wahl et al. c)

1
th Calafat F.M. et al. (2012), Mechanisms of decadal sea level variability in the eastern North Atlantic and the Mediterranean Sea, J. Geophys. Res., 117, C09022. Dangendorf, S. et al. (2014a), A new atmospheric proxy for sea level variability in the southeastern North Sea: observations and future ensemble projections, Climate Dynamics, 43, 447-467. Dangendorf, S. et al. (2014b), Mean sea level variability in the North Sea: processes and implications, J. Geophys. Res., 119, 6820- 6841. Ishii, M., & M. Kimoto (2009), Reevaluation of historical ocean heat content variations with time-varying XBT and MBT depth bias corrections, J. Oceanogr., 65, 287–299. Wahl, T. et al. (2011), Improved estimates of mean sea level changes in the German Bight over the last 166 years, Ocean Dynamics, 61(5), 701-715. Wahl et al. Ishii and Kimoto a) Investigation area with tide gauge locations and sea surface height (SSH) anomalies from Altimetry. b) Monthly de-seasonalized time series (grey) of local MSL (GIA corrected). The thick lines represent the 12 months low pass filtered component. The different colors mark sub-regions used for the calculation of regional indices. Composite plot showing wind and pressure condi- tions (right) during times of particularly high (>2*stdv) minus particularly low (<2*stdv) monthly MSL events (left). The plots are given for four regional indices (averages). Coherence of decadal (48 months low pass filtered) sea level (black: obser- ved, blue: atmospheri- cally corrected) in the North Sea (a). Also shown (b) is the steric height obtained from hydrographic obser- vations near Sognes- joen in the Norwegian Trench (referenced down to 700m). a) Explained variability of barotropic atmospheric forcing esti- mated with a stepwise multiple linear regression model (LRM; grey bars). The colored dots mark the contribution given by each predictor alone. Only predictors explaining a significant fraction of variability are shown (95% confidence level). b) Stdv. of the observed (black) and atmospherically corrected (red) monthly MSL time series. The dashed lines with the shaded areas mark the mean and stdv. over all stations. a) Average wind and pressure conditions over the period 1945- 2011. b) Correlation between the low pass filtered (48 months) and atmospherically corrected North Sea index (NSI) and SSH from Altime- try, and c) the steric height for the upper 200m. Reconstruction of decadal MSL variability in the North Sea based on barotropically corrected Newlyin (NEW) and steric sea level west of the UK. NEW is used as a proxy for boundary wave forcing along the conti- nental slope of the North Atlantic [Calafat et al., 2012]. Model performance of MICOM. Shown is a comparison between the NSI from coastal tide gauges and the sea surface height (SSH) from the closest grid points in MICOM. The comparison sug-gests that MICOM reproduces the majority of coastal MSL in the North Sea (81%). Wavelet coher- ence between (top) the NSI and eastern boun- dary longshore winds (LSW, integrated from the equator north- wards), b) NEW MSL and LSW, and c) the NSI and NEW. The results suggest that winds play an important role in forcing decadal sea level along the eastern boundary of the North Atlantic. However, travelling further north- wards, the signal seems to be increasingly distur- bed by topography. Schematic of eastern boundary wind forcing and the respective changes in coastal sea level. Correlations