Journal of Marine Systems xxx (2014) xxx–xxx

MARSYS-02658; No of Pages 9

Contents lists available at ScienceDirect

Journal of Marine Systems

j ourna l homepage: www.e lsev ie r .com/ locate / jmarsys

The impact of wind mixing on the variation of bottom dissolved oxygenoff the Changjiang Estuary during summer

Xiaobo Ni a, Daji Huang a,c,⁎, Dingyong Zeng a, Tao Zhang a, Hongliang Li b, Jianfang Chen a,b,c

a State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, State Oceanic Administration, Hangzhou 310012, Chinab Laboratory of State Oceanic Administration for Marine Ecosystem and Biogeochemistry, Second Institute of Oceanography, State Oceanic Administration, Hangzhou 310012, Chinac Ocean College, Zhejiang University, Hangzhou 310058, China

⁎ Corresponding author at: State Key Laboratory ofDynamics, Second Institute of Oceanography, State OceanRoad, 310012 Hangzhou, China. Tel.: +86 571 81963088;

E-mail address: djhuang@sio.org.cn (D. Huang).

http://dx.doi.org/10.1016/j.jmarsys.2014.11.0100924-7963/© 2014 The Authors. Published by Elsevier B.V

Please cite this article as: Ni, X., et al., The imsummer, J. Mar. Syst. (2014), http://dx.doi.o

a b s t r a c t

a r t i c l e i n f o

Article history:Received 30 June 2014Received in revised form 15 September 2014Accepted 24 November 2014Available online xxxx

Keywords:Dissolved oxygenHypoxiaWind mixingStratificationChangjiang Estuary

Hypoxia off the Changjiang Estuary has been frequently reported using short time duration field data. However,its evolutionwas unknown because of a lack of long-termdata and its associated dominant factor. A 104-day longdatasetwas collectedwith a bottommounted systemoff the Changjiang Estuary in summer 2009. Themonitoredparameters were bottom dissolved oxygen (DO), temperature, pressure and current. Two hypoxia events wereidentified, showing that hypoxia was severe and lasted for more than a half month. The first event appearedon July 18 and lasted 17 days. During this hypoxia period, the minimum DO was down to 0.17 mg/L, whichbroke the historical record. The second hypoxia event appeared onAugust 30 and lasted 18dayswith aminimumDO of 1.29 mg/L. The variation of bottom DO was closely related to that of stratification. The monitored datashowed that almost every increase/decrease of DOwas associated with a weakening/enhancing of stratification,which were recorded as many as 12 times during the monitoring period. Wind mixing modulated or broke thestratification, which affected the variation of bottom DO and hypoxia events. Using a lagged correlation analysis,the stratification andwindmixingwere significantly correlatedwith a coefficient of determination r2=0.72, andstratification lagged wind by 35 h. The bottom DO and wind mixing were significantly correlated with a coeffi-cient of determination r2 = 0.65, and DO lagged wind by 33 h. The formation periods of two hypoxia eventsestimated from monitored data were 20 and 15 days, which were much shorter than that from on-boardexperiments. Strongwindmixing played a dual role on hypoxia. It could relieve hypoxia conditions by supplyingDO throughmixing. It accelerated the formation of hypoxia afterward as a result of the enhanced phytoplanktonbloom induced by wind mixing and high organic decomposition rates consuming more DO.

© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-SA license(http://creativecommons.org/licenses/by-nc-sa/3.0/).

1. Introduction

Dissolved oxygen (DO) threatens marine animals when its concen-tration is lower than 2 mg/L or 62.5 μmol/L (Boesch and Rabalais, 1991;Diaz and Rosenberg, 1995), which is defined as hypoxia (Montgomery,1969). Additionally, hypoxia could transform the community structureof zoo benthos (Ning et al., 2011) or damage the coastal ecosystem(Diaz and Rosenberg, 2008). Increasing coastal hypoxia was reportedin the last few decades around theworld because of enhanced anthropo-genic nutrient loading. The seasonal hypoxia off the Changjiang Estuaryis one of the most focused targets and has been a hot spot for interdisci-plinary studies in recent years (Zhang et al., 2010).

Although the seasonal hypoxia off the Changjiang Estuary has beenfrequently reported using data observed by ship based field surveys,

Satellite Ocean Environmentic Administration, 36 Baochubeifax: +86 571 88839374.

. This is an open access article under

pact of wind mixing on the varg/10.1016/j.jmarsys.2014.11.

which are generally performed annually, seasonally and monthly, wewere still unclear how it forms, maintains and disappears because of alack of continuous observations (Wang et al., 2012; Zhu et al., 2011).The hypoxia area expanded fast and became one of the world's largestcoastal hypoxia (Li et al., 2002; Wang, 2009; Wang et al., 2013; Zhouet al., 2009) when it was first reported in the 1950s (Gu, 1980). Theformation and evolution of hypoxia were closely related to complicatedphysical processes from the interaction of the Changjiang DilutedWater(CDW), Taiwan Warm Current (TWC) and Kuroshio water masses, inaddition to chemical and biological processes and bottom topography(Li et al., 2011; Ning et al., 2011; Wang, 2009; Wei et al., 2007; Zhouet al., 2010). The stratification of the water column and oxygen con-sumption through organic decomposition were two essential factorsfor the formation of hypoxia off the Changjiang Estuary (Wang et al.,2012). Typhoons, monsoons, cold air and tides are important factorsto mediate hypoxia through weakening or breaking stratification,resulting in bottom oxygen supplement from the surface (Chen et al.,2007; Wei et al., 2007). However, there was no direct evidence todemonstrate this hypothesis because of a lack of continuous data; the

the CC BY-NC-SA license (http://creativecommons.org/licenses/by-nc-sa/3.0/).

riation of bottom dissolved oxygen off the Changjiang Estuary during010

2 X. Ni et al. / Journal of Marine Systems xxx (2014) xxx–xxx

key obstacle was that it was hard to make long-term observationsbecause of heavy fishery activities off the Changjiang Estuary.

