1. Terrestrial carbon and nutrient fluxes and the biogeochemical responses in the Mississippi River plume and Northern Gulf of Mexico
Wei-Jun Cai
The University of Georgia
Department of Marine Sciences
Athens, GA, USA
Science Workshop, July 21-24, 2008, Woods Hole, MA, USA

2. Acknowledgement Co-authors & contributors
Xianghui Guo, Wei-Jen Huang, Yongchen Wang
Steve Lohrenz---USM
Michael Murrell---EPA
Rik Wanninkhof and T.H. Peng---NOAA-AOML
Minhan Dai---Xiamen Univ., China
Funding
NASA— ( )
NOAA— ( )
NSF—Chemical Oceanography ( )

3. outline River delivery of C and nutrients to coastal oceans
Global flux
Mississippi River flux
The effect of river loading on coastal metabolism
Biogeochemical responses in the river plume and nearby areas
Nutrient behavior
Biological production
Control on the CO2 system
Carbon budget in the Northern Gulf of Mexico

4. Global river delivery of C and nutrients to coastal oceans

5. Global riverine C, N, and P flux
DIC flux ~ TOC flux
Heterotrophic loading >> auto…
File copied from global_shelfC

6. Contrast in organic C input
deliver more heterotrophic loading to low-lat margins
Compiled by M. Dai and W-J. Cai;
also see Borges papers

7. DIN flux in world’s major rivers
Source: Dagg et al. 2004
Speculate: River loading drives tropical margins towards
heterotrophic, and mid lat margins to autotrophic.

8. Distribution of riverine HCO3- & Si flux
1. Latitudinal distribution in HCO3 flux
2. CaCO3 minerals in drainage basin
3. Contrast to silicate flux
From Cai et al. 2008
Continental Shelf Research 28:
Cai unpublished

9. Distribution of riverine HCO3- concentration
Maximum shifted to higher latitude as a result
of precipitation pattern
From Cai et al. 2008
Continental Shelf Research 28:

10. Influence on TAlk distr in Ocean Margins
Miss R (2500±500)
Amazon R
(300±50)
MAB
& North
SAB
( )

11. Mississippi River flux
Largest river basin in North America and third largest in the world
Drainage basin encompasses 41% of the lower 48 United States

12. DIC>>TOC 2. Heterotrophic loading ~ autotrophic loading
3. Disparate
time scales
Cai and Lohrenz (2008) in KK Liu et al book.

13. Biogeochemical responses in the river plume and nearby area —biological production
Satellite evidence points towards linkages between high chlorophyll and river outflow

14. Relationship Between River Inputs and Coastal Ecosystem Properties
3 May 2004
Satellite evidence points towards linkages between high chlorophyll and river outflow
Moderate Resolution Imaging Spectroradiometer (MODIS) true-color image (October 2006) of the northern Gulf of Mexico derived from data downloaded from NASA Ocean Color Home Page (Feldman and McClain, 2007) and processed using HDFLook (Version 5.0B, November 2004). The blue boxes show the boundaries of eastern and western regions of interest as discussed in text. The white crosses represent sites of in situ surface nutrient and chlorophyll sample collection. The blue boxes designate eastern (closer to the direct river outflow) and western regions.

15. Relationship Between River Inputs and Coastal Ecosystem Properties
Relationship between river DIN flux and satellite-derived chlorophyll
Eastern Box
Western Box
Western Box
Time series of DIN flux and SeaWiFS chlorophyll a product from Top panel corresponds to a box delimited by 29.0oN, 89.5oW and 28.5oN, 90.0oW. The bottom panel corresponds to a box delimited by 29.0oN, 90.0oW and 28.5oN, 90.5oW. The SeaWiFS chlorophyll values are the average over the specified regions of the 9 km 8-day composite product obtained using GIOVANNI (see methods for details). A Kruskal-Wallis test confirmed that the median chlorophyll in the western box was significantly lower than that in the eastern box (Chi-square statistic=46.74, p<0.0001).
Source: Lohrenz et al. (2008)

16. Biogeochemical responses — nutrients and CO2 system
The high nutrient loading and the short turnover time contribute to a strong biological pump for carbon uptake in the mid-field

