Volume loss from Antarctic ice shelves is accelerating328 17 APRIL 2015• VOL 348 ISSUE 6232 sciencemag.org SCIENCE Fig. 2. Variability in the rate of Antarctic ice-shelf thickness

17. J. A. Screen, Environ. Res. Lett. 8, 044015 (2013).18. S. Schubert, H. Wang, M. Suarez, J. Clim. 24, 4773–4792

(2011).19. J. Pedlosky, Geophysical Fluid Dynamics (Springer, New York,

1979).20. K. Fraedrich, H. Böttger, J. Atmos. Sci. 35, 745–750

(1978).21. G. J. Boer, T. G. Shepherd, J. Atmos. Sci. 40, 164–184 (1983).22. M. Blackmon, J. Atmos. Sci. 33, 1607–1623 (1976).23. D. Coumou, V. Petoukhov, A. V. Eliseev, Nonlinear Process.

Geophys. 18, 807–827 (2011).24. J. Lehmann, D. Coumou, K. Frieler, A. V. Eliseev, A. Levermann,

Environ. Res. Lett. 9, 084002 (2014).25. H. Teng, G. Branstator, H. Wang, G. Meehl, W. M. Washington,

Nat. Geosci. 6, 1–6 (2013).26. K. E. Trenberth, J. T. Fasullo, G. Branstator, A. S. Phillips,

Nat. Clim. Change 4, 911–916 (2014).27. D. P. Dee et al., Q. J. R. Meteorol. Soc. 137, 553–597

(2011).28. M. Murakami, Mon. Weather Rev. 107, 994–1013

(1979).

29. J. Lu, G. Chen, D. M. W. Frierson, J. Clim. 21, 5835–5851(2008).

30. J. P. Peixoto, A. H. Oort, Physics of Climate (American Instituteof Physics, New York, 1992).

31. T. Woollings, M. Blackburn, J. Clim. 25, 886–902 (2012).32. J. Kyselý, R. Huth, Theor. Appl. Climatol. 85, 19–36

(2005).33. J. Kyselý, Global Planet. Change 62, 147–163 (2008).34. R. Dole et al., Geophys. Res. Lett. 38, L06702 (2011).35. J. A. Screen, I. Simmonds, Nat. Clim. Chang. 4, 704–709

(2014).36. J. A. Screen, Nat. Clim. Chang. 4, 577–582 (2014).37. T. Schneider, T. Bischoff, H. Plotka, J. Clim. 28, 2312–2331

(2014).38. C. Huntingford, P. D. Jones, V. N. Livina, T. M. Lenton,

P. M. Cox, Nature 500, 327–330 (2013).

ACKNOWLEDGMENTS

We thank the CMIP5 climate modeling groups and the EuropeanCentre for Medium-Range Weather Forecasts and NCEP-NCARfor making their model and reanalysis data available. Comments

by three anonymous reviewers, S. Rahmstorf, and P. Eickemeier haveconsiderably improved the manuscript. Data presented in thismanuscript will be archived for at least 10 years by the PotsdamInstitute for Climate Impact Research. The work was supportedby the German Research Foundation (grant no. CO994/2-1)and the German Federal Ministry of Education and Research(grant no. 01LN1304A). D.C. designed the research; D.C., J.L.,and J.B. performed the analysis; and D.C., J.L., and J.B. wrotethe manuscript.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/348/6232/324/suppl/DC1Text S1 to S6Figs. S1 to S25Table S1ReferencesData Deposition

26 September 2014; accepted 26 February 2015Published online 12 March 2015;10.1126/science.1261768

ICE SHEETS

Volume loss from Antarctic iceshelves is acceleratingFernando S. Paolo,1* Helen A. Fricker,1 Laurie Padman2

The floating ice shelves surrounding the Antarctic Ice Sheet restrain the groundedice-sheet flow. Thinning of an ice shelf reduces this effect, leading to an increase in icedischarge to the ocean. Using 18 years of continuous satellite radar altimeter observations,we have computed decadal-scale changes in ice-shelf thickness around the Antarcticcontinent. Overall, average ice-shelf volume change accelerated from negligible loss at25 +– 64 cubic kilometers per year for 1994–2003 to rapid loss of 310 +– 74 cubic kilometersper year for 2003–2012. West Antarctic losses increased by ~70% in the past decade,and earlier volume gain by East Antarctic ice shelves ceased. In the Amundsen andBellingshausen regions, some ice shelves have lost up to 18% of their thickness in lessthan two decades.