In this paper, we analyze the relationship between bottom DOand sea surface wind (SSW) with a 104-day time-series of bottom DO,temperature, depth and current at a station in the northern ChangjiangEstuary. The evolution process of formation, maintenance and disap-pearance of hypoxia are shown alongwith the variation of stratification.The role of wind mixing and Chlorophyll-a (Chl-a) on the evolution ofhypoxia is also discussed.

2. Data and methods

The long-term time-series data of interdisciplinary parameters atmooring station B in the northern Changjiang Estuary during summer2009 were collected by a Trawl Resistant Bottom Mount (TRBM) (Niet al., 2014). Mooring station B is shown in Fig. 1 and is located at31.5°N/123°E. It was deployed on June 6 and recovered on September17, lasting 104 days. The bottom DO was measured with an AanderaaOxygen Optode 3835 sensor. The sea bottom temperature (SBT), salini-ty and pressure were measured with a RBR Conductivity-Temperature-Depth (CTD). The current profile was measured using a RDI-WHSAcoustic Doppler Current Profiler (ADCP). All of the instruments weremounted on the TRBM, with a sampling interval of 30 min. The DOand CTD sensors were calibrated before deployment and after recovery.They were reclaimed once for maintenance and calibration during acruise on August 4.

During the observation period at mooring, an interdisciplinary sur-vey was performed from August 17 to 18 in section C at 31°N during2009 (Fig. 1). During this survey, temperature and salinity profile datawere collected at nine stations using a SBE917plus CTD.

The Level 4 sea surface temperature (SST) data produced by NOAA'sNational Oceanographic Data Center were used in this study. TheSST data had a spatial resolution of 5 km × 5 km with a time intervalof one day. The SST data that covered the study area from June 6to September 17 in 2009 were downloaded from their website athttp://ghrsst.nodc.noaa.gov. The SST at station B was calculated usingthe spatial average of the four nearby SST points.

Sea surface wind (SSW) data produced by NOAA's National ClimaticData Center were used in this study. The 6-h data had a spatial resolu-tion of a 0.25-degree grid. The SSW data covering the study area from

Fig. 1. The location of mooring station B an

Please cite this article as: Ni, X., et al., The impact of wind mixing on the vasummer, J. Mar. Syst. (2014), http://dx.doi.org/10.1016/j.jmarsys.2014.11

June 1 to September 17 in 2009 were acquired from their website athttp://www.ncdc.noaa.gov. The east and north components of theSSW at station B were calculated using the spatial average of the fournearby SSW points.

Sea surface Chl-a concentration data derived usingMODIS-Terra and-Aqua from June 1st to 13th, 2009 were used in this study. The Level 3mapped data with a spatial resolution of 4 km × 4 km were acquiredfrom the U.S. National Aeronautics and Space Administration (NASA)website at http://oceancolor.gsfc.nasa.gov.

The main acronyms used in this paper are listed in Table 1.

3. Results

3.1. Temporal variation of bottom DO

The data atmooring station Bwasmonitored from June 6 to Septem-ber 17, which almost covered the whole summer in 2009.

Fig. 2a shows that the bottom DO fluctuated with multiple timescales ranging from a tidal period of 12-h to a 20-day period. Thetidal scale variation of DO was caused by the tidal current advectionof DO. In this paper, we focused on the synoptic and seasonal variationof DO. Therefore, the daily mean DO (Fig. 3) was used in the followingtext, which removed the tidal variation. The most apparent features inthe temporal variation of the daily mean DO were that there were twodecreasing processes of DO. The first was from early June to the endof July, and the second occurred from mid-August to mid-September.Additionally, there was one increasing process of DO from earlyAugust to mid-August after Typhoon Morakot. The other noticeablecharacteristicswere the 12DO increasing-decreasing events. The timingof the DO increasing events were on June 6th, June 18th, June 24th,July 6th, July 14th, July 22nd, July 28th, August 2nd, August 19th, August30th, September 6th, and September 15th. In general, the longerthe time of an increasing event, the higher its corresponding DO peak.In particular, the DO increased from 0.17 mg/L to 5.86 mg/L in a13-day increasing event starting on August 2nd. The decreasing rateof DO following each increasing event was proportional to that ofthe increasing rate.

Two episodes of hypoxia were observed during summer 2009. Thefirst hypoxia formed on July 18. Two mild increasing events occurredwithin this hypoxia period, with a minimum DO of 0.17 mg/L. This

d section C off the Changjiang Estuary.

riation of bottom dissolved oxygen off the Changjiang Estuary during.010

Table 1List of acronyms.