between the low pass filtered NSI (48 months) from MICOM and each grid point time series calculated for the a) SSH, b) ocean bottom pressure (OBP), and c) steric height. The correlation analysis suggests a coherent OBP signal extending from the west coast of Africa along the continental slope into the Arctic ocean. Warm Water Cold Water Slope Southward Winds Mass Transport Warm Water Cold Water Slope Northward Winds Mass Transport Explained Variability [%] Oostende Vlissingen Hoek van Holland IJmuiden Den Helder West−Terschelling Delfzijl Norderney Cuxhaven Helgoland Hörnum Esbjerg Hanstholm Hirtshals Tregde Stavanger Bergen Lerwick Wick Aberdeen North Shields Lowestoft a) SLP U−Wind V−Wind Bar. LRM 20 40 60 80 Standard Deviation [cm] b) observed corrected 2 4 6 8 10 12 14 16 Time [yr] MSL [cm] Individual Gauges [atmospherically corrected] North Sea Index [atmospherically corrected] Newlyn [IB corrected]: R² = 0.70 Newlyn [IB corrected] + steric West−UK: R² = 0.79 1900 1920 1940 1960 1980 2000 2020 −10 −8 −6 −4 −2 0 2 4 6 8 10 MSL [cm] Model Performance MSL Tide Gauge MSL MICOM (r=0.9; VARexp = 81%) Time [yr] 1950 1960 1970 1980 1990 2000 −40 −20 0 20 40 SSH a) OBP b) Correlation [−] −1 −0.8 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 0.8 1 Steric c) Time [yr] 1) Oostende [1.97±0.13] 2) Vlissingen [1.69±0.07] 3) Hoek van Holland [2.29±0.06] 4) IJmuiden [1.44±0.08] 5) Den Helder [1.17±0.06] 6) West−Terschelling [0.6±0.14] 7) Delfzijl [1.41±0.08] 8) Norderney [1.84±0.12] 9) Cuxhaven [1.99±0.1] 10) Helgoland [1.29±0.37] 11) Hörnum [1.63±0.28] 1800 1850 1900 1950 2000 Time [yr] 12) Esbjerg [0.69±0.13] 13) Hanstholm [3.01±0.52] 14) Hirtshals [1.15±0.11] 15) Tregde [1.29±0.1] 16) Stavanger [1.77±0.12] 17) Bergen [1.37±0.12] 18) Lerwick [0.98±0.23] 19) Wick [1.92±0.32] 20) Aberdeen [1.67±0.05] 21) North Shields [2.26±0.08] 22) Lowestoft [2.06±0.22] 1800 1850 1900 1950 2000 b) 70cm 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 PSMSL North Sea PSMSL North Atlantic AMSeL German Bight 1) Oostende 2) Vlissingen 3) Hoek van Holland 4) IJmuiden 5) Den Helder 6) West−Terschelling 7) Delfzijl 8) Norderney 9) Cuxhaven 10) Helgoland 11) Hörnum 12) Esbjerg 13) Hanstholm 14) Hirtshals 15) Tregde 16) Stavanger 17) Bergen 18) Lerwick 19) Wick 20) Aberdeen 21) North Shields 22) Lowestoft 23) Brest 24) Newlyn −10 −5 0 5 10 15 20 25 30 35 MSLA [cm] a) SLP [hPa] 1005 1010 1015 1020 Average wind and pressure pattern Correlation [-] −0.6 −0.4 −0.2 0 0.2 0.4 0.6 NSI vs. SSH −0.6 −0.4 −0.2 0 0.2 0.4 0.6 NSI vs. Steric Correlation [-] a) b) c) Mean Wind Stress: 0.05 N/m2 MSL [cm] Region 1 (Station 1−6) −40 −20 0 20 40 SLP [hPa] −20 −15 −10 −5 0 5 10 MSL [cm] Region 2 (Station 7−14) −40 −20 0 20 40 SLP [hPa] −20 −15 −10 −5 0 5 10 MSL [cm] Region 3 (Station 15−17) −40 −20 0 20 40 SLP [hPa] −20 −15 −10 −5 0 5 10 Time [yr] MSL [cm] Region 4 (Station 18−22) 1880 1920 1960 2000 −40 −20 0 20 40 SLP [hPa] −20 −15 −10 −5 0 5 10 0.4 N/m2 −2 0 2 NAO [−] −3 0 3 0.25 0.5 1 2 4 8 16 32 0 0.2 0.4 0.6 0.8 1 −3 0 3 MSL [cm] Period [yr] 0.25 0.5 1 2 4 8 16 32 0 0.2 0.4 0.6 0.8 1 −3 0 3 Time [yr] 1880 1900 1920 1940 1960 1980 2000 0.25 0.5 1 2 4 8 16 32 0 0.2 0.4 0.6 0.8 1 MSL [cm] IJmuiden Delfzijl Esbjerg Stavanger Aberdeen a) Observed Atmospherically corrected Steric height Sognesjoen North Sea Index 0 10 20 30 40 50 60 MSL [cm] Time [yr] b) r = 0.66 r = 0.68 r = 0.57 1940 1950 1960 1970 1980 1990 2000 2010 −5 −3 −1 1 3 5