The Antarctic Ice Sheet gains mass throughsnowfall and losesmass at itsmargin throughsubmarine melting and iceberg calving.These losses occur primarily from ice shelves,the floating extensions of the ice sheet.

Antarctica’s grounded-ice loss has increased overthe past two decades (1, 2), with the most rapidlosses being along the Amundsen Sea coast (3)concurrent with substantial thinning of adjoin-ing ice shelves (4, 5) and along the AntarcticPeninsula after ice-shelf disintegration events(6). Ice shelves restrain (“buttress”) the flow ofthe grounded ice through drag forces at the ice-rock boundary, including lateral stresses at side-walls and basal stresses where the ice shelf restson topographic highs (7, 8). Reductions in ice-shelf thickness reduce these stresses, leading toa speed-up of ice discharge. If the boundarybetween the floating ice shelf and the groundedice (the grounding line) is situated on a retro-

grade bed (sloping downwards inland), this processleads to faster rates of ice flow, with potential fora self-sustaining retreat (7, 9, 10).Changes in ice-shelf thickness and extent have

primarily been attributed to varying atmosphericand oceanic conditions (11, 12). Observing ice-shelf thickness variability can help identify theprincipal processes influencing how changinglarge-scale climate affects global sea level throughthe effects of buttressing on the Antarctic IceSheet. The only practical way tomap andmonitorice-shelf thickness for this vast and remote ice sheetat the known space and time scales of ice-shelfvariability is with satellite altimetry. Previousstudies have reported trends based on simple linefits to time series of ice-shelf thickness (or height)averaged over entire ice shelves or broad regions(4, 13) or for short (~5-year) time intervals (5, 14, 15).Here, we present a record of ice-shelf thicknessthat is highly resolved in time (~3 months) andspace (~30 km), using the longest available re-cord from three consecutive overlapping satelliteradar altimetermissions (ERS-1, 1992–1996; ERS-2,1995–2003; and Envisat, 2002–2012) spanning18 years from 1994 to 2012.

Our technique for ice-shelf thickness changedetection is based on crossover analysis of satelliteradar altimeter data, in which time-separatedheight estimates are differenced at orbit intersec-tions (13, 16, 17). To cross-calibrate measurementsfrom the different satellite altimeters, we used theroughly 1-year overlap between consecutive mis-sions. The signal-to-noise ratio of altimeter-derived height differences for floating ice inhydrostatic equilibrium is roughly an order ofmagnitude smaller than over grounded ice, re-quiring additional data averaging to obtain com-parable statistical significance. We aggregatedobservations in time (3-month bins) and space(~30-km cells). Because the spatial distributionof crossovers changes with time (due, for exam-ple, to nonexact repeat tracks and nadir mis-pointing), we constructed several records at eachcell location and stacked them in order to pro-duce a mean time series with reduced statisticalerror (18). We converted our height-change timeseries and rates to thickness changes by as-suming that observed losses occurred predomi-nantly at the density of solid ice (basal melting)(4, 5, 17). This is further justified by the relativeinsensitivity of radar measurements to fluctua-tions in surface mass balance (18). For volumechanges, we tracked the minimum (fixed) area ofeach ice shelf (18). We assessed uncertainties forall estimates using the bootstrap approach (re-sampling with replacement of the residuals ofthe fit) (19), which allows estimation of formalconfidence intervals. All our uncertainties arestated at the 95% confidence level [discussion ofuncertainties are provided in (18) and the severalcorrections applied are stated in (20)].We estimated 18-year trends in ice-shelf thick-

ness by fitting low-order polynomials (degreen ≤ 3) to the data using a combination of lassoregularized-regression (21) and cross-validation formodel-parameter selection (the shape of the fitis determined by the data). This combined ap-proach allowed us tominimize the effect of short-term variability on the 18-year trends. Relative toprevious studies (4, 5, 13, 22), we have improvedestimations by (i) using 18-year continuous re-cords, (ii) implementing a time series averaging