Acronym Expanded form

CDW Changjiang Diluted WaterChl-a Chlorophyll-aDO Dissolved oxygenDM Difference of h/W3and TMh/W3 Wind mixing parameterSBT Sea bottom temperatureSST Sea surface temperatureSSW Sea surface windTDSB Temperature difference between SST and SBTTM Difference of surface energy input and tidal mixingTWC Taiwan Warm Current

3X. Ni et al. / Journal of Marine Systems xxx (2014) xxx–xxx

hypoxia lasted 17 days until August 3. It ended with the strongest DOincreasing event because of mixing from typhoon Morakot.

The second hypoxia event formed on August 30 and lasted 18 days.The monitoring ended September 17, except for August 31 because of aslight increasing event. During this hypoxia event, the DO concentra-tions were above 1.29 mg/L and varied much less than those in thefirst event. At the end of this event, the DO began to increase, suggestingthat a new increasing event was going to begin.

3.2. Stratification and its impact on bottom DO

The study areawas affected by CDWand TWCduring summer (Zhouet al., 2009, 2010). Strong thermocline and halocline were formed fromhigh-temperature/low-salinity CDW on top of the low-temperature/high-salinity TWC water. They are shown in the vertical distribution oftemperature, salinity and density in section C during summer 2009

01/06 11/06 21/06 01/07 11/07 21/0701

2

3

4

5

6

DO

(m

g/L

)

01/06 11/06 21/06 01/07 11/07 21/0715

20

25

30

Tem

per

atu

re (

oC

)

01/06 11/06 21/06 01/07 11/07 21/070

5

10

15

20

Sea

Su

rfac

e W

ind

(m

/s)

01/06 11/06 21/06 01/07 11/07 21/070

100

200

300

400

Win

d D

irec

tio

n (

o)

Fig. 2. The temporal variation of bottom DO (a, mg/L), sea surface temperature/sea bottom tein 2009.

Please cite this article as: Ni, X., et al., The impact of wind mixing on the vasummer, J. Mar. Syst. (2014), http://dx.doi.org/10.1016/j.jmarsys.2014.11.

(Fig. 4). The figures indicate that the pycnocline was primary followedby halocline and thermocline, implying that stratification was primarilycontrolled by salinity and temperature, which agrees with Chen et al.(2007). Because the bottom mounted system only monitored thebottom parameters, such as bottom temperature and salinity, and wedid not have temperature and salinity profiles, and surface salinitydata. Therefore, we used remote sensing SST data and calculated thetemperature difference between SST and SBT (TDSB), which was usedas the index of stratification in this study.

Fig. 2b shows that during the observing period, SST and SBTincreased with fluctuations from early June to mid-August. SST thenbegan to fall but SBT remained unchanged to the end of monitoring.The minimum SST was 21.8 °C and the maximum SST was 28.4 °C,which appeared on June 9 and August 22, respectively. The minimumandmaximum temperatures of SBTwere 17.7 °C and 25.1 °C, which ap-peared on June 6 and August 16, respectively.

The most apparent pattern for SST and SBT was that each SBT in-creasing process was associated with a decreasing SST process, whichwas caused by the vertical mixing processes. This pattern was clearlyidentified from decreasing stratification in terms of TDSB (Fig. 5).The stratification intensified with fluctuations from June to late July,declining sharply during the first half month of August and weakeningin September. There were approximately 12 obvious decreasing pro-cesses of stratification. The 12 processes were on June 6th, June 14th,June 23rd, July 4th, July 13th, July 20th, July 28th, August 1st, August19th, August 26th, September 6th, and September 12th. The watercolumn was well mixed by the end of the 14-day decreasing processfrom August 1st to 15th.

The variation of the daily mean bottom DO was almost out of phasewith that of TDSB (Fig. 5a), which indicated the importance of thestratification on the formation of hypoxia. This correlation between

31/07 10/08 20/08 30/08 09/09

a

31/07 10/08 20/08 30/08 09/09

b

Sea Surface TemperatureSea Bottom Temperature

31/07 10/08 20/08 30/08 09/09

c

31/07 10/08 20/08 30/08 09/09

d

mperature (b,°C), sea surface wind speed (c, m/s), and wind direction (d, °) at station B

riation of bottom dissolved oxygen off the Changjiang Estuary during010

06/06 11/06 16/06 21/06 26/06 01/07 06/07 11/07 16/07 21/07 26/07 31/07 05/08 10/08 15/08 20/08 25/08 30/08 04/09 09/09 14/090

1

2

3

4

5

6D

O (

mg

/L)

0.50

-0.45

-0.12

0.06

-0.03

0.20

-0.20

-0.12

0.23

-0.34

0.18

-0.19

0.24

-0.48

-0.02

0.43

-0.26

0.75

0.30

-0.63

0.33

-0.36

-0.13

0.15

-0.14 -0.04

0.03

-0.04

0.10

Fig. 3. The temporal variation of the daily mean DO (mg/L) and its rate (mg/L/day) at station B in 2009. Daily mean DO rates are green for a negative rate and red for a positive rate.(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

4 X. Ni et al. / Journal of Marine Systems xxx (2014) xxx–xxx

bottom DO and stratification was discussed by Rabalais and Gilbert(2009). During the summer of 2009, each of the 12decreasingprocessesof stratification was associated with an increasing event of DO. Thedominant variation period of DO and stratification was 3.7 days. Basedon delayed linear correlation analyses, the DO and stratification werenegatively related with the coefficient of determination r2 = 0.95, andthe DO lagged for 8 h by stratification. However, DO variation wasnot sensitive to decreasing stratification processes during the secondhypoxia event, where the DO was relatively stable when stratificationsubstantially declined.