Upload: others

Post on 24-Feb-2022

5 views

Category:

Documents


0 download

TRANSCRIPT

th

Calafat F.M. et al. (2012), Mechanisms of decadal sea level variability in the eastern North Atlantic and the Mediterranean Sea, J. Geophys. Res., 117, C09022.

Dangendorf, S. et al. (2014a), A new atmospheric proxy for sea level variability in the southeastern North Sea: observations and future ensemble projections, Climate Dynamics, 43, 447-467.

Dangendorf, S. et al. (2014b), Mean sea level variability in the North Sea: processes and implications, J. Geophys. Res., 119, 6820-6841.

Ishii, M., & M. Kimoto (2009), Reevaluation of historical ocean heat content variations with time-varying XBT and MBT depth bias corrections, J. Oceanogr., 65, 287–299.

Wahl, T. et al. (2011), Improved estimates of mean sea level changes in the German Bight over the last 166 years, Ocean Dynamics, 61(5), 701-715.

Wahl et al.

Ishii and Kimoto

a) Investigation area with tide gauge locations and sea surface height (SSH) anomalies from Altimetry. b) Monthly de-seasonalized time series (grey) of local MSL (GIA corrected). The thick lines represent the 12 months low pass filtered component. The different colors mark sub-regions used for the calculation of regional indices.

Composite plot showing wind and pressure condi-tions (right) during times of particularly high (>2*stdv) minus particularly low (<2*stdv) monthly MSL events (left). The plots are given for four regional indices (averages).

Coherence of decadal (48 months low pass filtered) sea level (black: obser-ved, blue: atmospheri-cally corrected) in the North Sea (a). Also shown (b) is the steric height obtained from hydrographic obser-vations near Sognes-joen in the Norwegian Trench (referenced down to 700m).

a) Explained variability of barotropic atmospheric forcing esti-mated with a stepwise multiple linear regression model (LRM; grey bars). The colored dots mark the contribution given by each predictor alone. Only predictors explaining a significant fraction of variability are shown (95% confidence level). b) Stdv. of the observed (black) and atmospherically corrected (red) monthly MSL time series. The dashed lines with the shaded areas mark the mean and stdv. over all stations.

a) Average wind and pressure conditions over the period 1945-2011. b) Correlation between the low pass filtered (48 months) and atmospherically corrected North Sea index (NSI) and SSH from Altime-try, and c) the steric height for the upper 200m.

Reconstruction of decadal MSL variability in the North Sea based on barotropically corrected Newlyin (NEW) and steric sea level west of the UK. NEW is used as a proxy for boundary wave forcing along the conti-nental slope of the North Atlantic [Calafat et al., 2012].

Model performance of MICOM. Shown is a comparison between the NSI from coastal tide gauges and the sea surface height (SSH) from the closest grid points in MICOM. The comparison sug-gests that MICOM reproduces the majority of coastal MSL in the North Sea (81%).

Wavelet coher-ence between (top) the NSI and eastern boun-dary longshore winds (LSW, integrated from the equator north-wards), b) NEW MSL and LSW, and c) the NSI and NEW. The results suggest that winds play an important role in forcing decadal sea level along the eastern boundary of the North Atlantic. However, travelling further north-wards, the signal seems to be increasingly distur-bed by topography.

Schematic of eastern boundary wind forcing and the respective changes in coastal sea level.

Correlations between the low pass filtered NSI (48 months) from MICOM and each grid point time series calculated for the a) SSH, b) ocean bottom pressure (OBP), and c) steric height. The correlation analysis suggests a coherent OBP signal extending from the west coast of Africa along the continental slope into the Arctic ocean.

Warm Water

Cold Water

Slope

Southward Winds

Mass Transport

Warm Water

Cold Water

Slope

Northward Winds

Mass Transport

Explained Variability [%]

OostendeVlissingenHoek van HollandIJmuidenDen HelderWest−TerschellingDelfzijlNorderneyCuxhavenHelgolandHörnumEsbjergHanstholmHirtshalsTregdeStavangerBergenLerwickWickAberdeenNorth ShieldsLowestoft

a)

SLPU−WindV−WindBar. LRM

20 40 60 80Standard Deviation [cm]

b)

observedcorrected

2 4 6 8 10 12 14 16

Time [yr]

MSL

[cm

]

Individual Gauges [atmospherically corrected]North Sea Index [atmospherically corrected]Newlyn [IB corrected]: R² = 0.70Newlyn [IB corrected] + steric West−UK: R² = 0.79

