Growth rates of Florida corals from 1937 to 1996 and their response to climate change

Abstract

Ocean acidification causes declines in calcification rates of corals because of decreasing aragonite saturation states (Ω_arag ). Recent evidence also indicates that increasing sea surface temperatures may have already reduced growth and calcification rates because of the stenothermic threshold of localized coral populations. Density banding in coral skeletons provides a record of growth over the coral's lifespan. Here we present coral extension, bulk density and calcification master chronologies from seven subtropical corals (Montastraea faveolata) located in the Florida Keys, USA with a 60-year common period, 1937-1996. Linear trends indicate that extension increased, density decreased and calcification remained stable while the most recent decade was not significantly different than decadal averages over the preceding 50 years for extension and calcification. The results suggest that growth rates in this species of subtropical coral have been tolerant to recent climatic changes up to the time of collection (1996).

ARTICLE

Received 26 Aug 2010 | Accepted 2 Feb 2011 | Published 1 Mar 2011 DOI: 10.1038/ncomms1222

Kevin P. Helmle1,2,3, Richard E. Dodge1, Peter K. Swart4, Dwight K. Gledhill2,3 & C. Mark Eakin5

Ocean acidication causes declines in calcication rates of corals because of decreasing aragonite saturation states (arag). Recent evidence also indicates that increasing sea surface temperatures may have already reduced growth and calcication rates because of the stenothermic threshold of localized coral populations. Density banding in coral skeletons provides a record of growth over the corals lifespan. Here we present coral extension, bulk density and calcication master chronologies from seven subtropical corals (Montastraea faveolata) located in the Florida Keys, USA with a 60-year common period, 19371996. Linear trends indicate that extension increased, density decreased and calcication remained stable while the most recent decade was not signicantly different than decadal averages over the preceding 50 years for extension and calcication. The results suggest that growth rates in this species of subtropical coral have been tolerant to recent climatic changes up to the time of collection (1996).

Growth rates of Florida corals from 1937 to 1996 and their response to climate change

1 Oceanographic Center, National Coral Reef Institute, Nova Southeastern University, 8000 N Ocean Drive, Dania Beach, Florida 33004, USA. 2 Cooperative Institute for Marine and Atmospheric Studies, Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, Florida 33149, USA. 3 Ocean Chemistry Division, National Oceanic and Atmospheric Administration, Atlantic Oceanographic and Meteorological Laboratory, Ocean Chemistry Division, 4301 Rickenbacker Causeway, Miami, Florida 33149, USA. 4 Division of Marine Geology and Geophysics, Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, Florida 33149, USA. 5 NOAA Coral Reef Watch, Division of Satellite Applications and Research Satellite Oceanography and Climate, National Oceanic and Atmospheric Administration Center,

E/RA31, SSMC1, 1335 East West Hwy, Silver Spring, Maryland 20910, USA. Correspondence and requests for materials should be addressed to

K.P.H. (email: [email protected]).

NATURE COMMUNICATIONS | 2:215 | DOI: 10.1038/ncomms1222 | www.nature.com/naturecommunications

2011 Macmillan Publishers Limited. All rights reserved.

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms1222

Ocean acidication due to rising atmospheric carbon dioxide (pCO2,air) is cited as a threat to calcifying marine organisms110. When ocean acidication is combined with physiologi

cal stress from rising sea surface temperatures (SSTs), shis in community structure and accelerating degradation of coral reef ecosystems may result1114. The impacts of ocean acidication and SST on calcication rates remain topics of some debate1520. The projected trend of future ocean acidication and SST rise21,22 is expected to increasingly inhibit the calcifying potential of marine organisms in the near future16. However, taxa-specic calcication response23,24 and the spatial and temporal variability in ocean acidication6,2527 are likely to regionally modify predicted declines in calcication rates.

Historical declines in calcication are estimated at ~10% for the twentieth century with declines in calcication predicted to double during the twenty-rst century10,28. Many experimental studies demonstrate that reductions in pH and carbonate ion (CO32 ) concentration (similar to that expected because of increasing pCO2,air)

cause substantial and predictable declines in coral calcication29,30. These laboratory results are based on constraining limiting factors to known values of pH, aragonite saturation states (arag) or concentration of CO32 . In situ coral calcication is subject to complex interactions of limiting environmental variables such as temperature, light, nutrients and water chemistry; however, comparably few laboratory studies account for these interactions3134.

Cyclic variations in skeletal porosity of many corals produce annual density bands35 that provide a record of linear extension, bulk density and calcication. Identifying limiting environmental variables of coral growth is an aim of eective coral sclerochronology; however, individual variables commonly explain only a portion of the annual variability in growth rates. Multidecadal growth records from the Pacic spanning the twentieth century initially demonstrated a positive linear relationship between annual calcication and SST across a latitudinal range36 and over time (through 1990), with SST accounting for roughly 30% of the annual variability and indicating a ~4% increase in calcication per 1 C3739. More recently, declines in extension and calcication over the past two decades (ca 19852005) from the Pacic, Andaman Sea and Red Sea have primarily been attributed to increased SST suggesting the relationship is nonlinear, where both low and high temperatures are capable of reducing growth rates3943. In contrast, a linear negative relationship between extension rate and SST was reported for a 440 years coral record from the Bahamas and accounted for 45% of the variability in extension rates based on 6-year timescales44.