3.3. Wind mixing and its mediation on stratification and bottom DO

The temporal variation of SSW from June 1 to September 17, 2009 isshown in Fig. 2c and d. Wind speed changed periodically in time scalesof 3–4 days. The SSW experienced four typical stages during the sum-mer according to its speed and direction: 1) a southerly wind occurredbefore August, 2) a strong easterly wind induced by Typhoon Morakotoccurred from August 1 to August 12, 3) a southerly wind occurred

Fig. 4. The distribution of temperature (°C), salinity and density an

Please cite this article as: Ni, X., et al., The impact of wind mixing on the vasummer, J. Mar. Syst. (2014), http://dx.doi.org/10.1016/j.jmarsys.2014.11

until the end of August, and 4) a northeasterly wind occurred inSeptember. A strong wind with a speed of more than 8 m/s frequentlyoccurred, which enhances vertical mixing along with surface energyinput and tidal mixing (MacCready et al., 2009; Xuan et al., 2012).

Wind mixing in the study area was calculated using a stratificationand mixing model for shallow seas (Simpson et al., 1978). The windmixing was quantified by a parameter h/W3, which is the amount of en-ergy required to mix the water column influenced by the competitionbetween surface heat energy input and tidal mixing (Eq. (1); Xuanet al., 2012).

DM ¼ h=W3−TM

¼ h=W3− δksρs αgQ=2c− 3π=4ρkε � h=uT3

� �−1� �−1� � ð1Þ

The second term on the right side of Eq. (1) was quantified by TM toexpress the difference of surface energy input and tidal mixing. InEq. (1), h is the water depth, W is the wind speed, ks is the surfacedrag coefficient, ρs is the air density, δ is the rate of dissipated wind

omalies (σt, kg/m3) in section C from August 17 to 18, 2009.

riation of bottom dissolved oxygen off the Changjiang Estuary during.010

06/06 16/060

1

2

3

4

5

6

DO

(m

g/L

)

a

26/06 06/07 16/07 26/07 05/08 15/08 25/08 04/09 14/090

1.5

3

4.5

6

7.5

9

TD

SB

(oC

)

0

3

6

9

12

15

18

Sea

Su

rfac

e W

ind

(m

/s)

b

06/06 16/06 26/06 06/07 16/07 26/07 05/08 15/08 25/08 04/09 14/090

1.5

3

4.5

6

7.5

9

TD

SB

(oC

)

Fig. 5. The temporal variation of bottom DO (a, mg/L) with TDSB (a,°C), and sea surface wind (b, m/s) with TDSB (b, °C) at station B in 2009.

5X. Ni et al. / Journal of Marine Systems xxx (2014) xxx–xxx

energy, uT is the depth-averaged tide velocity, c and ρ are the specificheat and density of seawater, k and ε are two efficiencies of tidemixing,α is the volume expansion coefficient, and Q is the rate of heat input.All of the parameters in Eq. (1), except for h and uT, are discussed inXuan et al. (2012), which studied in the same area and season. The30-minute interval for monitoring time-series data of h and uT areshown in Fig. 6.

06/06 16/06 26/06 06/07 16/07 20

0.2

0.4

0.6

0.8

1

1.2

1.4

06/06 16/06 26/06 06/07 16/07 2

34

35

36

37

38

39

40

Wat

er D

epth

h (

m)

Vel

oci

ty u

T (

m/s

)D

epth

-ave

rag

ed T

ide

Fig. 6. The temporal variation of depth-averaged tide velocity

Please cite this article as: Ni, X., et al., The impact of wind mixing on the vasummer, J. Mar. Syst. (2014), http://dx.doi.org/10.1016/j.jmarsys.2014.11.

TheDM is defined as h/W3 minus TM. When DM is positive, i.e., h/W3

is greater than TM, the wind mixing is not strong enough to mix thewater column along with the surface energy input and tidal mixing.The water column is separately mixed at the top and bottom layers,and the mixing is stronger when the DM value was smaller. However,when DM is less than zero, i.e., h/W3 is less than or equal to TM, thewindmixing is able tomix thewater columncompletely alongwith sur-face energy input and tidal mixing. However, DM less than zero was

6/07 05/08 15/08 25/08 04/09 14/09

a

6/07 05/08 15/08 25/08 04/09 14/09

b

uT (a, m/s) and water depth h (b, m) at station B in 2009.

riation of bottom dissolved oxygen off the Changjiang Estuary during010

6 X. Ni et al. / Journal of Marine Systems xxx (2014) xxx–xxx

only a necessary condition for mixing the water well, whether a watercolumn could bemixed well or not was still dependent on the durationof wind mixing.

The results of TM and h/W3 are shown in Fig. 7. TM showed a similarfluctuation pattern of tidal current, but h/W3 varied in synoptic scalesimilar to wind speed. Generally, TM was much less in magnitude andvariation compared with that of h/W3 because the surface energyinput and tide current were relatively stationary compared with thechanging wind. Fig. 7 shows that h/W3 approached TM from time totime, indicating vertical mixing events appeared frequently. Verticalmixing may have occurred during June 7th–9th, June 12th–19th, June23th–27th, July 5th–18th, July 23th, July 28th, August 1st–11th, August18th–19th, August 28th–29th, September 6th–9th and September 12thwhen h/W3 approached TM. Especially strong vertical mixing may haveoccurred during June 8th–9th, June 27th and August 7th–11th whenh/W3 was less than TM.