SCIENCE sciencemag.org 17 APRIL 2015 • VOL 348 ISSUE 6232 327

1Scripps Institution of Oceanography, University of California,San Diego, CA, USA. 2Earth & Space Research, Corvallis,OR, USA.*Corresponding author. E-mail: [email protected]

RESEARCH | REPORTSon M

ay 30, 2020

http://science.sciencemag.org/

Dow

nloaded from

328 17 APRIL 2015 • VOL 348 ISSUE 6232 sciencemag.org SCIENCE

Fig. 2. Variability in the rate of Antarctic ice-shelfthickness change (meters per year).Maps for (columnsfrom left to right) Filchner-Ronne, Amundsen, and Rossice shelves (locations in the bottom right corner) showingaverage rate of thickness change for (rows) four consec-utive 4.5-year intervals (1994–1998.5, 1998.5–2003, 2003–2007.5, and 2007.5–2012). Shorter-term rates can be higherthan those from an 18-year interval. Ice-shelf perimeters arethin black lines, and the thick gray line demarcates the limitof satellite observations.

Fig. 1. Eighteen years of change in thicknessand volume of Antarctic ice shelves. Rates ofthickness change (meters per decade) are color-coded from –25 (thinning) to +10 (thickening).Circles represent percentage of thickness lost(red) or gained (blue) in 18 years. Only signifi-cant values at the 95% confidence level areplotted (table S1). (Bottom left) Time series andpolynomial fit of average volume change (cubickilometers) from 1994 to 2012 for the West (inred) and East (in blue) Antarctic ice shelves. Theblack curve is the polynomial fit for All Antarcticice shelves. We divided Antarctica into eightregions (Fig. 3), which are labeled and delimitedby line segments in black. Ice-shelf perimetersare shown as a thin black line. The central circledemarcates the area not surveyed by the satel-lites (south of 81.5°S). Original data were inter-polated for mapping purposes (percentage areasurveyed of each ice shelf is provided in tableS1). Background is the Landsat Image Mosaic ofAntarctica (LIMA).

RESEARCH | REPORTSon M

ay 30, 2020

http://science.sciencemag.org/

Dow

nloaded from

scheme so as to enhance the signal-to-noiseratio, and (iii) using a robust approach to trendextraction.The 18-year average rate of thickness change

varies spatially (Fig. 1). On shorter time scales,trends are highly variable but spatially coherent(Fig. 2 and movie S1). We divided our data setinto eight regions on the basis of spatial coher-ence of long-term ice-shelf behavior and calcu-lated time series of ice-shelf thickness change(relative to series mean) for each region (Fig. 3).The largest regional thickness losses were in theAmundsen and Bellingshausen seas, with aver-age (and maximum) thinning rates of 19.4 T 1.9(66.5 T 9.0) m/decade and 7.4 T 0.9 (64.4 T 4.9)m/decade, respectively. These values correspondto ~8 and 5% of thickness loss over the 18 yearsfor the two regions, respectively. These tworegions account for less than 20% of the totalWest Antarctic ice-shelf area but, combined,contribute more than 85% of the total ice-shelfvolume loss from West Antarctica. The area-averaged time records of ice-shelf thickness andvolume for the West and East Antarctic sectors(Fig. 1, bottom left), broad regions (Fig. 3), andsingle ice shelves (fig. S1) at 3-month timeintervals show a wide range of temporal re-sponseswith large interannual-to-decadal fluctu-ations, stressing the importance of long recordsfor determining the long-term state of the iceshelves. Comparing our long records with simplelinear trends obtained for the periods of singlesatellite missions [such as the 5-year ICESat timespan used in (5)] shows that it is often not pos-sible to capture the persistent signals in the shorterrecords (Fig. 3 and fig. S1).Ice-shelf average thinning rates from the 18-year