Declines in annual density and calcication were reported from 1934 to 1982 based on 35 Porites corals from the Great Barrier Reef (GBR)38; however, more recent temporal models, constructed from 328 GBR corals, indicated that extension and calcication increased from 1900 to 1970 and rst began to decline from 1990 to 200541. No signicant trend was present in extension or calcication from 1801 to 1990 for a single Porites coral from French Polynesia37. Declines in extension (30%) and calcication (18%) were reported in Red Sea corals (Diploria heliopora) from 1998 to 2008 relative to the 1976 1997 mean values43. Examination of long-term coral growth records from the literature suggest that growth trends have not consistently exhibited the long-term linear declines in calcication expected from ocean acidication over the extent of twentieth century and that the measured declines in growth have been limited to recent decades and largely attributable to increasing SST.

In this paper, we report on the trends, variability and decadal averages of historical coral growth records in relation to climate change and ocean acidication. The growth data are analysed against environmental variables, climatic/oceanic indices and modelled estimates of arag. The historical changes and projected trends in SST and arag are thought to have already limited coral growth or soon to be limiting controls on coral growth rates. Although declining coral

growth is becoming more apparent world-wide, these growth data from subtropical reefs of the Florida Keys indicate a unique historical response of sustained calcication rates and increasing extension rates up to 1996.

ResultsGrowth rates of Florida corals. Annual extension, density and calcication data are presented from seven coral cores of Montastraea faveolata colonies collected in the upper Florida Keys, USA, in May 1997 and June 1998. Over the 60-year period, 19371996, mean annual linear extension is 0.79 0.07 cm per year, mean annual bulk density is 1.18 0.06 g cm 3 and mean annual calcication is 0.91 0.06 g cm 2 per year. The coefficient of variation (CV) demonstrates the variability of annual values over time (inter-annual variability) irrespective of dierences in the means. The CV in extension was 20% (range of individual cores: 1524%), density was 11% (716%) and calcication was 17% (1423%). The CV of growth parameters are similar to those for Porites from the GBR, in which density (69%) is the most conservative and extension (2831%) and calcication (2830%) are more variable38,40. The combination of extension, density and calcication measurements provides the most complete perspective of coral growth variability45. The relationship between the growth parameters over the 60-year record indicates that extension was negatively related to density (r = 0.72, P < 0.01, n = 60), extension was positively related to calcication (r = 0.72, P < 0.01, n = 60) and no signicant relationship was present between calcication and density (r = 0.07, P = 0.62, n = 60), which is in agreement with other coral growth studies36,38,45,46.

Trends based on the linear regressions of the standardized anomaly data (STDA; Fig. 1) indicated a signicant increase in extension rates (P = 0.048, n = 60), a signicant decrease in skeletal density (P < 0.001, n = 60) and no signicant change in calcication rates (P = 0.395, n = 60). Coefficients of determination (r2) for linear regression of the growth data are low because of the high inter-annual variability of growth parameters and indicated that a linear regression poorly explained the annual variability in growth parameters. For this reason, it is useful to assess the growth data as decadal averages (Fig. 2). No signicant dierences were present between the decadal averages beginning with 19371946 and ending with 19871996 for extension or calcication rates (single-factor analysis of variance at P < 0.05, n = 10). Signicant dierences were present in decadal averages of density data; most notably, the decade 19771986 was signicantly lower than the other ve decades (t-test at P < 0.01, n = 10), whereas density of the most recent decade 19871996 was not signicantly lower than the preceding ve decades.

Coral growth and environmental correlates. Coral growth master chronologies are compared with environmental and climatic records consisting of SST, rainfall, modelled arag, Atlantic multidecadal oscillation (AMO)47 and North Atlantic oscillation (NAO)48 indices. Correlation of growth parameters with SST indicated signi-cant positive relationships between calcication and annual average temperature and minimum monthly temperature, but no signicant relationship to maximum monthly temperature (Table 1). Annual SST for the study site exhibited no signicant linear trend over the period 19371996 (r2 = 0.05 P = 0.087, n = 60). Figure 3 illustrates the intermittent agreement between annual calcication anomaly and annual temperature anomaly. Rainfall was not signicantly related to extension, density and calcication, nor was the NAO signicantly correlated to any of the growth parameters on annual scales.

