sign in sign up
<<
Home , Paleotempestology

rates reveal response to past

Valerie Troueta,1, Grant L. Harleyb, and Marta Domínguez-Delmásc,d

aLaboratory of Tree-Ring Research, University of Arizona, Tucson, AZ 85721; bDepartment of Geography and Geology, University of Southern , Hattiesburg, MS 39402; cDepartment of Botany, University of Santiago de Compostela, 27002 Lugo, Spain; and dDepartment of History I, University of Huelva, 21071 Huelva, Spain

Edited by Kerry A. Emanuel, Institute of Technology, Cambridge, MA, and approved February 2, 2016 (received for review October 2, 2015) Assessing the impact of future change on North Atlantic most severe change in in documented history (7, tropical cyclone (TC) activity is of crucial societal importance, but the 8), is of particular interest in this context, but TC records limited quantity and quality of observational records interferes with that cover this period are scarce, often present a conservative the skill of future TC projections. In particular, North Atlantic TC estimate of the total number of storm events (9), and largely have response to radiative forcing is poorly understood and creates the insufficient time resolution to distinguish the MM (6, 10–12). dominant source of uncertainty for twenty-first-century projections. Documentary data sets are the main source of paleotempestology Here, we study TC variability in the Caribbean during the Maunder information of appropriate temporal resolution, but most docu- Minimum (MM; 1645–1715 CE), a period defined by the most severe ment-based TC studies have primarily focused on long-term TC reduction in solar irradiance in documented history (1610–present). (e.g., seasonality, recurrence intervals) rather than For this purpose, we combine a documentary time series of Spanish interannual or decadal-scale variability (13, 14). Here, we combine in the Caribbean (1495–1825 CE) with a tree-growth two annual resolution proxy records—a documentary time series suppression chronology from the Keys (1707–2009 CE). We of Spanish shipwrecks in the Caribbean (TCship)andatree-growth find a 75% reduction in decadal-scale Caribbean TC activity during suppression chronology from the Florida Keys (TCsupp)—to ex- the MM, which suggests modulation of the influence of reduced tend the observational Caribbean TC (CTC) record back over the solar irradiance by the cumulative effect of cool North Atlantic sea last 500 y and thus to cover the MM. – surface temperatures, El Niño like conditions, and a negative phase Over the past centuries, TCs have caused destruction of hu- of the North Atlantic Oscillation. Our results emphasize the need man settlements and wreaked havoc at sea. In the Caribbean, to enhance our understanding of the response of these oceanic TCs were the primary documented cause of shipwrecks in the and atmospheric circulation patterns to radiative forcing and cli- sixteenth through eighteenth centuries (15) and they left their mate change to improve the skill of future TC projections. mark on regional history. For instance, Spanish hegemony over was secured in 1640 after a hurricane decimated a Dutch Caribbean | tropical cyclone | | | fleet poised to attack Havana (15), leading to an additional documentary data century of Spanish monopoly over trade between the Caribbean and Europe. We make use of the well-documented maritime TC EARTH, ATMOSPHERIC, AND PLANETARY SCIENCES andfalling tropical cyclones (TCs) bring devastation to natu- legacy in the Caribbean region to reconstruct CTC variability. Lral and human landscapes with floods, winds, and storm Our reconstruction (TCship) is based on a comprehensive docu- surges. In recent decades, TC-related human mortality and mentary compilation (16) of 657 ships of Spanish origin economic losses have risen in step with increasing populations in that wrecked in the Caribbean Basin (Fig. 1A and Tables S1 high-risk coastal communities (1). TC damage is expected to and S2) over the period 1495–1825 CE due to storms or further increase in the near future with rising exposure and unspecified factors. projected anthropogenic (2). This is particularly the case for the North Atlantic Basin, which is one of the most Significance TC-active basins globally. The development of successful adap- tation and mitigation strategies