El Niño-Southern Oscillation and global warming: new data from old corals

Transcription

1 El Niño-Southern Oscillation and global warming: new data from old corals Kim M. Cobb Georgia Inst. of Technology Chris Charles, Scripps Inst. of Oceanography Larry Edwards, Hai Cheng, UMN Emory University April 7, 2005

2 Why study tropical Pacific climate? Sea Surface Temperature Anomaly (oc) ENSO El Niño-Southern Oscillation A coupled ocean-atmosphere phenomenon that originates in the tropical Pacific but affects global climate patterns December 1997 SST Anomalies El Niño impacts - the impacts are not confined to the tropical Pacific -ENSO extremes carry serious economic and social costs - improved ENSO forecasts minimize the costs

3 The instrumental record of ENSO Temperature Deviation ( C) Eastern tropical Pacific Temperature 1982 El Niño 1997 El Niño El Niño La Niña Year Devastation caused by the 1982 El Niño made ENSO prediction a top priority for U.S. and other countries.

4 I. Short-term Predictions 6-month prediction: - requires careful monitoring of tropical Pacific ocean + atmosphere - first ENSO model unveiled in models wrongly predicted a weak El Nino event in 1997 The tropical Pacific observing system

5 Temperature Deviation ( C) Atmospheric CO 2 (ppm) II. Long-term Predictions Eastern tropical Pacific Temperature Average Global Temperature Atmospheric CO Year Ice Core Keeling curve 1982 El Niño 1997 El Niño El Niño La Niña Are severe El Niño events becoming more frequent? How will ENSO change as the Earth warms? Temperature ( C)

6 The instrumental record of ENSO is too short to answer some key questions: 1. Are late 20th century El Niño events more frequent and more severe than those of the recent past? 2. Is there a correlation between average global temperature and El Niño activity? 3. How much and how fast has ENSO changed in the past?

7 Corals: The geologic record of ENSO CORALS from the tropical Pacific record ENSO in the geochemistry of their skeletons Living corals provide records for the last 200 years Fossil corals enable us to extend the record (ex A.D.) CORALS from 125 thousand years ago contain the signature of ENSO!

8 A new dot that extends back 10 centuries - Palmyra Palmyra Island

9 Research Objective: To generate >100-yr-long, high-resolution, high-fidelity climate proxy records from the tropical Pacific Ocean; to extend the record of ENSO back in time Materials: Methods: Modern and Fossil Corals Dating: U-Th radioactive decay series Climate proxy: Coral skeletal oxygen isotopes Site December 1997 SST And Rainfall Anomalies

10 Generating climate reconstructions from the Palmyra corals: 1) Recover the corals. 2) Prove that the coral geochemistry tracks large-scale climate. ie. Calibrate the modern coral record against the instrumental record of climate. 3) Assess dating accuracy and reproduciblity of climate signals in fossil corals. ie. Generate overlapping coral climate records when possible.

11 The Palmyra Island Coral Collection Modern Medieval Warm Period (MWP) Greenland green Little Ice Age (LIA) canals frozen in Europe Date (A.D.)

12 Climate Proxy: Coral oxygen isotopes (δ 18 O) 18 δ O = ( O/ O) spl ( ( O/ O) O/ std 16 O) std x1000 Palmyra Coral δ 18 O ( ) O ( ) Drilled in May Coral δ 18 O=ƒ(SST,δ 18 O SW ) 1990 Palmyra Coral Calibration Coral δ 18 O = -0.23(SST) R = SST ( C)

13 How well does Palmyra coral δ 18 O record ENSO? 3 2 R = Red = instrumental record of ENSO Black = modern coral δ 18 O Less smoothed SST Anomoly ( C) δ 18 O ( ) More smoothed SST Anomoly ( C) R = NIÑO3.4 SST Palmyra coral δ 18 O ( ) Year (A.D.)