Modelled values of aragonite saturation state are presented in Figure 4 along with the predicted response in calcication based on an assumption of 100% calcication in 1937, which declines by 13% through 1996. The measured calcication data (Fig. 1) failed to demonstrate this decline and was not signicantly correlated with arag on an annual basis (r = 0.09, P = 0.517, n = 60). Extension

NATURE COMMUNICATIONS | 2:215 | DOI: 10.1038/ncomms1222 | www.nature.com/naturecommunications

2011 Macmillan Publishers Limited. All rights reserved.

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms1222

ARTICLE

a

y = 0.0065x 12.699 r 2 = 0.07, P = 0.048

y = 0.0127x + 24.89 r 2 = 0.23, P < 0.001

y = 0.0028x + 5.4293 r 2 = 0.01, P = 0.395

1.0

19371946

3

2

0.5

1

STDA

0

STDA

0.0

1

2

0.5

3

1935

1940

1950

1945

1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

1.0

19471956

19571966

19671976

19771986

19871996

b

3

2

1

STDA

0

Figure 2 | Decadal coral growth averages. Decadal averages from master chronologies of annual extension in green, density in blue and calcication in red are presented in standardized anomaly (STDA) units with error bars representing the standard deviation for each decade (n = 10).

1

2

3

1935

1940

1950

1945

1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

Table 1 | Correlation of growth parameters with environmental records.

Extension Density Calcication

SST mean r=0.07, P=0.586 r=0.23, P=0.077 r=0.28, P=0.029 SST max r=0.05, P=0.682 r=0.17, P=0.190 r=0.22, P=0.094 SST min r=0.09, P=0.473 r=0.21, P=0.109 r=0.27, P=0.036 Rainfall r= 0.05, P=0.693 r=0.10, P=0.485 r=0.10, P=0.486 arag r= 0.25, P=0.057 r=0.42, P < 0.001 r=0.09, P=0.517

AMO r= 0.25, P=0.050 r=0.57, P < 0.001 r=0.19, P=0.151 NAO r=0.18, P=0.164 r= 0.04, P=0.740 r=0.25, P=0.053

Abbreviations: arag, Aragonite saturation state; AMO, Atlantic multidecadal oscillation; NAO,

North Atlantic oscillation; SST, sea surface temperature.

Correlation for period 19371996, n=60, bold values signicant at P < 0.05.

c

3

2

1

STDA

0

1

2

1935

3 1940

1950

1945

1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

Figure 1 | Coral growth master chronologies 19371996. Master chronologies (n = 7 corals) are presented in standardized anomaly (STDA) units for the period of 19371996 and depicted by bold black lines withthe individual coral records comprising the master chronology depictedby grey lines. Linear regression lines for the master chronology data are depicted by red lines. Annual linear extension (a) exhibits a slight, but signicant increasing trend. Annual skeletal bulk density (b) exhibits a signicant decreasing trend. Annual calcication (c) exhibits no signicant trend over the 60-year period.

was negatively related to arag, though non-signicantly (r = 0.25, P = 0.057, n = 60). Density was the only growth parameter that exhibits a signicant correlation with arag (r = 0.42, P < 0.001, n = 60).

The AMO exhibited a signicant positive relationship with density (r = 0.57, P < 0.001, n = 60) explaining 33% of the annual variability. The AMO was negatively related to extension (r = 0.25, P = 0.050, n = 60) and positively related to calcication (r = 0.19, P = 0.151, n = 60). The relationship between extension, density and the AMO is presented in Figure 3 and illustrates the positive relationship between the density anomaly values (negative relationship for extension) and the positive phase (19371965) and negative phase (19651996) of the AMO.

Discussion

The growth records from these subtropical corals do not exhibit signicant declines in calcication or extension on the basis of linear trends or decadal averages. Age eects on coral growth39, on the basis of 43 Porites records over at least 100 years, indicate that extension may increase by time and density and may decrease by time independent of environmental trends such as increasing SST or decreasing arag; whereas, calcication exhibits no detectable age eect through time. In this study, annual extension exhibited a signicant increase and annual density exhibited a signicant decrease over the 60-year period based on linear trends. However, following methods for determining age eects39, the 10-year

averages of the seven cores in this study with start years between 1906 and 1937, show no evidence of signicant age eects over a 60-year period for extension (y = 0.0072x + 0.7635, r2 = 0.388, P = 0.186), density (y = 0.0041x + 1.211, r2 = 0.0822, P = 0.582) or calcication (y = 0.0023x + 0.9133, r2 = 0.0282, P = 0.750).

Considering the annual variability in extension and calcication of these coral records, change over time is best assessed by evaluating the combination of long-term variability, trends and decadal averages, rather than the dierence between the mean values of individual years. For example, the dierence in annual calcication between the years 1937 and 1996 represents a 15% decline in calcication; however, this change is neither supported by a signicant linear trend nor by signicant dierences between decadal averages. The lack of declines in extension or calcication does not necessarily suggest that corals or coral reefs at this study site are healthy49, but rather demonstrates that the measured corals have historically been able to maintain rates of extension and calcication over the 60-year period from 1937 to 1996 under the combination of local environmental and climatic changes. The subtropical location of these reefs may contribute to their persistent ability to grow and calcify under less than optimal environmental conditions. The lack of decline in coral extension and calcication rates raises important questions as to whether these corals exhibit a particularly high tolerance to changing conditions, whether the subtropical location of these corals is not representative of global estimates of past environmental or climatic change and variability, or whether environmental variables have yet to cross critical thresholds for sustaining coral growth and calcication.