relies on skillful projections of North Atlantic TC activity, as well as an improved understanding Twenty-first-century North Atlantic tropical cyclone (TC) pro- of the drivers of its variability. jections are crucial for the development of adaptation and Modeling studies of twenty-first-century global TC activity mitigation strategies, but they are subject to large uncer- generally converge in their projections of increased TC intensity tainties, particularly with respect to TC response to radiative forcing. We used a combination of tree-ring data and historical and decreased frequency, but the magnitude range of projected shipwreck data to show that TC activity in the Caribbean was North Atlantic TC variability is wide (3). Uncertainties in twenty- distinctly suppressed during the Maunder Minimum (1645– first-century North Atlantic TC projections are largely driven by 1715 CE), a period when solar irradiance was severely reduced. the chaotic nature of the and by our limited un- This solar fingerprint on decadal-scale Caribbean TC variability derstanding of TC response to radiative forcing, including an- implies modulation by a combination of basin-wide climatic thropogenic greenhouse gases and aerosols, as well as natural phenomena. Our findings highlight the need to enhance our variability in volcanic and solar activity (4). Response uncertainty understanding of the response of atmospheric circulation pat- is the dominant source of uncertainty toward the end of the terns to radiative forcing and climate change to improve the twenty-first century (4), with different model runs resulting in TC skill of future TC projections. responses of opposing sign to projected radiative forcing (3). Our understanding of TC response to radiative forcing—and thus the Author contributions: V.T., G.L.H., and M.D.-D. designed research, performed research, skill of future TC projections—is restricted by limitations in the analyzed data, and wrote the paper. time-series length and quality of observational records (5) that The authors declare no conflict of interest. hinder trend detection and attribution (3). This article is a PNAS Direct Submission. To attribute significant TC changes to specific climate forc- 1To whom correspondence should be addressed. Email: [email protected]. ings, recent TC activity needs to be placed in a longer-term This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. context (6). The Maunder Minimum (MM; 1645–1715 CE), the 1073/pnas.1519566113/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1519566113 PNAS | March 22, 2016 | vol. 113 | no. 12 | 3169–3174 Downloaded by guest on September 26, 2021 TCship does not overlap in time with the North Atlantic TCsupp captures 40 of the 44 storms—occurring during 31 out Hurricane Database (HURDAT; 1851–2010 CE) (17) and there- of a total of 34 storm years—that tracked within 160 km of Big fore we used TCsupp (1707–2009 CE) to assess its validity as a Pine Key over the instrumental period (Figs. 1A,2A, and S1; CTC proxy. TCsupp is based on 38 south Florida slash pine (Pinus Tables S3 and S4). Widespread suppressions in tree growth (high 2 elliottii var. densa) trees from Big Pine Key that show common TCsupp values) corresponded to storm years (χ = 255, P < 0.001) 0 patterns of suppressed growth (Fig. 1B). The primary causes of and typically occurred the same year (t ; P < 0.001) or 1 y after +1 tree-growth suppressions in the Florida Keys are high-energy (t ; P < 0.01) a TC event (Figs. 1C and 2A). We interpret the winds and storm-surge-induced and TCsupp double peak in the superposed epoch analysis (SEA) results to thus reflects interannual variability in landfalling CTCs. Owing represent the immediate or delayed effect of wind and storm to the geography and regional positioning of the Florida Keys surge damage on tree growth. The timing of a TC event during relative to the Caribbean Sea and (Fig. 1A) and the hurricane season (August to October) relative to the growing to the decadal-scale relationship between landfalling and basin- season of trees (February to November) (20) likely dictates 0 +1 wide TC dynamics (18, 19), TCsupp canalsobeinterpretedtore- whether suppression in growth occurs at t or t , but a signifi- flect Caribbean basin-wide TC dynamics. cant pattern was not found.