Stratification was the result of the competition between windmixing, surface energy input and tidal mixing in the study area. Xuanet al. (2012) noted that vertical mixingwould be significantly enhancedwhen wind speed was greater than 8 m/s, including surface energyinput and tidal mixing factors. The relationship between the windspeed and TDSB demonstrated similar results (Fig. 5b). Although strongwind events (N8 m/s) were not always followed by decreasing pro-cesses of stratification, stratification declined obviously when h/W3

approached TM, indicating that vertical mixing event occurred, andthe DO increased accordingly (Fig. 7). Strong vertical mixing occurredon June 8th–9th, June 27th, July 14th, July 23th, July 28th, and August7th–11th when h/W3 was less than TM, suggesting the water columncould be completely mixed with enough time. The stratification de-creased and DO increased significantly during these periods. The dura-tion of their variation was proportional to duration of the negativeDM. For example, stratification was almost destroyed and DO reached

10-3

10-2

10-1

100

101

102

h/w

3 (m

-2s3 )

0

1

2

3

4

5

6

7

8

06/06 16/06 26/06 06/07 16/0710

-3

10-2

10-1

100

101

102

TM

(m

-2s3 )

10-3

10-2

10-1

100

101

102

h/w

3 (m

-2s3 )

0

1

2

3

4

5

6

DO

(m

g/L

)

06/06 16/06 26/06 06/07 16/0710

-3

10-2

10-1

100

101

102

TM

(m

-2s3 )

TD

SB

(oC

)

Fig. 7. The temporal variation of TDSB (a, °C) with h/W3 (a, m−2 s3) and TM (a, m−2 s3), dail

Please cite this article as: Ni, X., et al., The impact of wind mixing on the vasummer, J. Mar. Syst. (2014), http://dx.doi.org/10.1016/j.jmarsys.2014.11

its maximum during the longest strong mixing induced by TyphoonMorakot during August 7th–11th, which terminated the first hypoxiaevent.

Fig. 8 shows the variation of log10(DM), TDSB daily changing ratesand DO daily changing rates, when wind, stratification and DO pro-cesses occurred. However, Fig. 8 shows the delayed effects of windmixing on stratification and DO. Based on delayed correlation analyses,the coefficient of determination between TDSB and DM was r2 = 0.72for a 3.4-day period with TDSB lagged 35 h relative to DM. The co-efficient of determination between DO and DM was r2 = 0.65 for a3.1-day period with DO lagged 33 h relative to DM.

4. Discussion

4.1. The process and status of hypoxia

The bottom DO in our dataset demonstrated typical seasonal andintra-seasonal variations. Fig. 3 shows that fromearly summer, as strongstratification formed because of high runoff and enhanced solar radia-tion, bottom DO gradually declined with the effects of the inhibition ofDO vertical mixing and the decomposition of organic matter. Untilmid-summer, the DO decreased as stratification persisted and becamestronger, and hypoxia occurred on July 18th for the first time. Thehypoxia lasted 17 days until August 2 when it disappeared because ofstrong vertical mixing induced by Typhoon Morakot. The bottomwater was re-oxygenated and the DO reached its peak magnitudefrom vertical mixing. The DO repeated the process before August 2,and hypoxia reappeared rapidly on August 30 because of a higher DOdecreasing rate. It lasted 18 days to the end of the monitoring period.The formation of the second hypoxia was much shorter at 15 days(from August 15 to 30) compared with that of 38 days (from June 10to July 18) from the first event.

26/07 05/08 15/08 25/08 04/09 14/09

a

26/07 05/08 15/08 25/08 04/09 14/09

b

y averaged DO (b, mg/L) with h/W3 (b, m−2 s3) and TM (b, m−2 s3) at station B in 2009.

riation of bottom dissolved oxygen off the Changjiang Estuary during.010

06/06 16/06 26/06 06/07 16/07 26/07 05/08 15/08 25/08 04/09 14/09-4

-3

-2

-1

0

1

2

log

10(D

M)

(m- 2

s3 ) a

06/06 16/06 26/06 06/07 16/07 26/07 05/08 15/08 25/08 04/09 14/09-3

-2

-1

0

1

2b

06/06 16/06 26/06 06/07 16/07 26/07 05/08 15/08 25/08 04/09 14/09-1

-0.5

0

0.5

1

1.5c

Rat

e (o

C/d

ay)

TD

SB

Ch

ang

ing

Rat

e (m

g/L

/day

)D

O C

han

gin

g

Fig. 8. The temporal variation of log10(DM) (a, m−2 s3), TDSB changing rate (b, °C/day) and daily averaged DO changing rate (c, mg/L/day) at station B in 2009. The data less than 10−4 ofDM are not shown in the subgraph (a).

7X. Ni et al. / Journal of Marine Systems xxx (2014) xxx–xxx

However, the evolution process of DO was not a simple seasonalphenomenon, as previously thought, based on discrete seasonal ormonthly data. Wind events disturbed the process in synoptic scales.There were six increasing–decreasing events of DO during the first for-mation period of hypoxia and two increasing–decreasing events in thesecond one. On average, the DO increasing–decreasing event occurredonce a week following the synoptic variation of wind.