Calcication rates were positively related to SST, similar to results of Porites corals from Tahiti37 and the GBR38; however, SST only explained about 7% of the annual variability in calcication of these corals as opposed to ~30% for the Pacic locations. A negative response to high SST3943 was not present up to 1996, although this

NATURE COMMUNICATIONS | 2:215 | DOI: 10.1038/ncomms1222 | www.nature.com/naturecommunications

2011 Macmillan Publishers Limited. All rights reserved.

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms1222

a

1.0

4.7

1935

1940

1945

1950

1955

1960

1965

1970

105

4.6

0.5

4.5

100

STDA

0.0

4.4

arag

4.3

95

0.5

4.2

% Calcification

1.0

90

4.1

1935

1940

1945

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

4.0

85

3.9

3.8

80

b

1.0

1975

1980

1985

1990

1995

2000

0.5

Figure 4 | Aragonite saturation state and calcication. Modelled aragonite saturation state, arag, (left y axis) is presented for the period of the coral growth record (19371996) encompassing the study site (2425 N and 8081 W) along with the calculated percent change in calcication (right y axis) assuming 100% calcication in 1937.

STDA

0.0

0.5

1.0

1935

1940

1945

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

Figure 3 | Coral growth and environmental correlates. Coral growth parameters are plotted against environmental correlates. In a, annual calcication is depicted in red as standardized anomaly (STDA) unitsand is positively related to the annual sea surface temperature anomaly depicted in grey. In b, annual density, depicted in blue, is positively related to the AMO index, depicted in grey, whereas annual extension, depicted in green, is negatively related to the AMO index.

does not preclude a potential negative response associated with the 19971998 mass-bleaching event that aected corals in this region50. Although the ability of these subtropical corals to calcify and grow has not been limited as of 1996, it is important to note that extreme events such as bleaching or disease outbreaks have led to widespread mortality in spite of otherwise sustainable growth conditions.

The lack of a signicant positive relationship between arag and extension or calcication, or signicant declines in extension and calcication, suggest that estimated declines in arag have not, as of 1996, measurably limited the ability of these subtropical corals to accrete skeletal material (calcication) or construct new skeletal structures (extension). The decline in skeletal density over this period may suggest changes in how the corals are constructing the skeleton. The inter-annual variability of arag (CV = 3%) does not appear to be a dominant limiting environmental factor controlling the large annual variations in growth parameters (CV = 1117%). It is useful to consider that the modelled arag data are based on oceanic parameters, which could dier substantially within the coastal zone and may not accurately describe the near-reef carbonate chemistry dynamics; nevertheless, the lack of declines in extension and calcication suggests that the global oceanic secular declines of arag

are not reected in the calcication and growth of these subtropical corals over time.

The positive relationship between density and arag could suggest a weakening of coral skeleton in response to declines in arag

over time; however, a corresponding reduction in calcication is not present. Skeletal density is the quotient of variations in the amount of skeletal accretion (calcication) and the rate of upward growth (extension). Considering the opposite linear trends in density and extension and the lack of signicant trend in calcication, it seems that the decline in density is the result of increasing extension over time, as opposed to potential declines in the calcication of these corals.

A stretching modulation response51, in which corals exposed to higher environmental stress respond by increasing extension although maintaining calcication with a resulting decrease in skeletal density, may provide an alternative interpretation to the increased extension. However, the concept of stretching modulation

was based on the response of corals, from dierent locations, to increasing stress along a spatial gradient, whereas the corals in this study exhibit changes in extension and density on a temporal basis. It could be inferred that increasing extension and decreasing density indicate a response to potentially increasing environmental stress (that is, declining arag) over time. If this stress response was responsible for the linear trends over time, it would be expected that the most recent decade (19871996) would have the highest extension rate and the lowest density; however, as illustrated in Figure 2, this is not the case.

Inuence of the AMO on surface climate and weather varies regionally, but a positive phase is generally characterized by warmer temperatures and increased rainfall over Florida along with increased Atlantic hurricane activity47,52. The high correlation between density and the AMO might suggest that large-scale oceanic and climatic cycles are driving the variability in extension and calcication based on complex interactions of temperature, rainfall and related aects from cloud cover, storm activity and light availability that can vary on seasonal, annual and decadal timescales, and which are ultimately reected in the skeletal density. Because of the multidecadal cycle of the AMO, it is not possible to test the agreement of multiple phases using a 60-year coral record; however, considering that the AMO shied back to a positive phase in the mid 1990s, it may be posited that growth parameters would return to levels similar to those from 1937 to 1957 (that is, higher skeletal density, lower extension rates and neutral to higher calcication rates) and on the basis of the decadal growth averages this seems to be a reasonable hypothesis.

