OS36B10

1
OS36B-10
Hydrographic changes in the coastal northern Gulf
of Alaska during Northeast Pacific GLOBEC
(1997-2004)
Thomas C. Royer, Chester E. Grosch, Thomas J.
Weingartner and Seth Danielson
Center for Coastal Physical Oceanography, Old
Dominion University, Norfolk, VA, 23529, U.S.A.
Institute of Marine Science, University of
Alaska, P.O. Box 757220, Fairbanks, AK,
99775-7220, U.S.A.
E-mailroyer_at_ccpo.odu.edu
Interannual Changes in Hydrography Upper Layer
Salinity Anomaly (0-100 m) The upper layer
salinity anomaly (seasonal signal removed) (Fig.
2c) begins with a freshening in 1997-8 followed
by abnormally high salinities in late 1998 at the
midshelf locations and a relatively strong high
salinity near to the coast in early 1999. Higher
than normal salinities then follow until about
2003 when below normal salinities are found
across the shelf through 2004. Lower Layer
Salinity Anomaly (100-250 m) The salinity
anomalies here (Fig. 2d) are very similar to the
upper layer in timing, position and
magnitude. Upper Layer Temperature Anomaly (0-100
m) Upper layer temperatures (Fig.3c) were
elevated in 97-98 especially offshore near to
the shelf break. High temperatures continued in
1998 especially at the shelf break and offshore.
For 1999 - 2002, declining temperatures were
present across the shelf followed by a rather
abrupt warming in late 2002 into 2003 across the
shelf. In late 2003, upper temperatures returned
to normal. Lower Layer Temperature Anomaly
(100-250 m) Similar to Fig. 3c, temperatures
near the shelf break in 97 were above normal
(Fig. 3d). However, the nearshore deep water was
relatively warm throughout 1998 especially within
40 km of the coast. The period between early
1999 and to early 2002 was one of near normal
temperatures in the lower layer, followed by a
sharp cooling in mid-late 2002 over the inner 2/3
of the shelf. A very brief warming over the
inner half of the shelf occurred in Spring 2003
followed by a return to normal conditions
thorough the end of the record. Possible
Hydrographic Forcing Functions Coastal Freshwater
Discharge The monthly coastal freshwater
discharge has been determined over the Southeast
and Southcoast regions of Alaska (after Royer,
1982) and recent, enhanced glacial ablation has
been added (Arendt et al., 2002) (Fig. 4).
Relative to the 74 year discharge record, the
Feb. 99 discharge is one of the lowest
especially since ablation is now included. In
2002 there are consistently low discharges from
January through April. High discharges occurred
in October (4th highest since 1931) and
throughout the fall of 2002 (September (5th
highest), October and November). Alongshore Wind
Stress The monthly mean mesoscale alongshore
stress at 59.5 N, 149.5 W (about GAK3) from NSCAT
and QuikSCAT data (Fig. 5), contain 5 unusual
events. The first (Dec.97-Feb.98) consisted of
a positive wind stress (toward the east
upwelling) followed by a strong westward stress
(downwelling). A similar event (positive
followed by negative) took place from Nov.03 to
Feb.04. The other three events were all
negative (downwelling) and took place
Feb.-Apr00, Nov.-Dec.00 and Sept.03-Jan.04
(the most intense one).
Figure 1. GLOBEC LTOP Seward Line
Figure 2 Salinity
Figure 3 Temperature
Abstract
a
a
The hydrographic conditions (temperature and
salinity versus depth) during the GLOBEC sampling
(October 1997-2004) across the shelf of the
northern Gulf of Alaska reveal significant
seasonal and interannual temperature and salinity
changes during this period. The Northeast Pacific
Long Term Observation Program (NEP LTOP)
conducted approximately 6 cruises per year during
the 7 year field program. The interdisciplinary
sampling included depth profiles of salinity,
temperature, photosynthetically available
radiation (PAR), and light transmission across
the Seward Line beginning at the mouth of
Resurrection Bay and terminating off the shelf in
the deep Gulf of Alaska. The initial sampling
period took place shortly after an El
Nino-Southern Oscillation (ENSO) event and a
rapid temperature elevation was observed in early
1998 especially at depth (75-150 m) and centered
approximately 80 km offshore. The water
temperature at depth declined overall until 2003
when it rose to its highest level. As compared
with the 34 year time series of temperature and
salinity versus depth to 250 m at the mouth of
Resurrection Bay, Alaska (60 N, 149 W) (GAK 1),
at 2 C above normal, it was the highest anomaly
yet observed. However, it follows a general
trend of increasing water temperatures throughout
the water column. The salinity anomaly across
the shelf in the 75-150 m depth range was
relatively fresh in 1997-8 followed by increasing
salinities to their maxima in 2000-2001 and then
a return to more normal conditions. The abrupt
warming in 2003 was accompanied by a freshening
of 0.9, exceeding the 0.5 freshening that
accompanied the 1997-8 ENSO event. Generally, the
water temperatures have increased throughout the
water column over the Seward Line during the LTOP
period. During the same period, the salinity of
the uppermost 100 m of the water column has
decreased while the lower 150 m has a trend of
increasing salinity. This is consistent with an
increase in the stratification and baroclinic
circulation over this shelf which likely affects
production of the lower tropic levels.
b
b
The Data Hydrographic sampling on 40 cruises
from October 1997 to December 2004.Northeast
Pacific GLOBEC Long Term Observing Program (LTOP)
(Fig. 1) (Weingartner, et al. 2002). The water
column was divided into two layers representing a
directly forced surface layer with the mixed
layer (0-100 m) and a lower, more isolated, layer
100 250 m (or the bottom when lt than 250 m).
c
c
Seasonal Hydrography Upper Layer Salinity (0-100
m) The upper layer salinity (Fig.2a) (St. 1 at
coast and 13 in deep water), has a narrow band of
relatively freshwater ( less than 32) extending
about 55-60 km offshore seasonally with the
freshest water and maximum extent in the fall.
The greatest nearshore salinity contrast occurred
in fall 2002 with the least in fall 2000 and
2001. Offshore, the upper layer salinity has a
seasonal maximum (gt32.5) in the late winter/early
spring. The highest salinities offshore occurred
in early in both 2000 and 2001. No cross shelf
temporal delays are apparent at least on a
temporal scale of the cruises (one month or
more). Lower Layer Salinity (100-250 m) The lower
layer salinity (Fig.2b) has a coherent pattern of
low salinity (lt32.8) extending about 110 km
offshore, usually most apparent in the late
winter-early spring. The lower salinities occur
later in the year than the upper layers.
Offshore, there is an intrusion of high salinity
water either coincident with or immediately
following the nearshore low salinity intrusion.
No consistent cross shelf time delays are
noted. Upper Layer Temperature (0-100 m) There
is seasonal warming across the shelf without a
significant temporal delay (Fig. 3a). The
maximum heating coincides with the late summer
and the majority of this seasonal heating is
solar. The warmest year is 1998, following the
1997-8 ENSO event with the coolest winter
occurring in 2002. Lower Layer Temperature
(100-250 m) The highest temperatures in the
lower layer (Fig. 3b) occur seasonally after the
high temperatures in the upper layer with the
maximum temperatures are found at depth of in the
midst of winter (Dec.-Mar.). This warm water is
coincident with the low salinities seen in Fig.
2a and is likely a result of coastal downwelling
and is found primarily at a similar distance
offshore as the low salinity lower layer water
(about 110 km).