The status of the two episodes of hypoxia showed that hypoxia waslikely to occur in stratified water during summer and was severe andlasted longer time. During the first hypoxia period, the minimum DOwas down to 0.17mg/L, which was the lowest in recorded history com-pared with previously published data (Gu, 1980; Li et al., 2011; Wang,2009; Wang et al., 2013; Zhou et al., 2009). This serious hypoxiamay have occurred from the late 1990s when the hypoxia area beganto enlarge significantly. This was not known because of a lack of contin-uous monitoring. The second hypoxia lasted 18 days until the end ofmonitoring on September 17. The DO began to increase at the end.However, the hypoxiamay have continued for a longer time, suggestingthat the increasing event was relatively weak with an increasing rateof 0.1 mg/L/day. Hypoxia may have occurred from July to Septemberand persisted for a long period off the Changjiang Estuary, instead ofappearing in August and lasting for a short duration, as suggested byhistorical discrete observations.

4.2. Impact of wind mixing on hypoxia

The duration of hypoxia formation was much shorter than previ-ously thought. The observed DO decreasing rate (0.20 mg/L/day onaverage) was much higher than the DO consumption rate of 0.018 to0.054 mg/L/day from on board organic matter degradation

Please cite this article as: Ni, X., et al., The impact of wind mixing on the vasummer, J. Mar. Syst. (2014), http://dx.doi.org/10.1016/j.jmarsys.2014.11.

experiments (Zhu et al., 2013). Based on experimental results, the du-ration of hypoxia formation fromDO of 4.7 mg/L was 50–150 days (Zhuet al., 2013), which was much longer than the 20 days (from June 28 toJuly 18) and 15 days (from August 15 to August 30) observed from thetwo hypoxia events with similar starting DO concentrations.

Although strong wind events always raised the bottom DO, the DOdecreasing rate was always high for several days after strong windmixing processes, and then it declined significantly (Figs. 3, 9c). TheDOdecreasing rate after strongwind eventswas slightly lower or higherthan the DO increasing rate during wind process from time to time butwere the same on average. This suggests that the DO consumption ratewas approximately the same order of the DO increasing rate induced bywind mixing in several days, and the DO decreasing rate was higherwhen the wind processes were stronger. The second hypoxia formedin 15 days because of a high DO decreasing rate after the first hypoxiawas disturbed by strong wind processes.

The high DO decreasing ratewas the result of enhanced phytoplank-ton bloom induced by wind mixing, and high organic decompositionrates in high DO conditions (Zhu et al., 2013). Strong wind processescould trigger strong vertical mixing and upwelling, which suppliedlarge amounts of nutrients to upper euphotic layers from the bottom,promoting primary productivity and phytoplankton blooms (Babinet al., 2004; Herbeck et al., 2011; Hung et al., 2013; Lin et al., 2003).The high DO decreasing rate was the result of the decomposition oflarge amounts of organic matter descending from upper layers. Contin-uous satellite Chl-a data from June 1–13 showed that the Chl-a variationwas similar towind processes (Fig. 9). Chl-a concentrations increased toa peak value of 6-fold and 3.4-fold within two days after two windevents began on June 1st and June 8th, respectively. During the secondwind event, the bottom DO concentration started to decrease when

riation of bottom dissolved oxygen off the Changjiang Estuary during010

10-3

10-2

10-1

100

101

102

h/W

3 (m

-2s3 )

2.5

3

3.5

4

4.5

01/06 02/06 03/06 04/06 05/06 06/06 07/06 08/06 09/06 10/06 11/06 12/06 13/06 14/060

0.5

1

1.5

2

2.5

3

3.5

4

Ch

l-a

(mg

/m3 )

a

10-3

10-2

10-1

100

101

102

h/W

3 (m

-2s3 )

3.5

4

4.5

5

5.5

DO

(m

g/L

)

01/06 02/06 03/06 04/06 05/06 06/06 07/06 08/06 09/06 10/06 11/06 12/06 13/06 14/060

0.5

1

1.5

2

2.5

3

3.5

4

Ch

l-a

(mg

/m3 )

b

TD

SB

(oC

)

Fig. 9. The temporal variation of Chl-a (a, mg/m3) with h/W3 (a, m−2 s3) and TDSB (a, °C), Chl-a (b, mg/m3) with h/W3 (b, m−2 s3) and daily mean DO (b, mg/L) at station B in 2009.

8 X. Ni et al. / Journal of Marine Systems xxx (2014) xxx–xxx

the Chl-a concentration increased, and declined sharply when Chl-areached its peak on June 11th. Stronger phytoplankton bloom wasfound after Typhoon Marakot in the southern East China Sea to thesouth of our study area (Li et al., 2013; Tang and Sui, 2014), which in-creased the Chl-a concentrations to 10-fold to 60-fold based on fieldand satellite data. This result supported the highest DO decreasingrate after Typhoon Marakot in our data.

We could not conclude that strong wind events would relieve hyp-oxia off the Changjiang Estuary. Strong wind processes during TyphoonMorakot removed the first hypoxia event during summer 2009. How-ever, the second hypoxia event occurred 15 days later on August 30thbecause of a high DO decreasing rate induced by this wind process. Thefirst hypoxia event formed in 20 days from June 28 to July 18 with alower starting DO value. Wind mixing could relieve hypoxia, but maymake hypoxia more serious. At the end of the first hypoxia event, theDO concentration may not drop to 0.17 mg/L if there were no mixingprocesses on July 22th and 28th, which may have brought more fleshorganisms from the top. Approximately 3–6 typhoons pass over theChangjiang Estuary annually (Su, 2005; Wei et al., 2007), which inter-vene in hypoxia events. However, effects of typhoon on hypoxia dependon its path and strength and duration.