Previous coral growth studies53,54 in the upper Florida Keys consisted only of extension rates and generally agree with the relationship to AMO cycles presented in this paper. Extension rate data from Biscayne National Park54 were at or below average from 1860 to 1880 (positive AMO), above average from 1880 to 1940 (negative AMO), below average and declining from 1940 to 1965 (positive AMO) and below average, but increasing from ~1965 to the end of the record in 1984 (negative AMO). Extension data from Key Largo were reported as 5-year averages53 and exhibited a general decreasing trend from 1953 to 1968 (positive AMO) and a general increasing trend from 1968 to 1978 (negative AMO). However, without accompanying calcication and density data, which exhibits the highest correlation to the AMO in the present study, it is difficult to draw strong conclusions about AMO and coral growth relationships.

Considering the potential threats from rising temperatures and ocean acidication, there are two fundamental explanations for the increasing extension and stable calcication over the 19371996 period at this location: the degree of change in SST, rainfall, arag and

AMO at this study site was insufficient to limit the growth potential

NATURE COMMUNICATIONS | 2:215 | DOI: 10.1038/ncomms1222 | www.nature.com/naturecommunications

2011 Macmillan Publishers Limited. All rights reserved.

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms1222

ARTICLE

of these corals or the measured corals were capable of tolerating these changes in environmental and climatic variables. The AMO represented the strongest environmental correlate with coral growth parameters and it would be useful to assess this relationship over multi-century coral records, which pre-date recent global warming and rise in carbon dioxide to determine the potential long-term natural variability associated with the AMO. Relative to ocean acidication, these Florida Keys corals maintained their rates of calcication through 1996, despite the expected declines in calcication based on modelled arag. Explanations for the lack of decline in calcication over time as expected from ocean acidication could be that: massive reef-building corals are not as susceptible to declines in arag (ref. 55) as demonstrated by laboratory experiments; local process, such as high seasonal variation in arag in the Florida Keys25, may be temporarily enabling these corals to maintain their historical rates of calcication; the role of arag in controlling calcication is masked amidst considerable natural inter-annual variability; or the actual in situ reef-site carbonate chemistry is decoupled from the oceanic values, which could occur as a result of shis in benthic community metabolism, mineral buering and/or coastal biogeochemical processes. Ongoing retrospective monitoring of coral growth rates from existing cores and updated coral core records is necessary to better identify growth rate changes over time associated with potential local variability across the Florida reef tract and ultimately to identify large-scale variability throughout the western Atlantic, Caribbean and Gulf of Mexico.

Methods

Coral samples and X-radiography. Coral cores collected in May 1997 and June 1998 from seven Montastraea faveolata colonies located in the upper Florida Keys, USA, (2425 N and 8081 W) were drilled at the location of maximum vertical growth of the colony using a hydraulic drill with 10-cm-diameter barrel. Cores were sectioned into 0.40.7 cm-thick parallel-sided slabs oriented along the primary growth axis, based on the orientation of corallites on the sample surface. Slabs were X-rayed on Kodak Industrex AA400 810 Ready-pack II lm (Kodak) using a self-rectied dental X-ray machine with settings of 70 kV and 15 mA, a source to object distance of 1.5 m and exposure times of 510 s. Films were manually developed and digitized with a medical lm scanner at the native resolution of the CCD, which was 357 d.p.i. Digitized X-radiographs were assembled into a mosaic image for the entire core. Chronologies were constructed from the annual density bands and it was determined that all cores had a 60-year common period from 1937 to 1996.

Coral densitometry and growth measurements. Coral growth parameters were measured from digitized X-radiographs based on a process of relative optical densitometry45,56 using a computer programme, CoralXDS + (http://www.nova. edu/ocean/coralxds/index.html). Relative optical densitometry is based on the premise that when two materials (for example, a coral sample and an aluminium standard) are X-rayed and the exposed lm exhibits equal optical density for both materials, the density of the sample material can be calculated56,57 from:t r h t r h

ms mst st st

s s =

where ms is the relative mass absorption coefficient of the sample, mst is the relative mass absorption coefficient of the standard, s is the density of the sample, st is the density of the standard, s is the thickness of the sample and st is the thickness of the standard. Solving for the density of the coral sample yields

r tt r

h h

height = 0.9975 cm and slope = 0.1743) was included in all X-radiographs and used for density calibration.

Using the CoralXDS + program, a transect running normal to the orientation of density bands was selected on the coral X-radiograph image. The pixels within the transect were converted to skeletal bulk density, averaged over the width of the transect and plotted as a function of distance along the length of the transect. High-density, low-density and annual bands were determined by the second derivative zero crossing (inection point) of a cubic spline of the density transect data. Bulk density was measured in g cm 3, linear extension was measured in cm per year and calcication, in g cm 2 per year, was calculated from the product of extension and density. Master chronologies of extension, density and calcication were constructed from STDA in which the mean growth over the common period (19371996) was subtracted from annual values and then divided by the standard deviation of the common period for each individual core.

