Microsoft Word - DOS_PNAS_SI_resubmission.docx

Supplementary Information for

Novel sulfur isotope analyses constrain sulfurized porewater fluxes as

a minor component of marine dissolved organic matter

Alexandra A. Phillips, Margot E. White, Michael Seidel, Fenfang Wu, Frank F. Pavia, Preston C.

Kemeny, Audrey C. Ma, Lihini I. Aluwihare, Thorsten Dittmar, Alex L. Sessions

Alexandra A. Phillips

Email:

phillips.alexandra.a@gmail.com

This PDF file

includes:

Supplementary

ext

Figures

Tables S1

Legends

for

Datasets

References

Other

supplementary materials

for this manuscript

include

the following

Datasets S1

Station Specific

Methods

N Atlantic Gyre

Samples (

= 12) of 10 L were collected aboard the

R/V Atlantic Explorer

BATS Station (31° 40'N, 64° 10'W) during the October 2020 leg of the Bermuda Atlantic Time

Series (BATS) cruise (Figure S1).

Nutrient and DOC

concentrations were provided by the

Bermuda Atlantic Time Series, with details available online:

http://bats.bios.edu/

Note: the BATS

protocols measure total organic carbon, which (due to low particulate organic matter in the gyre)

are indistinguishable from dissolved organic carbon. We therefore use DOC, rather t

han TOC, for

consistency

with the rest of the manuscript

NE Pacific OMZ

Samples were collected aboard the

R/V Oceanus

on two cruises to the NE

Pacific oxygen minimum zone (OMZ) region off the coast of Mexico in April 2017 and June 2018

(Figure S2)

. Samples were collected from five sites across the two cruises: Station 2 (16° 30'N,

107° 12'W;

=10), Station 3 (16° 00'N, 1

10° 00'W;

=7), Station 4 (21° 30'N, 109° 30'W;

=5),

Station 5 (24° 41'N, 113° 17'W;

=7), and Station 6 (27° 23'N, 117° 30'W;

=1). Station 6 was

only sampled in 2017. Stations 3 and 4 were only sampled in 2018. Stations 2 and 5 were

sampled in both year

s. 20 L of seawater was extracted per depth on the 2017 cruise, and 5 L of

seawater per sample on the 2018 cruises. One additional sample was collected in April 2018

aboard the

R/V Western Flyer

at Station M (34° 30'N, 123° 00'W;

=1): this 15L sample was

collected by ROV

Doc Ricketts

1 m above the seafloor.

Details on nutrients and DOC

measurements are detailed elsewhere

(1)

NE Pacific and NE Pacific Shelf

Seawater samples for DOM

SPE

analysis were collected aboard

the

R/V Sally Ride

at two stations in the NE Pacific off the coast of Southern California in

November 2020

(Figure S3)

: on the shelf (33° 11'N, 118° 37'W;

= 7) and offshore (32° 33'N,

120° 31'W;

= 12). ~7

9 L of seawater

was collected per sample

. No nutrient or DOC data are

reported from these cruises.

N Pacific Gyre

Samples (

= 12) were collected aboard the

R/V Kilo Moana

at Station ALOHA

(22° 45'N, 158° 00'W) during the November 2020 leg of the Hawaii Ocean Time

eries (HOT

;

Figure S4

). 10 L seawater samples were collected for DOM

SPE

analysis.

Metadata

provided

by the Hawaii Ocean Time

series,

which are available publicly or by request. Details on

methodology can be accessed online:

https://hahana.soest.hawaii.edu/hot/protocols/protocols.html

San Pedro Basin

Samples

(

=9) were collected aboard the

R/V Yellowfin

on the February 2021

leg of the San Pedro Ocean Time

series (SPOT) cruise off the coast of Southern California to one

station (33° 33'N, 118° 24'W

; Figure S5

). ~7

9 L seawater were collected for DOM

SPE

analysi

s per

sample.