1900 1920 1940 1960 1980 2000 2020−10

−8

−6

−4

−2

0

2

4

6

8

10

MSL

[cm

]

Model Performance

MSL Tide GaugeMSL MICOM (r=0.9; VARexp = 81%)

Time [yr]1950 1960 1970 1980 1990 2000

−40

−20

0

20

40

SSHa)

OBPb)

Corre

latio

n[−

]

−1

−0.8

−0.6

−0.4

−0.2

0

0.2

0.4

0.6

0.8

1Steric

c)

Time [yr]

1) Oostende[1.97±0.13]

2) Vlissingen[1.69±0.07]

3) Hoek van Holland[2.29±0.06]

4) IJmuiden[1.44±0.08]

5) Den Helder[1.17±0.06]

6) West−Terschelling[0.6±0.14]

7) Delfzijl[1.41±0.08]

8) Norderney[1.84±0.12]

9) Cuxhaven[1.99±0.1]

10) Helgoland[1.29±0.37]

11) Hörnum[1.63±0.28]

1800 1850 1900 1950 2000Time [yr]

12) Esbjerg[0.69±0.13]

13) Hanstholm[3.01±0.52]

14) Hirtshals[1.15±0.11]

15) Tregde[1.29±0.1]

16) Stavanger[1.77±0.12]

17) Bergen[1.37±0.12]

18) Lerwick[0.98±0.23]

19) Wick[1.92±0.32]

20) Aberdeen[1.67±0.05]

21) North Shields[2.26±0.08]

22) Lowestoft[2.06±0.22]

1800 1850 1900 1950 2000

b)

70cm

1 23

456

7 8 910 11

12

131415

16

1718

19

20

21

22

23

24 PSMSL North SeaPSMSL North AtlanticAMSeL German Bight

1) Oostende2) Vlissingen3) Hoek van Holland4) IJmuiden5) Den Helder6) West−Terschelling7) Delfzijl8) Norderney9) Cuxhaven10) Helgoland11) Hörnum12) Esbjerg13) Hanstholm14) Hirtshals15) Tregde16) Stavanger17) Bergen18) Lerwick19) Wick20) Aberdeen21) North Shields22) Lowestoft23) Brest24) Newlyn

−10

−5

0

5

10

15

20

25

30

35

MSL

A[c

m]

a)SLP [hPa]

1005

1010

1015

1020

Average wind and pressure pattern

Corr

elat

ion

[-]

−0.6

−0.4

−0.2

0

0.2

0.4

0.6

NSI vs. SSH

−0.6

−0.4

−0.2

0

0.2

0.4

0.6

NSI vs. Steric

Corr

elat

ion

[-]

a)

b)

c)

Mean Wind Stress:0.05 N/m2

MSL

[cm

]

Region 1 (Station 1−6)

−40

−20

0

20

40

SLP

[hPa

]

−20−15−10−50510

MSL

[cm

]

Region 2 (Station 7−14)

−40

−20

0

20

40

SLP

[hPa

]

−20−15−10−50510

MSL

[cm

]

Region 3 (Station 15−17)

−40

−20

0

20

40

SLP

[hPa

]

−20−15−10−50510

Time [yr]

MSL

[cm

]

Region 4 (Station 18−22)

1880 1920 1960 2000

−40

−20

0

20

40

SLP

[hPa

]

−20−15−10−50510

0.4 N/m2

−202

NA

O[−

]

−303

0.250.5

1248

1632

0

0.2

0.4

0.6

0.8

1

−303

MSL

[cm

]Pe

riod

[yr]

0.250.5

1248

1632

0

0.2

0.4

0.6

0.8

1

−303

Time [yr]1880 1900 1920 1940 1960 1980 2000

0.250.5

1248

1632

0

0.2

0.4

0.6

0.8

1

MSL

[cm

]

IJmuiden

Delfzijl

Esbjerg

Stavanger

Aberdeen

a)

ObservedAtmospherically correctedSteric height SognesjoenNorth Sea Index

0

10

20

30

40

50

60

MSL

[cm

]

Time [yr]

b)

r = 0.66r = 0.68r = 0.57

1940 1950 1960 1970 1980 1990 2000 2010−5

−3

−1

1

3

5