Environmental records. SST for the period of 19371996 was obtained from the 11 gridded HadISST 1.1 dataset centered on 24.5 N and 80.5 W, which consist of blended in situ observations and satellite AVHRR (UK Meteorological Office, Hadley Centre. HadISST 1.1Global sea-Ice coverage and SST, 1870-Present, British Atmospheric Data Center, 2006, http://badc.nerc.ac.uk/data/hadisst/). Annual and seasonal rainfall records were constructed from daily rainfall data collected at Tavernier, USA (24 59 N, 80 31 W) for the period of 19371996 (South Florida Water Management District, www.sfwmd.gov). arag from 1937 to 1996 is derived using the OAPS model58 applied to HadISST 11 SST and WOA05 monthly salinity climatologies and is driven using global atmospheric CO2 estimates provided by Scott Doney (WHOI) with the predicted calcication response based on the rst-order empirical relationship of Langdon and Atkinson31 in which calcication rate (G, mmol CaCO3 m 2 h 1) = (8 1)(arag 1). Climate time series data were obtained from NOAAs National Climate Data Center (www.ncdc.noaa.gov) and Earth System Research Laboratory (www.cdc.noaa.gov). The AMO index is constructed from the Kaplan SST dataset47 and the NAO index is constructed from the dierence between instrumental pressure observations between Gibraltar, Azores and southwest Iceland48.

References

1. Orr, J. C. et al. Anthropogenic ocean acidication over the twenty-rst century and its impact on calcifying organisms. Nature 437, 681686 (2005).

2. Caldeira, K. & Wickett, M. E. Ocean model predictions of chemistry changes from carbon dioxide emissions to the atmosphere and ocean. J. Geophys. Res. (Oceans) 110, 112 (2005).

3. Sabine, C. L. et al. The oceanic sink for anthropogenic CO2. Science 305, 367371 (2004).

4. Feely, R. A. et al. Impact of anthropogenic CO2 on the CaCO3 system in the oceans. Science 305, 362366 (2004).

5. Hughes, T. P. et al. Climate change, human impacts, and the resilience of coral reefs. Science 301, 929933 (2003).

6. Guinotte, J., Buddemeier, R. & Kleypas, J. Future coral reef habitat marginality: temporal and spatial eects of climate change in the Pacic basin. Coral Reefs 22, 551558 (2003).

7. Guinotte, J. M. & Fabry, V. J. Ocean acidication and its potential eects on marine ecosystems. Ann. N. Y. Acad. Sci. 1134, 320342 (2008).

8. Fabry, V. J., Seibel, B. A., Feely, R. A. & Orr, J. C. Impacts of ocean acidication on marine fauna and ecosystem processes. ICES J. Mar. Sci. 65, 414432 (2008).

9. Fabricius, K. Theme section on Ocean Acidication and Coral Reefs. Coral Reefs 27, 455457 (2008).

10. Kleypas, J. A. et al. Geochemical consequences of increased atmospheric carbon dioxide on coral reefs. Science 284, 118120 (1999).

11. Carpenter, K. E. et al. One-third of reef-building corals face elevated extinction risk from climate change and local impacts. Science 321, 560563 (2008).12. Silverman, J., Lazar, B. & Erez, J. Eect of aragonite saturation, temperature, and nutrients on the community calcication rate of a coral reef. J. Geophys. Res. (Oceans) 112, 05004 (2007).

13. Hoegh-Guldberg, O. et al. Coral reefs under rapid climate change and ocean acidication. Science 318, 17371742 (2007).

14. Mumby, P. J., Hastings, A. & Edwards, H. J. Thresholds and the resilience of Caribbean coral reefs. Nature 450, 98101 (2007).

15. Kleypas, J. A. et al. Comment on Coral reef calcication and climate change: The eect of ocean warming. Geophy. Res. Lett. 32, L08601 (2005).

16. McNeil, B. I., Matear, R. J. & Barnes, D. J. Coral reef calcication and climate change: the eect of ocean warming. Geophy. Res. Lett. 31, 14 (2004).

17. McNeil, B. I., Matear, R. J. & Barnes, D. J. Reply to comment by Kleypas et al. on Coral reef calcication and climate change: the eect of ocean warming. Geophys. Res. Lett. 32, 13 (2005).

18. Pelejero, C. et al. Preindustrial to modern interdecadal variability in coral reef pH. Science 309, 22042207 (2005).

19. Matear, R. J. & McNeil, B. I. Comment on Preindustrial to modern interdecadal variability in coral reef pH. Science 314, 595 (2006).20. Pelejero, C. et al. Response to comment on Preindustrial to modern interdecadal variability in coral reef pH. Science 314, 595c (2006).

(1)(1)

(2)(2)

(3)(3)

s = mst

ms

st st

s

For a given working range of exposure parameters, mst/ms can be converted to a constant relative mass absorption coefficient ratio and the value mr substituted:

r t r hh

s

= mr st st

Hence, the density of the coral sample (skeletal bulk density) can be determined if the relative mass absorption coefficient ratio, density of the standard and thicknesses of the sample and standard are known. The mr was determined to be 0.622 by plotting relative optic density curves along the lengths of an aluminium wedge (Al) and a Tridacna aragonite wedge (Ar) against and determining the mr

value that produced the lowest residual dierence, calculated as |(mrAl Ar)|.