polynomial fits in the Amundsen Sea region (AS)range from 1.5 T 0.9 m/decade for Abbot to 31.1 T5.4 m/decade for Crosson, with local maximumthinning of 66.5 T 9.0 m/decade on Getz (fig. S1and table S1). Crosson and Getz have lost ~18and 6% of their thicknesses, respectively, overthe 18-year period. If this thinning persists forthese two ice shelves, we can expect volumelosses of ~100 and 30%, respectively, in the next100 years. Getz is the single largest contributor tothe overall volume loss of Antarctic ice shelves,with an average change of –54 T 5 km3/year,accounting for ~30% of the total volume lossfrom theWest Antarctic ice shelves (table S1).Wefind the most dramatic thickness reduction onVenable Ice Shelf in the Bellingshausen Sea (BS),with an average (andmaximum) thinning rate of36.1 T 4.4 (64.4 T 4.9) m/decade, respectively (fig.S1 and table S1). This ice shelf has lost 18% of itsthickness in 18 years, which implies completedisappearance in 100 years.For the ice shelves in the AS, observed rates

are highest near the deep grounding lines, withlower rates found toward the shallower ice fronts(Fig. 2, table S1, and movie S1). This pattern isconsistent with enhanced melting underneaththe ice shelf forced by an increased flux of cir-cumpolar deep water (CDW) from across the con-tinental shelf and into the sub–ice-shelf cavity(12, 23, 24). The consequent loss of ice-shelf but-

tressing from increased ocean-forcedmeltingmayhave driven the grounding lines inland (25) toa point on a retrograde bed slope at which themarine ice-sheet instability mechanism can takeover the dynamics of ice export (7, 26). Hence,observed ice-shelf thinning reflects both ocean-induced basal melting and increased strain ratesresulting from faster flows. Our analysis showsthat thinning was already under way at a sub-stantial rate at the start of our record in 1994.On the eastern side of the Antarctic Peninsula

[comprising Larsen B (Scar Inlet remnant), Larsen

C, and Larsen D], the regional ice-shelf thinningrate of 3.8 T 1.1 m/decade (Fig. 3) is about half ofthat on thewestern side (BS) (Fig. 1). The onset ofthinning for Larsen C has progressed southward(Fig. 4), which is consistent with climate-drivenforcing discussed in earlier studies (22, 27). Thehighest thinning rates on Larsen C (with localmaximum thinning of 16.6 T 8.1 m/decade) arenear Bawden Ice Rise (Figs. 1 and 4). Assumingthat half of this observed thinning is due to airloss within the firn column, and consideringthat the ice shelf is ~40 m above flotation over

SCIENCE sciencemag.org 17 APRIL 2015 • VOL 348 ISSUE 6232 329

Fig. 3. Time series ofcumulative thicknesschange relative to seriesmean for Antarcticice-shelf regions(1994–2012). Time seriescorrespond to averagesfor all ice-shelf data withinthe Antarctic regionsdefined in Fig. 1. Dotsrepresent average thick-ness change every3 months. Error bars aresmall (in many cases,smaller than the symbolsthemselves, thus omittedfrom the plots), makingthe interannual fluctuationshown by the dots signifi-cant. The blue curve is thelong-term trend frompolynomial regressionwith the 95% confidenceband (18), and the red lineshows the regression lineto the segment of ourdata set that overlaps withthe period used for a priorICESat-based analysis(2003–2008) (5). Aver-age rates (in meters perdecade) are derived fromthe end points of thepolynomial models.