A

BC0.6

0.4

0.2 simulated) 0

PDSI (departure actual from -0.2 -3 -2 -1 01 23 Lag year

Fig. 1. Geographical location, storm-induced tree growth suppression, and its climatic signal at Big Pine Key, Florida. (A) Geographical location of the BPK tree-ring site (red dot), HURDAT-derived (1851–2010 CE) (17) category 1–5 TCs (white lines; n = 44) that tracked within 160 km (red buffer ring) of the site, and countries/states for which shipwrecks were recorded (black shading) (16); (B) Slash pine (P. elliottii var. densa) section (dated over 1707–1829 CE) with tree- growth suppressions (multiple consecutive narrower-than-average annual rings) resulting from TC events indicated by date of recorded TC event; (C)SEA

(1895–2009 CE) of TCsupp event series (defined as years when >75% of samples showed growth suppression; n = 12) with July–November Palmer Severity Index (PDSI) for Florida climate division 7.

3170 | www.pnas.org/cgi/doi/10.1073/pnas.1519566113 Trouet et al. Downloaded by guest on September 26, 2021 AB

C

Fig. 2. Instrumental and reconstructed CTC activity. SEA of the TCsupp time series with instrumental [HURDAT, ref. 17; 1851–2010 CE] (A); and reconstructed (TCship; 1495–1825 CE) (B) CTC events. SEA was performed for TCsupp contemporaneous to (year 0) and lagging (1 y before, 5 y following) the TC event years. HURDAT TC event years were selected based on category 1–5 storms that tracked within 160 km radius from Big Pine Key (A). TCship event years were defined as years with five or more shipwrecks (16) (B). The combined event list (C) consisted of TCship, HURDAT, and documented historical storms (14) (1826–1849 CE; asterisks above TCsupp).

A similar pattern occurs when comparing TC to TC recorded in an array of Caribbean and lower- EARTH, ATMOSPHERIC, supp ship AND PLANETARY SCIENCES events: anomalously high TCsupp values occurred in the year of resolution TC proxies (10–12, 23–25) and is concurrent with 0 +1 (t ; P < 0.001) or the year after (t ; P < 0.05) severe TCship increased ocean stratification—possibly linked to reduced TC- events (χ2 = 470, P < 0.0001; Fig. 2B), suggesting a high syn- induced mixing—near Great Bahama Bank (26) and increased chronicity between years of damage at sea (shipwrecks) and on aridity in various proxy records from the circum-Caribbean re- land (pines). Comparable patterns were found when selecting gion (27, 28). shorter but better replicated TCsupp time series (Fig. S2) and Modern CTC variability is influenced by four key large-scale – i when using a TCship time series based only on shipwrecks that climate state variables (3, 29 31): ( ) absolute tropical Atlantic were documented to be caused by storms (Fig. S3). Furthermore, sea surface temperatures (SSTs) (32); (ii) tropical Atlantic SSTs this “double peak” SEA pattern also occurs when using a com- relative to tropical mean SSTs (relative SSTs) (33); (iii) the El – iv bined storm event list (1707–2010 CE) including TCship, docu- Niño Southern Oscillation (ENSO); and ( ) the North Atlantic mented historical storms (14), and HURDAT data (Fig. 2C). Oscillation (NAO). A combination of proxy records suggests that TCsupp and TCship combined provide an annual resolution the states of these climate factors converged during the Little Ice record of CTC variability that covers the (ca. Age, and the MM in particular, to create a drastically unfavor- 1500–1850 CE), a period characterized by unusually cool global able environment for CTC development. The distinct reduction temperatures driven largely by a combination of strong volcanic in CTC activity during the MM can be attributed to cool absolute activity and fluctuations in solar irradiance (21). Over this pe- SSTs in the Caribbean (34, 35) and in the main development riod, TCship shows distinct decadal-scale variability (Fig. 3A) that region (36): warmer ocean waters provide more thermal and corresponds to decadal-scale variations in an independent ship- kinetic energy and thus increase the potential for a greater wreck compilation (Fig. S4). By calculating decadal-scale per- number of severe storms (31) (Fig. 3E). Moreover, west Atlantic centages of shipwrecks per number of active North Atlantic SST anomalies during the MM were on average −0.21 °C cooler Spanish ships (TCship%), we verified that decadal-scale fluctua- than SST anomalies in the other (Indian, west Pacific, and east tions in TCship were not a function of fluctuations in Spanish Pacific) tropical ocean basins (based on data from ref. 37; Fig. transatlantic shipping activity (Figs. 3B and S5). Both TCship and 3E), resulting in cool relative SSTs. Anomalously cool absolute TCship% reveal a distinct reduction of CTC activity during the and relative SSTs resulting from generally low sunspot activity MM (Fig. 3 A–D), which is also expressed as a reduction in the during the MM (8) could therefore have impacted the thermo- number of years with shipwrecks per decade (Fig. 3C). This dynamic environment, vertical wind shear, and tropospheric finding is supported by historical evidence from ship logbooks stability, thus creating conditions adverse to CTC development. (14) suggesting increased CTC activity during the 1550s–1640s The MM was likely also a period of broad-scale climate shifts. followed by a marked reduction until the 1720s (Fig. 3B) and Model simulations suggest a southerly position of the intertropical by a lake sediment geochemical record from (22) (Fig. convergence zone, an El Niño–like mean state, and negative NAO 3E). Reduced seventeenth-century CTC activity has also been conditions (7) during the late seventeenth century, that have been

Trouet et al. PNAS | March 22, 2016 | vol. 113 | no. 12 | 3171 Downloaded by guest on September 26, 2021 A