The aluminium wedge standard (density = 2.7029 g cm 3, length = 5.1629 cm,

s

NATURE COMMUNICATIONS | 2:215 | DOI: 10.1038/ncomms1222 | www.nature.com/naturecommunications

2011 Macmillan Publishers Limited. All rights reserved.

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms1222

21. Houghton, J. T. & Intergovernmental Panel on Climate Change Working Group I in 881 (Cambridge University Press, 2001).

22. Parry, M. L. & Intergovernmental Panel on Climate Change Working Group II in 976 (Cambridge University Press, 2007).

23. Ries, J. B., Cohen, A. L. & McCorkle, D. C. Marine calciers exhibit mixed responses to CO2-induced ocean acidication. Geology 37, 11311134 (2009).

24. Hendriks, I. E., Duarte, C. M. & lvarez, M. Vulnerability of marine biodiversity to ocean acidication: a meta-analysis. Estuar. Coastal Shelf Sci. 86, 157164 (2010).

25. Gledhill, D. K., Wanninkhof, R. & Eakin, M. Observing ocean acidication from space. Oceanography 22, 4859 (2009).

26. Manzello, D. P. Ocean acidication hot spots: spatiotemporal dynamics of the seawater CO2 system of eastern Pacic coral reefs. Limnol. Oceanogr. 55,

239248 (2010).27. Bates, N. R., Amat, A. & Andersson, A. J. Feedbacks and responses of coral calcication on the Bermuda reef system to seasonal changes in biological processes and ocean acidication. Biogeosciences 7, 25092530 (2010).

28. Gattuso, J.- P., Allemand, D. & Frankignoulle, M. Photosynthesis and calcication at cellular, organismal and community levels in coral reefs: a review on interactions and control by carbonate chemistry. Am. Zool. 39, 160183 (1999).

29. Langdon, C. Review of experimental evidence for eects of CO2 on calcication of reef builders. Proceedings 9th International Coral Reef Symposium, Bali, Indonesia. 2, 10911098 (2002).

30. Kleypas, J. & Langdon, C. in Coral Reefs and Climate Change: Science and Management (eds Phinney, J.T., Hoegh-Guldberg, O., Kleypas, J., Skirving, W. & Strong, A.E.) 73110 (American Geophysical Union, 2006).

31. Langdon, C. & Atkinson, M. J. Eect of elevated pCO2 on photosynthesis and calcication of corals and interactions with seasonal change in temperature/ irradiance and nutrient enrichment. J. Geophys. Res. 110, C09S07 (2005).32. Reynaud, S. et al. Interacting eects of CO2 partial pressure and temperature on photosynthesis and calcication in a scleractinian coral. Glob. Change Biol. 9, 16601668 (2003).

33. Renegar, D. A. & Riegl, B. M. Eect of nutrient enrichment and elevated CO2 partial pressure on growth rate of Atlantic scleractinian coral Acropora cervicornis. Mar. Ecol. Prog. Ser. 293, 6976 (2005).

34. Cohen, A. L. & Holcomb, M. Why corals care about ocean acidication: uncovering the mechanism. Oceanography 22, 118127 (2009).

35. Knutson, D. W., Buddemeier, R. W. & Smith, S. V. Coral chronometers: seasonal growth bands in reef corals. Science 177, 270272 (1972).

36. Lough, J. M. & Barnes, D. J. Environmental controls on growth of the massive coral Porites. J. Exp. Mar. Biol. Ecol. 245, 225243 (2000).

37. Bessat, F. & Buigues, D. Two centuries of variation in coral growth in a massive Porites colony from Moorea (French Polynesia): a response of ocean-atmosphere variability from south central Pacic. Palaeogeogr. Palaeoclimatol. Palaeoecol. 175, 381392 (2001).

38. Lough, J. M. & Barnes, D. J. Several centuries of variation in skeletal extension, density and calcication in massive Porites colonies from the Great Barrier Reef: a proxy for seawater temperature and a background of variability against which to identify unnatural change. J. Exp. Mar. Biol. Ecol. 211, 2967 (1997).

39. Lough, J. M. Coral calcication from skeletal records revisited. Mar. Ecol. Prog. Ser. 373, 257264 (2008).

40. Cooper, T. F., Death, G., Fabricius, K. E. & Lough, J. M. Declining coral calcication in massive Porites in two nearshore regions of the northern Great Barrier Reef. Glob. Change Biol. 14, 529538 (2008).

41. Death, G., Lough, J. M. & Fabricius, K. E. Declining coral calcication on the Great Barrier Reef. Science 323, 116119 (2009).

42. Tanzil, J., Brown, B., Tudhope, A. & Dunne, R. Decline in skeletal growth of the coral Porites lutea from the Andaman Sea, South Thailand between 1984 and 2005. Coral Reefs 28, 519528 (2009).