• Discussion
• The negative upper layer salinity anomaly (Fig.
  2c) is a response to cumulative high rates of
  freshwater discharge (Fig. 4) throughout Fall
  97. In early 99, relatively low freshwater
  discharge caused increased salinity out to about
  GAK 4. One of the highest rates of discharge
  followed in October but was modulated by the
  relatively quiescent, possibly upwelling winds
  (Fig.5). Moderate downwelling winds returned
  Feb.-Apr. 00. However the salinity increases on
  the outer shelf were the result of a mesoscale
  eddy that impinged on the Seward Line in May 00
  (Okkonen, et al, 2001). Late in 00 abnormally
  high salinities returned to the inner shelf out
  of phase with the higher than normal freshwater
  discharge that peaked in October 00. Possibly
  this is a consequence of the eddy that propagated
  across the shelf at this time. It could be that
  the eddy entered the offshore waters in Spring
  00 causing the high salinities at the shelf
  break. Relatively low freshwater flows Jan.-Apr.
  02 were not clearly reflected as an increased
  the upper layer salinity anomaly. This could be a
  consequence of the lack of wind strength for this
  period. The Fall 02 high discharge produces
  the strongest negative salinity anomaly at the
  coast out to GAK5 for late 02 and early 03.
  Upper and lower layer (Fig. 2d) salinity
  anomalies have a very similar temporal and
  spatial patterns, suggesting that alongshore
  advection plays an important role in their
  formation.
• The temperature anomalies (Figs. 3c, 3d) have
  similar vertical coherences (maybe slightly less
  than salinity) but have more uniform cross-shelf
  spatial structures. The temperature anomalies
  are higher than normal in 97-8 especially at the
  shelf break. Since this was time of the 97-8
  ENSO event, it could be the result of the
  propagation of an alongshore Kelvin wave. A
  double signal, the first arriving in early 98
  and the second about 9-10 months later coincides
  with the double SOI signal, accounting for the
  Kelvin wave propagation (Royer, 2005). Below
  normal temperatures in 02, are due to weak
  downwelling. Downwelling here forces warmer,
  upper layer water downward. Subsequently, the
  relatively strong downwelling in late 03-early
  04 created a pool of relatively warm water
  across the shelf.
• During the GLOBEC period at GAK1, the salinity
  anomaly in July 00 in the upper layer reached
  the second highest value (0.61) in the 35 year
  hydrographic record. The salinity in the lower
  layer was the highest ever (0.59) in March 00.
  