B

C

D

E

Fig. 3. Reduced CTC activity during the Maunder Minimum. (A) Decadal-scale TCship and instrumental (HURDAT) time series sums and TCsupp averages; (B)TCship% and decadal documentary Caribbean storm counts (14); (C) number of years per decade with more than one or four shipwrecks in TCship;(D) spectral solar irradiance (8); (E) tropical Atlantic SST, relative SST proxies (calculated as the residual between tropical Atlantic SST anomalies and averaged Indian, west, and east Pacific Oceans SST anomalies), and storm proxies from Cariaco Basin (34), tropical Atlantic (35), a Jamaica lake sediment geochemical record (22), mean SST anomalies over the Main Development Region (36), and tropical west Atlantic, west and east Pacific, and Indian Oceans corals (37). Gray-shaded area indicates the MM (1645–1715 CE). Smoothing techniques in E include loess (34) and 30-y spline (36, 37). The extended hurricane activity index in E is calculated over the region 15–25°N and 70–80°W (22).

confirmed by proxy evidence (35, 36, 38, 39). Modern CTC negative NAO during the MM might have provided a pathway climatology—with generally suppressed TC activity during El Niño for an unfavorable effect of reduced solar activity on decadal- years and negative NAO phases—alludes to a likely contribution scale CTC activity. However, on interannual and even daily time of the state of these major circulation patterns to TC suppression scales, the solar–TC relationship is complex, spatially explicit, in the Caribbean during the MM. The majority of sedimentary TC and modulated by oceanic heat content (40). In regions with high proxies on the northeastern US coast, however, did not record a oceanic heat content (e.g., the twentieth-century Caribbean), TC TC suppression during the MM (9, 36), suggesting an important activity is typically reduced during years of strong solar activity, influence of regional oceanic conditions and shifting genesis lo- when tropopause temperatures are warmed and the vertical cations as well as a potential modulation—and northeastward temperature differential responsible for tropospheric convection redirection—of the prevailing storm tracks related to the above- is weakened (41). In contrast, in regions where ocean heat content described changes in atmospheric circulation patterns (30). is limited in summer (e.g., the twentieth-century east Atlantic), Our high-resolution CTC proxy records suggest that a com- strong solar activity can provide a marginal SST increase necessary bination of cold Caribbean SSTs, El Niño–like conditions, and a for , resulting in more TCs. This modulation of the

3172 | www.pnas.org/cgi/doi/10.1073/pnas.1519566113 Trouet et al. Downloaded by guest on September 26, 2021 solar–TC relationship by ocean heat content can also be applied to decadal-scale Florida time series correlated strongly (r = 0.77, decadal temporal scales rather than spatial scales: tropical western n = 31, P < 0.001) with the Florida component of the TCship time Atlantic SST reconstructions suggest that MM Caribbean SSTs series (Fig. S4). were up to 1 °C cooler than present (34, 37; Fig. 3E) and as such represent a potential MM shift in the ocean heat budget and the TCsupp. We aimed our sampling strategy at using radial growth solar–TC relationship. anomalies from Pinus elliottii var. densa (hereafter slash pine) Our findings suggest an abrupt CTC suppression in response trees to detect TCs in the Florida Keys/Caribbean region. For to a severe reduction in solar irradiance and thus provide a this purpose, we collected partial (live trees) or full (dead trees) benchmark for modeling studies of North Atlantic TC response cross-sections from 38 mature old-growth pines that showed no to radiative forcing. The unequivocal strength and sign of the signs of visible injury to the tree (i.e., fire scar, beetle gallery, CTC response to reduced solar activity imply modulation by lightning scar) and were located at the highest elevations (≥ 2m synchronized oceanic and atmospheric circulation patterns. To a.s.l.) within the island interior of Big Pine Key (BPK; Fig. 1A). further improve our understanding and to constrain estimates Trees located at higher BPK elevations are able to survive of future North Atlantic TC activity, additional high-resolution multiple storm surge events because salt water has a short resi- proxy records and targeted climate modeling experiments are dence time at the surface (43). Moreover, BPK slash pine trees needed that can support a process-based investigation of the exhibit a different growth form compared with individuals on responses of North Atlantic TC activity, SSTs, ENSO, and NAO mainland Florida and windthrow-induced mortality associated to radiative forcing and future climate change. with TCs is uncommon. We used standard dendrochronological techniques (44) for sample preparation, measurement, and cross- Data dating of the tree-ring data. TCship. We compiled a total of 656 Spanish shipwreck events Slash pine ring-width data can contain a weak late summer (1495–1825 CE) that occurred in the Caribbean realm and were (September) moisture availability signal (45), but this was not the included in Marx (16). This data set is not complete, as some case at the BPK site (Fig. S6) and we focused our analysis on shipwrecks were never documented and some documents likely tree-growth suppressions (Fig. 1B) in the slash pine data set. We lost due to deterioration with age, war, or fire (15), but it is a expected a TC signal due to loss of tree biomass (leaves, comprehensive collection of shipwrecks in the Americas that branches) to be manifested within the tree-ring record as a pe- includes ca. 4,000 entries cataloged by year and location. Ships of riod of tree-growth suppression (46) and therefore defined British, French, Dutch, and other nationalities also crossed the suppression events as periods in which raw ring width was ≥25% Atlantic and wrecked in the Caribbean from the seventeenth below the 10-y running mean of all growth rings (n = 1,228 for 38 century onward, but to avoid trends in TCship due to increasing tree samples over the period 1707–2009 CE). We than calculated numbers of ships, we compiled only shipwrecks of Spanish origin. an annual time series TCsupp as the percentage of samples in a given We compiled all Spanish shipwreck events that were recorded to year t that contained a suppression. Widespread tree growth sup- have occurred (i) due to storm activity (66%) or for unspecified pressions (defined as suppression in ≥75% of samples) occurred causes (34%), (ii) during the hurricane season (July–November) during wet [positive Palmer Drought Severity Index (PDSI)] sum- or with unknown seasonality, and (c) in Florida (21%), on the mers, most likely caused by TC-induced rainfall (Fig. 1C). Atlantic Coast of Mexico (25%), in (22%), Cuba EARTH, ATMOSPHERIC,