RESEARCH | REPORTSon M

ay 30, 2020

http://science.sciencemag.org/

Dow

nloaded from

the ice rise (28), we can expect Larsen C to fullyunground from this pinning point within thenext 100 years, with potential consequences onthe ice-shelf stability (29).The regional time-varying trends for the ice

shelves in the three East Antarctic regions (QueenMaud, Amery, and Wilkes) are coherent (Fig. 3).Ice shelves in the Wilkes region are challengingfor conventional radar altimeters because manyof them are small, contained in narrow embay-ments, and have rough surfaces so that altimeter-derived height changes do not necessarily reflectthickness change accurately. Our estimate of over-all thickness change for the Wilkes ice shelves is1.4 T 1.5 m/decade, which is not significantlydifferent from zero. The Queen Maud region iceshelves show an overall increase in thickness of2.0 T 0.8 m/decade.Like the AS ice shelves, Totten and Moscow

University ice shelves in the Wilkes region but-tress a large marine-based section of the EastAntarctic ice sheet so that their stability is poten-tially important to grounded-ice loss. Althoughthese ice shelves were previously reported asthinning (5) on the basis of a straight-line fit toa 5-year record from a satellite laser altimeter(ICESat, 2003–2008), our results show that thoseestimates are not representative of the longer-term trends (fig. S1B). Our estimate of thicknessloss during 2003–2008 is similar to the ICESat-based result, but the full 18-year period showsthickness trends that are not significantly differ-ent from zero (fig. S2).For most ice shelves, our estimates are signif-

icantly different from previous results (table S2).Several factors contribute to this. (i) The areas ofice shelves over which measurements are aver-aged vary between studies, affecting estimates onsmall ice shelves with large thickness-changesignals. (ii) Because of our grid resolution, iceshelf mask, and limited data coverage, we cannotsample near the grounding line of some ice shelves(such as Pine Island or Dotson); in such cases,

our estimated changes are likely to represent alower bound (changes could be larger). (iii) Radaraltimeters are less sensitive than are laser altim-eters to variations in surface mass balance owingto penetration of the radar signal into the firnlayer. (iv) Short records and previous trend-extraction approaches could not capture and ac-count for fluctuations in the underlying trend (fig.S3). This is the dominant factor affecting compar-isons between our results and previous studies.The total volume of East Antarctic ice shelves

increased during 1994–2003 by 148 T 45 km3/year,followed by moderate loss (56 T 37 km3/year),whereas West Antarctic ice shelves exhibitedpersistent volume loss over the 18 years, withmarked acceleration after 2003 (Fig. 1). Beforeand after 2003, this region lost volume by 144 T45 and 242 T 47 km3/year, respectively, corre-sponding to ~70% increase in the average lossrate. The total circum-Antarctic ice-shelf volumeloss was negligible (25 T 64 km3/year) during1994–2003 and then declined rapidly by 310 T74 km3/year after 2003. Overall, from 1994 to2012 Antarctic ice-shelf volume changed on aver-age by –166 T 48 km3/year, withmean accelerationof –31 T 10 km3/year2 (–51 T 33 km3/year2 for theperiod 2003–2012).We have shown that Antarctic ice-shelf vol-

ume loss is accelerating. In the Amundsen Sea,some ice shelves buttressing regions of groundedice that are prone to instability have experiencedsustained rapid thinning for almost two decades.If the present climate forcing is sustained, weexpect a drastic reduction in volume of the rap-idly thinning ice shelves at decadal to centurytime scales, resulting in grounding-line retreatand potential ice-shelf collapse. Both of these pro-cesses further accelerate the loss of buttress-ing, with consequent increase of grounded-icedischarge and sea-level rise. On smaller scales,ice-shelf thickness variability is complex, dem-onstrating that results from single satellite mis-sions with typical durations of a few years are

insufficient to draw conclusions about the long-term response of ice shelves. Large changes occurover a wide range of time scales, with rapid var-iations of ice-shelf thickness suggesting that iceshelves can respond quickly to changes in oce-anic and atmospheric conditions.