Nutrient data was provided by the San Pedro Ocean Time

series and is available online

or by request. Details on methodology can be accessed online:

https://dornsife.usc.edu/spot/methods/

S Pacific Gyre

Seawater samples (

=7) of 5

7 L were collected aboard the

R/V Ronald Brown

during the first leg of the P18 GO

SHIP (US Global Ocean Ship

based Hydrographics

Investigations Program) cruise at a single station (12° 59'S, 103° 00'W) in Novem

ber to

December 2016 off the coast of Peru

(Figure S6)

Nutrient data from this cruise is available online

https://cchdo.ucsd.edu/

while DOC measurements followed the same protocols as the NE

Pacific OMZ samples

(1)

Caeté Estuary and Caeté Estuary Porewater

Water samples were collected by boat in the dry

season of October 2018 along a 35

km salinity transect from riverine (1° 1'S, 46° 45'W) to marine

(0° 54'S, 46° 36'W), ending at the Amazonian mangrove

fringed estuary near Bragança City

(Pará State, Brazil) in the Caeté Estuary

(2)

. Porewater samples (~3

6 L) were taken in a tidal

creek close to the marine end of the transect, at the mangrove

fringed estuary. Water samples

were filtered using a thoroughly pre

rinsed 1.0 μm Causapure filter cartridge (CPR

001

DOX,

PP, Infiltec).

OC and nutrient data are not reported for these stations.

North Sea Porewater

Porewater from an anoxic intertidal flat w

collected in the Northern

German Wadden Sea at the Janssand site close to the barrier island Spiekeroog in 2010

(3)

Samples were collected in 10 L acid

cleaned polycarbonate carboys after digging a hole and

scooping porewater out of the holes. The samples were filtered through pre

combusted (450°C, 4

hours) glass fi

ber filters (Whatman 1 μm GMF and 0.7 μm GFF) and solid phase extracted

(4)

DOC and nutrient data are not reported for these stations.

Station Specific Results

N Atlantic Gyre

Samples were analyzed between 5 and 4800 m water depth

(Figure S

)

values decreased with depth from a maximum of 18.

‰ in the surface (5 m) to a minimum value

of 14.

‰ below the photic zone. C:S ratios generally increased with depth, from a surface

minimum of 191 to a maximum value of 303 in the deep ocean (2000 m). DOM

C values

were constant with depth, averaging

22.3 ± 0.1‰.

NE Pacific OMZ

Samples were collected at six stations largely within the upper

50 m of the

water column

(Figure S9)

. DOM

SPE

S values showed no change with depth in the upper

averaging 19.0 ± 0.

‰.

Deeper samples were limited, but within error,

averaging 1

± 1.0‰.

DOM

SPE

C values were also constant with depth, averaging

22.2 ± 0.3‰. C:S values

generally decreased w

ith depth but had a large range in the surface ocean (< 1

0m), between

and 291. Variations in C:S values and δ

S values were not related to station number or

sampling year. NE Pacific OMZ samples displayed no correlation with dissolved oxygen

concentrations, which ranged from ~0 to 250 μM across the sample set (Figure S

). However,

samples did show a significant negative correlation between δ

S values and C:S ratios (R

0.47,

<.001).

NE Pacific and NE Pacific Shelf

Samples were analyzed f

rom between 5 and 1275 m on the

shelf and between 5 and 3790 m in the coastal ocean (Figure S

). Carbon isotopes were similar

across stations, with lower DOM

SPE

C values in the upper 1

0 m (

22.6 ± 0.3‰) than below the

photic zone (

22.1 ± 0.1‰). δ

values were largely constant in the NE Pacific, averaging 17.

0.5‰ throughout the water column. Meanwhile, δ

S values from the NE Pacific shelf decreased

from a maximum of

19.1

± 0.

‰ to 1

5.0

± 0.2‰ from surface to deep. DOM

SPE

C:S ratios

fluctuated,

with a minimum of 160 ± 4 to a maximum of 275 ± 3.