(13%), the (5%), and , , South Methods AND PLANETARY SCIENCES

Carolina, , Mississippi, Jamaica, the Cayman Islands, The relationship between TCsupp and instrumental and reconstructed TC , and (less than 5% each; Fig. 1A). events was evaluated in a contingency analysis and in an SEA. We compared Quantities were recorded as “many,”“several,” or in similar the frequency distribution of growth suppressions in TCsupp to storms in the terms on six occasions in the record and we counted these as five HURDAT (17) and shipwrecks in the TCship time series in a contingency 2 2 shipwrecks, which is the median number of ships reported to analysis and calculated statistical significance levels using a Pearson’s χ (χ ) test. The HURDAT time series (1851–2010 CE) consisted of the number of have wrecked during storms when more than one ship wrecked – (range: 2–29 shipwrecks; n = 39). The TC time series thus category 1 5 storms that tracked within 160 km of BPK per year (Fig. 2A). ship Large storms within this radius are documented to cause major storm surges consists of the number of Spanish ships that wrecked per year in on BPK (47). For instance, Hurricane Wilma passed at 145 km from BPK and the Caribbean region due to storm activity or for unspecified caused growth suppression in 75% of the TCsupp samples in 2005 (Table S3). causes (Fig. 2). We compared lagged relationships in an SEA by superposing a window We calculated decadal sums based on TCship (Fig. 3A) and of contemporaneous and lagged annual growth suppressions over each