REFERENCES AND NOTES

1. A. Shepherd et al., Science 338, 1183–1189 (2012).2. T. C. Sutterley et al., Geophys. Res. Lett. 41, 8421–8428

(2014).3. I. Joughin, R. B. Alley, Nat. Geosci. 4, 506–513

(2011).4. A. Shepherd et al., Geophys. Res. Lett. 37, L13503

(2010).5. H. D. Pritchard et al., Nature 484, 502–505

(2012).6. T. A. Scambos, J. A. Bohlander, C. A. Shuman, P. Skvarca,

Geophys. Res. Lett. 31, L18402 (2004).7. C. Schoof, J. Geophys. Res. 112, F03S28 (2007).8. D. Goldberg, D. M. Holland, C. Schoof, J. Geophys. Res. 114,

F04026 (2009).9. L. Favier et al., Nat. Clim. Change 4, 117–121 (2014).10. I. Joughin, B. E. Smith, B. Medley, Science 344, 735–738

(2014).11. T. Scambos, C. Hulbe, M. Fahnestock, in Antarctic Peninsula

Climate Variability: Historical and PaleoenvironmentalPerspectives, E. Domack et al., Eds. (American GeophysicalUnion, Washington, DC, 2003), vol. 79, pp. 79–92.

12. P. Dutrieux et al., Science 343, 174–178 (2014).13. H. J. Zwally et al., J. Glaciol. 51, 509–527

(2005).14. E. Rignot, S. Jacobs, J. Mouginot, B. Scheuchl, Science 341,

266–270 (2013).15. M. A. Depoorter et al., Nature 502, 89–92 (2013).16. C. H. Davis, A. C. Ferguson, IEEE Trans. Geosci. Rem. Sens. 42,

2437–2445 (2004).17. D. J. Wingham, D. W. Wallis, A. Shepherd, Geophys. Res. Lett.

36, L17501 (2009).18. Materials and methods are available as supplementary

materials on Science Online.19. B. Efron, R. J. Tibshirani, An Introduction to the Bootstrap,

vol. 57 of Monographs on Statistics and Applied Probability(Chapman and Hall, New York, 1993).

20. Corrections include lag of the satellite’s leading-edge tracker(retracking), surface scattering variations, surface slope, dryatmospheric mass, water vapor, the ionosphere, solid Earthtide, ocean tide and loading, atmospheric pressure, andregional sea-level variation (18).

21. R. Tibshirani, J. R. Stat. Soc., B 58, 267–288(1996).

330 17 APRIL 2015 • VOL 348 ISSUE 6232 sciencemag.org SCIENCE

Fig. 4. Evolution of the rate of thicknesschange in theAntarctic Peninsula. Instantaneousrate-of-thickness change (meters per year) forfour specific times (1994, 1997, 2000, and 2008)is calculated as the derivative of the polynomial fitto the thickness-change time series. The rateincreases spatially with time from north to southin the Larsen Ice Shelf (movie S1). The eastern(Weddell Sea) side of the Antarctic Peninsula (top)shows independent behavior from the western(Bellingshausen Sea) side (bottom).

RESEARCH | REPORTSon M

ay 30, 2020

http://science.sciencemag.org/

Dow

nloaded from

22. H. A. Fricker, L. Padman, J. Geophys. Res. 117, C02026(2012).

23. S. S. Jacobs, A. Jenkins, C. F. Giulivi, P. Dutrieux, Nat. Geosci.4, 519–523 (2011).

24. M. Thoma, A. Jenkins, D. Holland, S. Jacobs, Geophys. Res.Lett. 35, L18602 (2008).

25. E. Rignot, J. Mouginot, M. Morlighem, H. Seroussi, B. Scheuchl,Geophys. Res. Lett. 41, 3502–3509 (2014).

26. J. Weertman, J. Glaciol. 13, 3–11 (1974).27. A. J. Cook, D. G. Vaughan, Cryosphere 4, 77–98

(2010).28. P. R. Holland et al., Cryosphere Discuss. 9, 251–299 (2015).