5. Conclusions

For the first time, we depicted the process of how summer hypoxiaformed, maintained and disappeared using 104-day time-series dataof bottom DO in the northern Changjiang Estuary. During the summerof 2009, two hypoxic events occurred, lasting 17 days in July and18days from late August to September. TheminimumDOconcentrationwas down to 0.17 mg/L during the first hypoxia period, which was thelowest in historical record compared with the published data.

Please cite this article as: Ni, X., et al., The impact of wind mixing on the vasummer, J. Mar. Syst. (2014), http://dx.doi.org/10.1016/j.jmarsys.2014.11

Windmixing played an important role on vertical mixing comparedwith small changes of surface energy input and tidalmixing. The resultsof correlation analyses showed that variation of DO and stratificationwere controlled bywind processes. The strongest wind process inducedby Typhoon Morakot promoted bottom DO by breaking down thepycnocline, which ended the first hypoxia event. Strongwind processesinduced a high DO decreasing rate because of phytoplankton bloomscaused by wind mixing and a high organic decomposition rate in highDO concentrations, forming the second hypoxia event. It was difficultto conclude that strong wind events would definitely relieve hypoxiaoff the Changjiang Estuary.

Acknowledgments

Weare thankful for the SST, SSWand Chl-a satellite data provided bytheNOAA andNASA. This workwas supported by theNational Basic Re-search Program of China under Grant No. 2011CB409803, the NationalNatural Science Foundation of China under Grant Nos. 41206085,41321004, Y506287 and 41203085; the Key Projects in the NationalScience & Technology Pillar Program under Grant Nos. 2008BAC42B01,2008BAC42B03, the Public Science and Technology Research FundsProjects of Ocean under Grant Nos. 201205015, 201105014, the “908”Project under Grant No. 908-01-BC06, Zhejiang Province Public Tech-nology Application Research under Grant No. 2014C33093, and theScientific Research Fund of the Second Institute of Oceanography,State Oceanic Administration, SOA under Grant No. JT1005.

References

Babin, S.M., Carton, J.A., Dickey, T.D., Wiggert, J.D., 2004. Satellite evidence of hurricane-induced phytoplankton blooms in an oceanic desert. J. Geophys. Res. 109, C03043.http://dx.doi.org/10.1029/2003JC001938.

riation of bottom dissolved oxygen off the Changjiang Estuary during.010

9X. Ni et al. / Journal of Marine Systems xxx (2014) xxx–xxx

Boesch, D.F., Rabalais, N.N., 1991. Effects of hypoxia on continental shelf benthos: com-parisons between the New York Bight and the Northern Gulf of Mexico. In: Tyson,R.V., Pearson, T.H. (Eds.), Modem and Ancient Continental Shelf Anoxia. Geol. Sot.London, Spec. Publ. 58, pp. 27–34.

Chen, C.C., Gong, G.C., Shiah, F.K., 2007. Hypoxia in the East China Sea: one of the largestcoastal low-oxygen areas in the world. Mar. Environ. Res. 64 (4), 399–408.

Diaz, R.J., Rosenberg, R., 1995. Marine benthic hypoxia: a review of its ecological effectsand the behavioral responses of benthic macrofauna. Oceanogr. Mar. Biol. Annu.Rev. 33, 245–303.

Diaz, R.J., Rosenberg, R., 2008. Spreading dead zones and consequences for marine ecosys-tems. Science 321, 926–929.

Gu, H.K., 1980. Maximum dissolved oxygen concentration in the vertical distribution ofthe Yellow Sea. Acta Oceanol. Sin. 2, 70–79 (in Chinese with English abstract).

Herbeck, L.S., Unger, D., Krumme, U., Liu, S.M., Jennerjahn, T., 2011. Typhoon-induced pre-cipitation impact on nutrient and suspended matter dynamics of a tropical estuaryaffected by human activities in Hainan, China. Estuar. Coast. Shelf Sci. 93, 375–388.

Hung, C.C., Chung, C.C.C., Gong, G.C., Jan, S., Tsai, Y., Chen, K.S., Chou, W.C., Lee, M.A.,Chang, Y., Chen, M.H., Yang, W.R., Tseng, C.J., Gawarkiewicz, G., 2013. Nutrient supplyinthe southern East China Sea after Typhoon Morakot. J. Mar. Res. 71, 133–150.

Li, D.J., Zhang, J., Huang, D.J., Wu, Y., Liang, J., 2002. Oxygen deficit out of the ChangjiangEstuary. Sci. China Ser. D Earth Sci. 32 (8), 686–694.

Li, H.L., Chen, J.F., Lu, Y., Jin, H.Y., Wang, K., Zhang, H.S., 2011. Seasonal variation of DO andformation mechanism of bottom water hypoxia of Changjiang River Estuary. J. Mar.Sci. 29 (3), 78–87 (in Chinese with English abstract).

Li, Y., Ye, X., Wang, A., Li, H., Chen, J., Qiao, L., 2013. Impact of Typhoon Morakot on chlo-rophyll a distribution on the inner shelf of the East China Sea. Mar. Ecol. Prog. Ser.483, 19–29.