17. pCO2 distribution

18. Salinity pCO2 (spring & summer)
Salinity explains some
but not all features
Inside river channels & embayment, low S high pCO2 (turbidity)
2. In the plume, low S
low pCO2 (bio uptake)
salinity
pCO2
3. Variability

19. Mississippi River plume, June 2003
Great DIC removal & nutrient removal at S=15.
No TA removal.
3. At S=15, maxDDIC ~ 430 uM. Appling a Redfield ratio of 6.6, we would predict a max NO3 removal of 65 uM.
(Rodney Powell)
(W.-J Cai)

20. Mississippi River plume, August 2004
DDIC = 320 uM, DTA=210 uM
DIC removal due to OC = /2 = 215 uM
Predicted NO3 removal = 33 uM

21. DIC and TAlk in the Mississippi River plume
DIC and TA removal

22. Net community production (NCP) & net calcification
Mixing layer depth and water residence
time taken from Green et al. (2006)

23. evidence of coccolithophores
Robert Stavn, Naval Research Laboratory
EDS
Summer 2004
SEM
Summer 2008
W-J Cai (unpub)

24. Significance of potential coccolith bloom in the Miss R plume
It increases pCO2 (opposite to diatom)
2HCO3- + Ca2+  CO2 + CaCO3
Diatom bloom creates a low pCO2, high pH & high CO32- condition for coccolith bloom (ecosystem succession)
No such response reported in the Amazon plume. Why? (ecosystem shift under anthropogenic pressure)

25. pCO2 in the Amazon plume
Cooley et al. GBC : 21, GB3014, doi: /2006GB002831, 2007

26. pCO2 in the Changjiang plume
The East
China Sea
salinity pH
pCO2
pCO2 in the East China Sea in summer 1998 during the great flood period
(Results from the Chinese JGOFS, L. Zhang pers. comm.)

27. pCO2 in the Pearl River plume
Dai et al. CSR (2008)

28. Carbon budget in the Northern GOM?
18 atm CO2
Basin process 1
N.GOM
PP=20?, R=1.15PP
NCP<-3.5 ?
NCP=5.66/2-4= -1.2
Turbidity plume
PP=5.7, R=0.6PP
NCP=2.3
CO2 (0.2?)
20%PP
15%PP
DIN =1.0
River input
DIC =21
TOC =5.0
DIC?
DIN?
DIC, DIN
(unit = 1012 gC/yr)

29. GOMECC-Mississippi Transect
Great DIC enrichment in the bottom water

30. Summary
Globally, river loadings are heterotrophic, in particular, this is the case for tropical rivers.
Large river plumes are autotrophic and a sink of CO2. This is particularly true for rivers with high anthropogenic DIN loading.
River HCO3- affects coastal TAlk distribution.
pCO2 in large river plumes has a great spatial and seasonal variability.
Potential coccolith bloom may be an important mechanism controlling surface water pCO2 for some large river plumes.

31. Thank you!

32. Distribution of riverine HCO3- concentration
The Mississippi River
The Changjiang
(Yangtze River)
Maximum shifted to north as a result
Of precipitation pattern
From Cai et al. 2008
Continental Shelf Research 28:

33. Mixing in the Amazon River plume
vs. in the MR plume
Amazon River water has low TA & low buffer capacity; in contrast, MR water has high TA and high buffer capacity.

34. October 2005
Ca2++CO32- = CaCO3
mixing
2:1
1:1 TA
DIC

35. Satellite image of CaCO3
Brown & Yoder, JGR-Ocean 1994

36. salinity pCO2 (late summer & fall)
Seasonal variation
Low pCO2 in spring and early summer
2. High CO2 in late summer and fall

37. CO2 sink Large River Margin ? ? River Plume Net Autotrophic
uptake by
planktons ? ?
Net Autotrophic
Low DICex
TOC, DIC, high DIN
River Inputs
CO2 deficit
High (labile) OCex
River Plume
NCP> 0 (net autotrophic) or GPP > R
OC exports > OC imports & DIC import > DIC export
Results: low DIC, & low pCO2