. Marine Biology 108:341348.
Hager,
S.W. 1993. Dissolved nutrient and suspended particulate matter data
for the San Francisco Bay Estuary, California. October 1988 through
September 1991. U.S. Geological Survey Open-File Report 93-57.
Jassby,
A.D., B.E. Cole, and J.E. Cloern. 1997. The design of sampling transects
for characterizing water quality in estuaries. Estuarine, Coastal
and Shelf Science 45:285302.
Lorenzen,
C.J. 1967. Determination of chlorophyll and phaeopigments: spectrophotometric
equations. Limnology and Oceanography 12:343346.
Luoma,
S.N., A. van Geen, B.G. Lee, and J.E. Cloern. 1998. Metal uptake
by phytoplankton during a bloom in South San Francisco Bay: Implications
for metal cycling in estuaries. Limnology and Oceanography 43:10071016.
Riemann,
B. 1978. Carotenoid interference in spectrophotometric determination
of chlorophyll degradation products from natural populations of
phytoplankton. Limnology and Oceanography 23:10591066.
Rodgers,
K.M., D.L. Garrison, and J.E. Cloern. 1996. Toxic phytoplankton
in San Francisco Bay. In 1995 Annual Report: San Francisco Estuary
Regional Monitoring Program for Trace Substances. San Francisco
Estuary Institute, Oakland, CA. pp. 5666.
Roos,
M. 1997. Water year 1996 in review. Newsletter, Interagency Ecological
Program for the Sacramento-San Joaquin Estuary, v. 10:912.
Schoellhamer,
D.H. 1996. Factors affecting suspended-solids concentrations in
South San Francisco Bay, California. Journal of Geophysical Research
101, No. C5:12,08712,095.
Schoellhamer,
D.H. 1997. Time series of suspended-solids concentration, salinity,
temperature, and total mercury concentration in San Francisco Bay
during water year 1996. In 1996 Annual Report: San Francisco Estuary
Regional Monitoring Program for Trace Substances. San Francisco
Estuary Institute, Oakland, CA. pp. 6577.
|
RMP:
RMP
Home | Program Information
| Publications | RMP
Data