N Pacific Gyre

Samples were analyzed between 5 and 4800 m (

=12; Figure S

). DOM

SPE

C values were constant throughout the water column, averaging

22.7 ± 0.1‰. DOM

SPE

values were also fairly constant, with an average of 18.1

± 0.8‰, but with higher variability. C:S

ratios were similarly variable, with an average of 230 ± 25. δ

S values were also negatively

correlated with C:S ratios (

2

= 0.47,

<.05).

San Pedro Ba

sin

Samples were analyzed between 2 and 880 m

(Figure S13)

. San Pedro is a

persistently hypoxic (<1 mL/L dissolved oxygen) restricted basin off the coast of Los Angeles,

Southern California. DOM

SPE

samples showed no change in δ

C values, averaging

22.5 ±

0.2‰. In contrast, C:S ratios and δ

S values were strongly dependent on depth. C:S ratios

increased from 153 ± 3 in the surface (2 m) to 287 ± 6 at depth (880 m) while δ

S values from

the same d

epths decreased from

18.6 ± 0.2‰ to 16.7 ± 0.3‰.

C:S and δ

S values were also

significantly negatively correlated (

2

= 0.71,

<.01).

S Pacific Gyre

Samples were analyzed between 435 and 4411 m (Figure S

). DOM

SPE

values fluctuated between a minimum of

22.7 ± 0.1‰ and a maximum of

21.5 ± 0.1‰, although

without depth dependence. DOM

SPE

S values decreased from 1

± 0.2‰ to 16.

± 0.2‰ with

depth while C:S ratios increased from 235 ± 3 to 277 ± 3.

Fig. S1.

CTD profile from October 2020 Bermuda Atlantic Time Series cruise at BATS station in

the N Atlantic Gyre. Profiles of salinity (psu, yellow line) and temperature ( ̊C, magenta

line) are

plotted on the upper x

axis, dissolved oxygen concentration (μM, blue line) on the second upper

axis and fluorescence (relative fluorescence units) in green on the lower x

axis.

ampling

depths

for DOM

are marked on the right side of the y

axis

with grey triangles.

Fig. S2

Physical properties from stations sampled in the NE Pacific OMZ from 2017

2018. Note

that while x

axes are constant for each subplot, y

axes are not.

Colors and units are the same as

Figure S1.

Sampling depths for DOM are marked on the right side of the y

axis with grey

triangles.

Fig

CTD profile from November 2020 cruise to the NE Pacific and NE Pacific shelf. Note the

change of scale in the y axis.

Colors and units are the same as Figure S1.

Sampling depths for

DOM are marked on the right

side of the y

axis with grey triangles.

Fig

CTD profile from

November

2020

Hawaii Ocean Timeseries

cruise at

Aloha

station in the

Pacific

Gyre.

Colors and units are the same as Figure S1.

Sampling depths for DOM are

marked on the right side of t

he y

axis with grey triangles.

Fig

CTD profile from San Pedro Basin during the February 2021 SPOT cruise.

Colors and

units are the same as Figure S1.

Sampling depths for DOM are marked on the right side of the y

axis with grey triangles.

Fig

CTD profile from November 2016 GO

SHIP cruise to the S Pacific Gyre. Colors and units

are the same as Figure S1.

Sampling depths for DOM are marked on the right side of the y

axis

with grey triangles.

Fig

Metadata analysis of

potential correlations between DOM

SPE

S values against (A)

values, (B) C:S ratios (C) DOC

(D) DOS calculated from DOS (apparent DOS) and DOM

SPE

C:S

(

)

oxygen

(F)

salinity

(G)

CTD

fluorescence

(a proxy for chlorophyll a)

(H)

temperature

(I)

nitrat

(J)

phosphate and (K) silicate

. Only C:S ratios had a significant negative correlation with

S values. Temperature

and

DOC were positively correlated. However, C:S ratios, δ

S values,

temperature, and DOC concentration all also vary with depth. Sample

s from different depths were

binned into 0

100 m (magenta), 100

250 m (pink), 250

500 m (yellow), 500

1000 m (green)

1000

2500 m (light blue), and 2500

5000 m (dark blue).