Lin, I.I., Liu, W.T., Wu, C.C., Chiang, J.C.H., Sui, C.H., 2003. Satellite observations of modula-tion of surface winds by typhoon-induced upper ocean cooling. Geophys. Res. Lett.30, 1131. http://dx.doi.org/10.1029/2002GL015674.

MacCready, P., Banas, N.S., Hickey, B.M., Dever, E.P., Liu, Y., 2009. A model study of tide-and wind-induced mixing in the Columbia River Estuary and plume. Cont. ShelfRes. 29 (1), 278–291.

Montgomery, R.B., 1969. The words naviface and oxyty. J. Mar. Res. 27 (1), 161–162.Ni, X., Huang, D., Zhang, T., Zeng, D., 2014. A kind of trawl-resistant bottom mount for

multi-disciplinary parameters observing: developing and application. Ocean Technol.33 (2), 87–92.

Ning, X., Lin, C., Su, J., Liu, C., Hao, Q., Le, F., 2011. Long-term changes of dissolved oxygen,hypoxia, and the responses of the ecosystems in the East China Sea from 1975 to1995. J. Oceanogr. 67 (1), 59–75.

Please cite this article as: Ni, X., et al., The impact of wind mixing on the vasummer, J. Mar. Syst. (2014), http://dx.doi.org/10.1016/j.jmarsys.2014.11.

Rabalais, N.N., Gilbert, D., 2009. Distribution and consequences of hypoxia. In: Urban, E.,Sundby, B., Malanotte-Rizzoli, P., Melillo, J.M. (Eds.), Watersheds, Bays and BoundedSeas. Island Press, Washington, DC, pp. 209–226.

Simpson, J.H., Allen, C.M., Morris, N.C.G., 1978. Fronts on the continental shelf. J. Geophys.Res. 83 (C9), 4607–4614. http://dx.doi.org/10.1029/JC083iC09p04607.

Su, J.L. (Ed.), 2005. The Oceanography in the Chinese Marginal Seas. Ocean Press, Beijing(in Chinese).

Tang, D.L., Sui, G.J., 2014. Typhoon impact and crisis management. Adv. Nat. Technol.Hazards Res. 40. http://dx.doi.org/10.1007/978-3-642-40695-9_13 (@ Springer-Verlag Berlin Heidelberg).

Wang, B., 2009. Hydromorphological mechanisms leading tohypoxia off the ChangjiangEstuary. Mar. Environ. Res. 67 (1), 53–58.

Wang, B., Wei, Q., Chen, J., Xie, L., 2012. Annual cycle of hypoxia off the Changjiang(Yangtze River) Estuary. Mar. Environ. Res. 77, 1–5.

Wang, B., Chen, J.F., Jin, H.Y., Li, H.L., Xu, J., 2013. Inorganic carbon parameters respondingto summer hypoxia outside the Changjiang Estuary and the related implications.J. Ocean Univ. China 12 (4), 568–576.

Wei, H., He, Y., Li, Q., Liu, Z., Wang, H., 2007. Summer hypoxia adjacent to the ChangjiangEstuary. J. Mar. Syst. 67 (3), 292–303.

Xuan, J.L., Huang, D., Zhou, F., Zhu, X.H., Fan, X., 2012. The role of wind on the detachmentof low salinity water in the Changjiang Estuary in summer. J. Geophys. Res. 117,C10004. http://dx.doi.org/10.1029/2012JC008121.

Zhang, J., Gilbert, D., Gooday, A.J., Levin, L., Naqvi, S.W.A., Middelburg, J.J., Scranton,M., Ekau,W., Pena, A., Dewitte, B., Oguz, T., Monteiro, P.M.S., Urban, E., Rabalais, N.N., Ittekkot, V.,Kemp,W.M., Ulloa, O., Elmgren, R., Escobar-Briones, E., Van der Plas, A.K., 2010. Naturaland human-induced hypoxia and consequences for coastal areas: synthesis andfuture development. Biogeosciences 7, 1443–1467. http://dx.doi.org/10.5194/bg-7-1443-2010.

Zhou, F., Ni, X., Xuan, J., Huang, D., 2009. A preliminary study of variations of theChangjiang Diluted Water between August of 1999 and 2006. Acta Oceanol. Sin. 28(6), 1–11.

Zhou, F., Huang, D., Ni, X., Xuan, J., Zhang, J., Zhu, K., 2010. Hydrographic analysis on themulti-time scale variability on hypoxia adjacent to the Changjiang River estuary.Acta Ecol. Sin. 30 (17), 4728–4740 (in Chinese with English abstract).

Zhu, Z., Zhang, J., Wu, Y., Zhang, Y., Lin, J., Liu, S., 2011. Hypoxia off the Changjiang(Yangtze River) Estuary: oxygen depletion and organic matter decomposition. Mar.Chem. 125 (1), 108–116.

Zhu, Z., Zhang, J., Wu, Y., Zhang, Y., Lin, J., Ji, Q., 2013. Early degradation rate particulateorganic carbon and phytoplankton pigments under different dissolved oxygen leveloff the Changjiang (Yangtze) River Estuary. Oceanol. Limnol. Sin. 44 (1), 1–8.

riation of bottom dissolved oxygen off the Changjiang Estuary during010