29. C. P. Borstad, E. Rignot, J. Mouginot, M. P. Schodlok,Cryosphere 7, 1931–1947 (2013).

ACKNOWLEDGMENTS

This work was funded by NASA [awards NNX12AN50H 002(93735A), NNX10AG19G, and NNX13AP60G]. This is ESRcontribution 154. We thank J. Zwally’s Ice Altimetry groupat the NASA Goddard Space Flight Center for distributing theirRA data sets for all satellite radar altimeter missions (http://icesat4.gsfc.nasa.gov). We thank C. Davis and D. Wingham forRA-processing advice. We thank A. Shepherd and anonymousreviewers for their comments on the manuscript.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/348/6232/327/suppl/DC1Materials and MethodsFigs. S1 to S4Table S1 and S2References (30–46)Movie S1

17 October 2014; accepted 11 March 2015Published online 26 March 2015;10.1126/science.aaa0940

OCEAN CHEMISTRY

Dilution limits dissolved organiccarbon utilization in the deep oceanJesús M. Arrieta,1,2* Eva Mayol,1 Roberta L. Hansman,3 Gerhard J. Herndl,3,4

Thorsten Dittmar,5 Carlos M. Duarte1,2,6

Oceanic dissolved organic carbon (DOC) is the second largest reservoir of organiccarbon in the biosphere. About 72% of the global DOC inventory is stored in deep oceaniclayers for years to centuries, supporting the current view that it consists of materials resistantto microbial degradation. An alternative hypothesis is that deep-water DOC consists of manydifferent, intrinsically labile compounds at concentrations too low to compensate for themetabolic costs associated to their utilization. Here, we present experimental evidenceshowing that low concentrations rather than recalcitrance preclude consumption of asubstantial fraction of DOC, leading to slow microbial growth in the deep ocean. Thesefindings demonstrate an alternative mechanism for the long-term storage of labile DOC inthe deep ocean, which has been hitherto largely ignored.

The accepted paradigm is that recalcitrant dis-solved organic carbon (DOC) is ubiquitousin the ocean and makes up the bulk of theDOC pool at depths of >1000 m and at DOCconcentrations below 42 mmol C liter−1 (1).

However, most of the components of the recalci-trant DOC pool remain unidentified (1, 2), andthere is little evidence of structural propertiesthat could make these compounds unavailableto microbial degradation. Conversely, the dilu-tion hypothesis (3, 4) postulates that most or-ganic substrates in the deep ocean are labile butcannot be used by prokaryotes at concentrationsbelow the levels matching the energetic invest-ment required for their uptake and degradation.

An early study (5) tested the dilution hypothesisby looking for microbial consumption in concen-trates of natural DOC from deep waters, butfound no substantial changes in DOC concentra-tions after a 2-month incubation. Those resultsled to the conclusion that deep-water DOC iscomposed of recalcitrant molecules and there-fore to the dismissal of the dilution hypothesis.Here, we revisit the dilution hypothesis usinga simple experimental approach, similar to thatused by Barber in 1968 (5) but using method-ologies not available at that time. Specifically,we tested the hypothesis that no significant in-crease in prokaryotic growth should be detectablewhen increasing DOC concentrations, which isas expected if deep oceanic DOC were struc-turally refractory.Natural prokaryotic communities collected at

14 stations between 1000 to 4200 m in the Pacificand Atlantic Ocean (fig. S1) were exposed toambient, 2-, 5- and 10-fold concentrations ofnatural DOC collected from their original loca-tion by means of solid phase extraction (6, 7) andincubated at in situ temperature. A consistentincrease in prokaryotic abundance over time wasobserved in response to increasing concentrationsof DOC in all 14 experiments (Fig. 1 and fig. S2).Maximum prokaryotic abundances obtained at~10-fold DOC concentrations were 3.6 to 11.7times higher than those observed in the corre-sponding controls (Fig. 1 and fig. S2). Unamended

