To the North Pole and back: a pan-Arctic barium synthesisLaura M. Whitmore(USM), Tristan Horner(WHOI), Phoebe Lam(UCSC), Chantal Mears(HZG),

Rob Rember(UAF),Frank Dehairs(VUB), Helmuth Thomas(HZG), Yang Xiang(UCSC), Alan Shiller(USM)

Background Take Home Messages

Acknowledgements: Thanks to the crews and chief scientists of the GN01, GN02, GN03, and GN04 expeditions. Funded by NSF Chemical Oceanography Program

Barium Distributions Mass Balance Carbon Proxy

Isotopes

Dissolved barium (dBa) has been used as a water mass tracer and Ba can be applied as a proxy for productivity due to a relationship between particulate Ba (pBa) and organic carbon. The utility of Ba depends on predictable behavior of the tracer.

We aim to describe and quantify non-conservative behavior (sources and sinks) of Ba in the Arctic marine system.

New data in this study are from GEOTRACES campaigns conducted throughout the Arctic Ocean from July – October 2015.

There are substantial non-conservative components to the dBa distribution in the Arctic Ocean.

• Ba flux from the shelf contributes the largest individual component to the surface 500 m of the basin.

We observed a large range of POC:pBa ratios and a log-linear relationship between pBa and POC; median and mean values from this study are higher than previous records.

A linear Ba isotope and 1/[dBa] trend indicates mixing is a dominant process, but mixing of what endmembers?

Contact: laura.whitmore@usm.eduUniversity of Southern Mississippi

Barium Behavior

𝑓𝑠ℎ𝑒𝑙𝑣𝑒𝑠 = 𝑓𝑡𝑟𝑎𝑛𝑠𝑝𝑜𝑟𝑡 − 𝑓𝑃𝑎𝑐𝑖𝑓𝑖𝑐 − 𝑓𝐴𝑡𝑙𝑎𝑛𝑡𝑖𝑐 − 𝑓𝑅𝑖𝑣𝑒𝑟𝑠

Simplified Form

Flux Term Flux (𝐦𝐨𝐥 𝐁𝐚 𝐲−𝟏) Budget

ftransport 2.2 × 1010 –

fPacific 2.0 × 109 9.1 %

fAtlantic 9.4 × 109 42.3 %

frivers 8.1 × 108 3.6 %

fshelf 𝟏. 𝟎 × 𝟏𝟎𝟏𝟎 45.0 %

Barium isotopes have a linear relationship, which indicates mixing – although of what endmembers is unclear. Two low [dBa] shelf samples deviate from the trend.

A full water column profile from the Canada basin has an isotopic maximum around 500 m, likely indicating formation of barite. The particle maximum is above the isotopic maximum and likely indicates lateral transport of pBa off the shelves.

BRINES?ICE MELT

4000

500

Sinking and remineralization

pBa formation associated with organic matter

Ba distribution is controlled predominantly by inputs from Atlantic, Pacific, and riverine contributions. The role of ice, pBa, and shelves on the Ba mass balance is poorly constrained.

We assume that ice and pBa input and removal terms are small; we solve for the shelf flux considering only the surface 500 m:

Study Range Median Mean

This Study 90 − 90,000 2,260 6,450

Thomas et al. (2011) 𝑅𝑒𝑝𝑜𝑟𝑡𝑒𝑑 𝑚𝑒𝑎𝑛 𝑜𝑓 𝑙𝑖𝑡𝑒𝑟𝑡𝑢𝑟𝑒 𝑣𝑎𝑙𝑢𝑒𝑠: 2,575

Francois et al. (1995) 240 − 7,200 916 1,350

Dymond et al. (1992) 263 − 1,410 630 2,600

Corg:pBanonlithogenic (mol:mol)

Dep

th (

m)

North Pole

Dep

th (

m)

Beaufort Sea Barents Sea

0

100

200

300

400

500

dB

a (n

M)