43. Cantin, N. E., Cohen, A. L., Karnauskas, K. B., Tarrant, A. M. & McCorkle, D. C. Ocean warming slows coral growth in the central Red Sea. Science 329, 322325 (2010).44. Saenger, C., Cohen, A. L., Oppo, D. W., Halley, R. B. & Carilli, J. E. Surface-temperature trends and variability in the low-latitude North Atlantic since 1552. Nat. Geosci. 2, 492495 (2009).

45. Dodge, R. E. & Brass, G. W. Skeletal extension, density and calcication of the reef coral, Montastrea annularis: St. Croix, U.S. Virgin Islands. Bull. Mar. Sci. 34, 288307 (1984).

46. Carricart-Ganivet, J. P., Beltran-Torres, A. U., Merino, M. & Ruiz-Zarate, M. A. Skeletal extension, density and calcication rate of the reef building coral Montastraea annularis (Ellis and Solander) in the Mexican Caribbean. Bull. Mar. Sci. 66, 215224 (2000).

47. Eneld, D. B., Mestas-Nunez, A. M. & Trimble, P. J. The Atlantic multidecadal oscillation and its relation to rainfall and river ows in the continental US. Geophys. Res. Lett. 28, 20772080 (2001).

48. Jones, P. D., Jonsson, T. & Wheeler, D. Extension to the North Atlantic oscillation using early instrumental pressure observations from Gibraltar and south-west Iceland. Int. J. Climatol. 17, 14331450 (1997).

49. Edinger, E. N. et al. Normal coral growth rates on dying reefs: are coral growth rates good indicators of reef health? Mar. Pollut. Bull. 40, 404425 (2000).

50. Somereld, P. et al. Changes in coral reef communities among the Florida Keys, 19962003. Coral Reefs 27, 951965 (2008).

51. Carricart-Ganivet, J. P. & Merino, M. Growth responses of the reef-building coral Montastraea annularis along a gradient of continental inuence in the southern Gulf of Mexico. Bull. Mar. Sci. 68, 133146 (2001).

52. Knight, J. R., Folland, C. K. & Scaife, A. A. Climate impacts of the Atlantic multidecadal oscillation. Geophys. Res. Lett. 33, L17706 (2006).

53. Hudson, H. Growth rates in Montastrea faveolata: a record of environmental change in Key Largo Coral Reef Marine Sanctuary, Florida. Bull. Mar. Sci. 31, 444459 (1981).

54. Hudson, J. H., Hanson, K. J., Halley, R. B. & Kindinger, J. K. Environmental implications of growth rate changes in Montastrea faveolata: Biscayne National Park, Florida. Bull. Mar. Sci. 54, 647669 (1994).

55. Manzello, D. P. Coral growth with thermal stress and ocean acidication: lessons from the eastern tropical Pacic. Coral Reefs 29, 749758 (2010).56. Chalker, B., Barnes, D. & Isdale, P. Calibration of x-ray densitometry for the measurement of coral skeletal density. Coral Reefs 4, 95100 (1985).

57. Buddemeier, R. W. Environmental controls over annual and lunar monthly cycles in hermatypic coral calcication. Proceedings second International Coral Reef Symposium, Brisbane, Australia. 2, 259267 (1974).

58. Gledhill, D. K., Wanninkhof, R., Millero, F. J. & Eakin, M. Ocean acidication of the Greater Caribbean Region 19962006. J. Geophys. Res. 113, C10031 (2008).

Acknowledgments

Support for this project was provided by the National Coral Reef Institute, NOAA Coral Reef Conservation Programme and NOAA Coral Reef Watch. This research was, in part, performed while KPH held a National Research Council Research Associateship Award at NOAA Atlantic Oceanographic and Meteorological Laboratory and part under the auspices of the Cooperative Institute for Marine and Atmospheric Studies (CIMAS), a Cooperative Institute of the University of Miami and the National Oceanic and Atmospheric Administration, cooperative agreement #NA17RJ1226. Cores were collected under a NOAA/NURP award to P.K.S. and R.E.D. We are especially gratefulto Kevin E. Kohler for programming and development of CoralXDS soware. We thank Otto Rutten and Daniel Anderegg for assistance during coring, Rik Wanninkhof for valuable comments during manuscript development, and Greta Mackenzie for manuscript editing. The manuscript contents are solely the opinions of the authors and do not constitute a statement of policy, decision or position on behalf of NOAA or the US Government.

Author contributions

Data collection and analysis were performed by K.P.H., R.E.D. and P.K.S. with modelled data contribution from D.K.G. Data interpretation and implications were contributed by K.P.H, R.E.D. and P.K.S. The manuscript was prepared by K.P.H., R.E.D., P.K.S., D.K.G. and C.M.E.

Additional information

Competing nancial interests: The authors declare no competing nancial interests.

Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/

How to cite this article: Helmle, K. P. et al. Growth rates of Florida corals from 1937 to 1996 and their response to climate change. Nat. Commun. 2:215 doi: 10.1038/ncomms1222 (2011).

NATURE COMMUNICATIONS | 2:215 | DOI: 10.1038/ncomms1222 | www.nature.com/naturecommunications

2011 Macmillan Publishers Limited. All rights reserved.