Fig. S

DOM samples

from the N Atlantic Gyre

against depth for (A) C:S

ratios, (B) δ

S values

C values. Note the scale breaks at 150 m 750 m on the y

axis

. 1

standard errors are

shown for

S values

, but

error bars for C:S and

C values

are within the size of the symbol.

Fig. S

DOM samples

from the NE Pacific OMZ

against depth for (A) C:S ratios, (B) δ

S values

C values. Note the scale breaks at 150 m 750 m on the y

axis

. 1

standard errors are

shown for

S values

, but

error bars for C:S and

C values

are within the size of the s

ymbol.

Fig

Map (A) of NE Pacific OMZ sampling locations in this study (magenta shapes), with the

sampling site from Raven et al. 2020

(5)

marked with a dark grey inverted triangle. Oxygen

concentrati

on is plotted against DOM

SPE

(B) δ

S values and (C) C:S ratios.

Fig. S1

DOM samples

from the NE Pacific (

light pink right

facing triangles

) and NE Pacific Shelf

(yellow circles)

against depth for (A) C:S ratios, (B) δ

S values (C) δ

values. Note the scale

breaks at 150 m 750 m on the y

axis

. 1

standard errors are shown for

S values

, but

error

bars for C:S and

C values

are within the size of the symbol.

Fig. S1

DOM samples

from the N Pacific Gyre

against depth for (A) C:S ratios, (B) δ

S values

C values. Note the scale breaks at 150 m 750 m on the y

axis

. 1

standard errors are

shown for

S values

, but

error bars for C:S and

C values

are within the size of the s

ymbol.

Fig. S1

DOM samples

from the San Pedro Basin

against depth for (A) C:S ratios, (B) δ

values (C) δ

C values. Note the scale breaks at 150 m 750 m on the y

axis

. 1

standard errors

are shown for

S values

, but

error bars for C:S and

C values

are within the size of the

symbol.

Fig. S1

DOM samples

from the S Pacific Gyre

against depth for (A) C:S ratios, (B) δ

S values

C values. Note the scale breaks at 150 m 750 m on the y

axis

. 1

standard errors are

shown for

S values

, but

error bars for C:S and

C values

are within the size of the symbol.

Table S1.

Ranges of parameters measured by this study on DOM

SPE

samples from water

columns and porewater, by station. Note that there is only one sample from the North Sea

porewater.

Station

DOM

SPE

C:S (molar)

DOM

SPE

S (‰)

DOM

SPE

C (‰)

North Sea Porewater

0.2

23.3

Caeté

Porewater

–

2.7

–

0.7

27.1

–

26.7

Caeté

Estuary

–

117

4.2

–

8.7

28.6

–

24.2

N Atlantic Gyre

191

–

303

14.9

–

18.9

22.4

–

22.1

NE Pacific OMZ

166

–

293

16.4

–

19.9

22.7

–

21.6

NE Pacific

186

–

275

17.1

–

18.8

23.2

–

21.9

NE Pacific Shelf

160

–

255

15.0

–

19.1

22.7

–

22.0

N Pacific Gyre

197

–

271

16.9

–

19.3

22.9

–

22.5

San Pedro Basin

153

–

287

15.8

–

18.8

22.8

–

22.4

S Pacific Gyre

235

–

277

16.7

–

18.1

22.7

–

21.5

Dataset

Supplementary

excel data file with all collected EA

IRMS

data (

DOM

SPE

DOM

SPE

DOM

SPE

C:S

), physical parameters (from CTD casts at each station), and chemical

data, sorted by location.