80

70

60

50

40

0

100

200

300

400

500

1.4

1

0.6

Beaufort Sea Barents Sea

Ba S

atu

ratio

n In

dex

De

pth

(m

)

0

100

200

300

400

500Beaufort Sea Barents Sea

Large Size Fraction (> 51 µm) Small Size Fraction (1 – 51 µm)

Beaufort Sea Barents Sea

No

nlith

og

en

ic

pB

a (n

M)

0.3

0.1

0.03

0.01

1/[dBa] (kg mol-1)

δ1

38

/134B

a (

‰)

0.070.01 0.050.03

0.3

50.4

50

.55

Yenisey

~ 75 nM

Mackenzie300 – 500 nM

Yukon

~ 300 nM Kolyma

~ 75 nM

Lena

~ 115 nM

Ob’

~ 120 nM

If dBa behaves conservatively then we expect:

𝐵𝑎𝑝𝑟𝑒𝑑 = 𝐵𝑎𝑟𝑖𝑣𝑒𝑟𝑠 f𝑟𝑖𝑣𝑒𝑟𝑠 + 𝐵𝑎𝑃𝑎𝑐 f𝑃𝑎𝑐 + 𝐵𝑎𝐴𝑡𝑙 𝑓𝐴𝑡𝑙 + 𝐵𝑎𝑖𝑐𝑒 𝑓𝑖𝑐𝑒

𝐵𝑎𝑎𝑛𝑜𝑚𝑎𝑙𝑦 = 𝑑𝐵𝑎𝑜𝑏𝑠 − 𝑑𝐵𝑎𝑝𝑟𝑒𝑑

We define Baanomaly as the difference between observed and predicted Ba:

Using water mass fractions from Pasqualini et al. for the US transect, we observe a positive Ba anomaly between 25 and 250 m, which is consistent with Pacific-derived waters. Is this evidence of a shelf Ba signal?

Dep

th (

m)

0

100

200

300

400

500

-20 -10 0 10 20

Baanomaly (nM)

Endmember dBa (𝐧𝐌)

Pacific 57

Atlantic 45

Rivers 100

Ice 0

0

1000

2000

3000

40000.35 0.45 0.05 0.25

δ138/134Ba

(‰)

Nonlithogenic

pBa (nM)

Dep

th (

m)

Beaufort

Barents

Ra-based Arctic fshelf estimate:

𝑓𝑠ℎ𝑒𝑙𝑓 =𝐵𝑎

𝑅𝑎× 𝑓𝑠ℎ𝑒𝑙𝑓 (𝑅𝑎)

𝟗. 𝟏 × 𝟏𝟎𝟗 𝐦𝐨𝐥 𝐁𝐚 𝐲−𝟏

Ra flux from Kipp et al. (2018)

Area scaled benthic flux chamber

estimate:

𝟐. 𝟓 × 𝟏𝟎𝟗 𝐦𝐨𝐥 𝐁𝐚 𝐲−𝟏

California shelf benthic flux from McManus et al. (1994)

Dep

th (

m)

No

nli

tho

ge

nic

pB

a(n

M) 1

0.1

0.01

0.001

100.10.010.001 1

POC (µM)

Large Size Fraction (> 51 µm)

No

nli

tho

ge

nic

pB

a(n

M)

1

0.1

0.01

0.001

0.10.010.001 1

POC (µM)

Small Size Fraction (1 – 51 µm)

Dep

th (m

)

3000

0𝑓𝑃𝑎𝑐𝑖𝑓𝑖𝑐 + 𝒇𝑨𝒕𝒍𝒂𝒏𝒕𝒊𝒄 +𝒇𝑹𝒊𝒗𝒆𝒓𝒔 +𝑓𝑖𝑐𝑒 + 𝑓𝑠ℎ𝑒𝑙𝑣𝑒𝑠 = 𝑓𝑝𝐵𝑎 + 𝑓𝑖𝑐𝑒 + 𝑓𝑡𝑟𝑎𝑛𝑠𝑝𝑜𝑟𝑡

Inputs Outputs