• In Feb. 03, the upper layer temperature anomaly
  was highest (2.28 C) in the 35 year record with
  the 2nd highest anomaly (1.41 C) in Feb. 98
  (ENSO conditions). These are consistent with the
  lower layer temperature anomalies of 1.90 C
  (highest) and 1.35 C (2nd highest) at the same
  times. The detection of the warmest waters at
  these depths at GAK 1 since 1970 could be a
  consequence of a finer temporal sampling though
  it could also be a consequence of long term
  increases in the water temperature throughout
  this water column (Royer, 2005).
• Conclusions
• Coastal freshwater discharge influences the
  salinity anomalies across the shelf in the
  northern Gulf of Alaska.
• Mesoscale eddies occasionally can play an
  important role in exchanging waters between the
  deep gulf and shelf.
• El Nino (ENSO) events and local winds influence
  temperature anomalies.
• Cross shelf ENSO propagation has been observed
  for the first time in the Gulf of Alaska. ENSO
  signals are largest at the shelf break and
  arrive there first.
• Highest anomalies of salinity and temperature
  since 70 were recorded at GAK1 during the GLOBEC
  sampling (97-04).

d
d

References Arendt, A. A, et al., 2002. Science,
297 382-386. Okkonen, S.R, et al., 2003.
J.Geophy. Res.108, doi10.1029/2002JC001342.
Royer, T. C. 1982. J. Geophys. Res.
872,017-2,021. Royer, T.C., 2005. Deep Sea Res.
II, 52, 267-288. Weingartner, T., et al. 2002,
Oceanography Magazine 1530-35.
Figure 4. Coastal Freshwater Discharge
Figure 5 Alongshore Wind Stress
(speed3) (positive eastward, offshore Ekman
transport)
Acknowledgements The satellite wind data products
were provided by Isaac Schroeder. Funding was
provided by the National Science Foundation and
travel support for the Ocean Sciences meeting was
provided by the U.S. Arctic Research Commission.