controls showed much lower, sometimes undetec-table, prokaryotic growth, comparable with thevalues observed in deep layers of the ocean inother studies (8, 9), whereas specific growth ratesin the higher DOC enrichments showed valuesup to 0.4 days−1, which is typical of productivesurface waters (Fig. 2 and fig. S3). No significantdifferences (t test, P > 0.05) in prokaryoticgrowth were observed between unamended con-trols and extraction controls, confirming that theobserved growth was due to the materials beingextracted from seawater and not the result of acontamination with labile organics during theextraction procedure (fig. S4). The solid phaseextraction method used to concentrate naturaldissolved organic matter (DOM) may have intro-duced some compositional bias toward smalland polar compounds while losing a substan-tial part of the DOM pool (extraction efficiency~40%), but this does not change the fact thatincreasing the concentration of the extractablecomponents of natural DOC resulted in enhancedprokaryotic growth. Chemical alterations of DOC,such as the disruption of supramolecular arrange-ments (10) or mild hydrolysis produced duringthe concentration procedure, are an unlikely ex-planation for the observed response becausethe treatments in which DOC concentrationwas doubled by adding concentrated DOC showedlittle or no enhancement of prokaryotic growthas compared with that of controls. Hence, wevalidated the dilution hypothesis tested, show-ing that dilution limits C utilization in the deepocean.

Specific growth rates increased with DOCconcentrations following a classical Monod mod-el [coefficient of determination (R2) of 0.71 to0.98] at all the locations studied (Fig. 2, fig. S3,and table S2), further confirming the hypothesisthat prokaryotic growth in deep waters is limitedby the concentration of DOC. In 9 out of 14experiments, the initial in situ DOC concentra-tion was low enough to capture the lower partof the curve and thus to give an estimate of theminimum DOC concentration necessary to sup-port prokaryotic maintenance metabolism (Fig.2, fig. S3, and table S2). According to these esti-mates, concentrations of natural DOC below30.7 T 5.4 mmol C L−1 (average T SE, n = 9 mea-surements) would not be sufficient to supportprokaryotic metabolism, a value not significantlydifferent (t test, P > 0.05) from the lowest con-centrations of DOC around 34 mmol C liter−1

reported for the deep ocean (11).

SCIENCE sciencemag.org 17 APRIL 2015 • VOL 348 ISSUE 6232 331

1Department of Global Change Research, Institut Mediterranid’Estudis Avançats (IMEDEA), Consejo Superior deInvestigaciones Científicas (CSIC)/Universidad de las IslasBaleares (UIB), 07190 Esporles, Spain. 2Red Sea ResearchCenter, King Abdullah University of Science and Technology(KAUST), Thuwal 23955-6900, Kingdom of Saudi Arabia.3Department of Limnology and Bio-Oceanography, DivisionBio-Oceanography, University of Vienna, Althanstr. 14, 1090Vienna, Austria. 4Department of Biological Oceanography,Royal Netherlands Institute for Sea Research (NIOZ),1790AB Den Burg, Netherlands. 5Research Group for MarineGeochemistry (ICBM-MPI Bridging Group), Institute forChemistry and Biology of the Marine Environment (ICBM),Carl von Ossietzky University Oldenburg, and Max PlanckInstitute for Marine Microbiology (MPI), Bremen, Germany.6The UWA Oceans Institute, University of Western Australia(UWA), Crawley, WA, Australia.*Corresponding author. E-mail: [email protected]

RESEARCH | REPORTSon M

ay 30, 2020

http://science.sciencemag.org/

Dow

nloaded from

Volume loss from Antarctic ice shelves is acceleratingFernando S. Paolo, Helen A. Fricker and Laurie Padman

originally published online March 26, 2015DOI: 10.1126/science.aaa0940 (6232), 327-331.348Science 

, this issue p. 327Sciencetheir disappearance may allow land-based ice to collapse and melt.climate continues to warm. If warming continues to cause ice shelves to thin, as they have for the past couple of decades,around the edge of the continent are losing mass. This result increases concern about how fast sea level might rise as

present satellite data showing that ice shelves in many regionset al.into the sea, are thinning at faster rates. Paolo The floating ice shelves around Antarctica, which buttress ice streams from the continent and slow their discharge

Disappearing faster around the edges

ARTICLE TOOLS http://science.sciencemag.org/content/348/6232/327

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2015/03/25/science.aaa0940.DC1

REFERENCES

http://science.sciencemag.org/content/348/6232/327#BIBLThis article cites 40 articles, 4 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Copyright © 2015, American Association for the Advancement of Science

on May 30, 2020

http://science.sciencem

ag.org/D

ownloaded from