14 Test #1: Young fossil vs. Modern coral Palmyra Island Coral Collection Test # Date (A.D.)

15 Modern Coral Fossil Coral Overlap Dating based on decay of 238 U to 230 Th U/Th date 1= U/Th date 2= Absolute date = 1915 U/Th date 3= U/Th date 4= U/Th date 5= Absolute date = To overlap corals, measure 2-5 ages for each coral sequence δ 18 O ( ) Fossil NB9 Modern Obtain ± <10y precision -4.4 Cobb et al, EPSL, 2003

16 Test #2: Fossil coral vs. Fossil coral Palmyra Island Coral Collection Date (A.D.)

17 17 th Century Splice SB3/SB13 R = 0.66 splice SB13/SB8 R = 0.62 SB SB SB8 + 0 δ 18 O( ) Year (A.D.) As number of overlapping corals 3 increases corals, 13 dates, 3,000 δ 18 O measurements

18 Palmyra Island Coral Collection Date (A.D.)

19 SB7 vs. CH9 R = th -15 th Century Splice SB5 vs. CH5 R = 0.71 SB6 vs. CH5 R = splice δ 18 O( ) SB CH SB SB CH Year (A.D.) 5 corals, 29 dates, 14,000 δ 18 O measurements

20 Palmyra Coral δ 18 O Sequences Single records 5-coral splice 3-coral splice Modern C warmer colder δ 18 O( ) Date A.D Cobb et al., Nature, 2003

21 An extended history of ENSO from the Palmyra coral records Years El Niño δ 18 O( ) El Niño La Niña ENSO characteristics can change in less than a decade Most frequent, intense El Niño events of reconstruction

22 Proxy-proxy comparison for last millennium 1 C - rate, magnitude of 20th century climate change unprecedented - no simple relationship b/t tropical Pacific climate and NH temp, solar, or volcanic forcing - warm during Little Ice Age, cool during Medieval Warm Period - changes in the meridional temperature gradients implied (inc. during LIA, dec. during MWP)

23 Summary Coral δ 18 O is a sensitive, reliable proxy for tropical Pacific climate change. Overlapping corals 1) enables the construction of longer records 2) increases confidence in climate reconstructions Most intense ENSO activity occurred during 17 th century, during Little Ice Age. ENSO characteristics can change dramatically from decade to decade.

24 Food for Thought Coral reefs are disappearing at alarming rates worldwide, due to the combined influence of rising ocean temperatures and human disturbances (pollution, dynamite fishing, etc). Even if ENSO does not change in a greenhouse world, Man has perturbed the environment to the point that an El Niño event could be the straw that breaks the camel s back. Web Resources My homepage: General El Niño info: NOVA El Niño page: Data available at:

25 Gunung Buda Speleothem Project Goal: to reconstruct annual to centennial-scale rainfall variability from an ENSOsensitive region over the last glacial cycle Materials: ~40 speleothems ( m long) Dating: U-series geochemistry Climate proxies: growth rate, δ 18 O, δ 13 C, Sr/Ca, Mg/Ca, optical properties Approach: 1. On-site geochemical calibration 2. High-precision U/Th dates 3. Reproducibility tests 4. billions and billions of datapoints

26

27 U/Th dating results to date gc04 (6cm) ssr03 (38cm) ssr02 (13cm) ssr01 (46cm) sch03 (36cm) gc03 (23cm) baby3 (5cm) baby2 (8cm) baby1 (10cm) sch01 (33cm) mj02 (20cm) mj01 (23cm) ba04b (89cm) sch02 (86cm) ba03 (135cm) ba02 (69cm) ssc01 (75cm) Age (kybp) NOTE: stalagmites have infinite resolution (1 drop every 30 seconds to 10 minutes), temporal resolution of climate records limited only by analytical capabilities (in other words, how low can you go?)

28 Borneo experiences severe drought during El Niño events Mg/Ca δ 18 O ( ) and δ 13 C ( ) Timeseries of rainwater and cave dripwater geochemistry core-top δ 13 C core-top δ 18 O Mulu dripwater Mg/Ca (slow) Mulu dripwater Sr/Ca (slow) Mulu dripwater [Ca] (slow) Mulu dripwater Mg/Ca (fast) Mulu dripwater Sr/Ca (fast) Mulu dripwater [Ca] (fast) Long Napir rainwater Mulu rainwater Mulu driwpater (fast) Mulu dripwater (slow) Buda dripwater (all) Buda rainwater Mulu dripwater δ 13 C (slow) Mulu dripwater δ 13 C (fast) 3.2e-4 3.0e-4 2.8e-4 2.6e-4 2.4e-4 Sr/Ca [Ca] (umol/l) Dai and Wigley, 2000 How are precipitation changes recorded in cave dripwater geochemistry? /01/03 12/01/03 02/01/04 04/01/04 Date 2.2e-4 800