verified that decadal-scale fluctuations were not a function of HURDAT (1851–2010 CE) or TCship-based (1707–1825 CE) TC event year (48). fluctuations in the number of active transatlantic Spanish ships Event years were defined as years in which category 1–5TCstrackedwithin by calculating decadal-scale percentages of shipwrecks per number 160 km of BPK (n = 34) for HURDAT and as years with five or more shipwrecks events (n = 9; Fig. 2B)forTC Results are consistent when using smaller (100 of active ships in the Caribbean (TCship%;Fig.3B). For this pur- ship. pose, we compiled a time series of the total number of ships km) and larger (300 km) tracking radii for HURDAT TCs (Fig. S1) or shorter but better replicated TCsupp time series (Fig. S2). In addition to this, we used a departing from Seville to the Caribbean and Gulf of Mexico during = – – combined event list of TCship, documented TCs/storms from ref. 14 (n 2; 1826 the period 1511 1784 CE based on the General Archive of the 1849 CE), and HURDAT (Fig. 2C). Monte Carlo simulations were used (n = Indies (AGI) (pares.mcu.es). Within the AGI, we tallied each ship 1,000) to develop bootstrapped confidence intervals to determine whether a departure among the 643 registers in the Registros de Ida. Because significantly higher than average number of trees showed growth suppression – + of discrepancies in the AGI related to low ship log entries during in a 7-y window (t 1 through t 5) centered around the TC event year (t0). early and late decades, we calculated annual ship occurrences for the period 1590–1780 CE only (n = 3,881; Fig. S5). The data set ACKNOWLEDGMENTS. We thank Justin L. Hart and Mary Glueck for was quality checked for potential duplicate records by considering assistance with data collection and development. We are grateful to P. J. Baker, L. Bakkensen, H. Diaz, and J. Esper for providing suggestions. G.L.H. the ship name, type, and captain name, and duplicate records were was supported by a grant from the Office of Research from the University excluded from the final time series. The record of annual ship of Southern Mississippi, by the National Science Foundation under Grants counts was binned into decadal sums (Fig. S5) and used to calcu- 1002479 and 0538420, and by the Fish and Wildlife Service. M.D.-D. was supported by a University of Arizona Agnese N. Haury Visiting late TCship% (Fig. 3B). Fellowship and by a Marie Sklodowska Curie Innovative Training Networks We further compared the decadal-scale TCship time series to a – = Fellowship (ForSEAdiscovery project, PITN-2013-GA-607545). The idea for shipwreck time series (1515 1815 CE; number of shipwrecks this collaborative project originated during the Second American 232) for Florida based on the data compiled in ref. 42. This Dendrochronology Conference.

Trouet et al. PNAS | March 22, 2016 | vol. 113 | no. 12 | 3173 Downloaded by guest on September 26, 2021 1. Pielke RA, et al. (2008) Normalized hurricane damage in the United States: 1900-2005. 25. Donnelly JP, Woodruff JD (2007) Intense hurricane activity over the past 5,000 years Nat Hazards Rev 9:29–42. controlled by El Niño and the West African . Nature 447(7143):465–468. 2. Mendelsohn R, Emanuel K, Chonabayashi S, Bakkensen L (2012) The impact of climate 26. Lund DC, Lynch-Stieglitz J, Curry WB (2006) Gulf Stream density structure and change on global tropical cyclone damage. Nat Clim Chang 2(3):205–209. transport during the past millennium. Nature 444(7119):601–604. 3. Knutson TR, et al. (2010) Tropical cyclones and climate change. Nat Geosci 3(3): 27. Burn MJ, Palmer SE (2013) Solar forcing of Caribbean drought events during the last 157–163. millennium. J Quat Sci 29(8):827–836. 4. Villarini G, Vecchi GA (2012) Twenty-first-century projections of North Atlantic trop- 28. Lane CS, Horn SP, Orvis KH, Thomason JM (2011) evidence of Little Ice ical storms from CMIP5 models. Nat Clim Chang 2(8):604–607. Age aridity on the Caribbean slope of the Cordillera Central, . 5. Landsea CW, Harper BA, Hoarau K, Knaff JA (2006) Climate change. Can we detect Quat Res 75(3):461–470. trends in extreme tropical cyclones? Science 313(5786):452–454. 29. Sabbatelli TA, Mann ME (2007) The influence of climate state variables on Atlantic 6. Mann ME, Woodruff JD, Donnelly JP, Zhang Z (2009) Atlantic hurricanes and climate Tropical Cyclone occurrence rates. J Geophys Res Atmos 112(D17), 10.1029/2007jd008385. over the past 1,500 years. Nature 460(7257):880–883. 30. Kozar ME, Mann ME, Camargo SJ, Kossin JP, Evans JL (2012) Stratified statistical 7. Shindell DT, Schmidt GA, Mann ME, Rind D, Waple A (2001) Solar forcing of regional models of North Atlantic basin-wide and regional tropical cyclone counts. J Geophys climate change during the Maunder Minimum. Science 294(5549):2149–2152. Res Atmos 117:D18103. 8. Lean J (2000) Evolution of the sun’s spectral irradiance since the Maunder Minimum. 31. Goldenberg SB, Landsea CW, Mestas-Nunez AM, Gray WM (2001) The recent increase Geophys Res Lett 27(16):2425–2428. in activity: Causes and implications. Science 293(5529):474–479. 9. Donnelly JP, et al. (2015) Climate forcing of unprecedented intense‐hurricane activity 32. Webster PJ, Holland GJ, Curry JA, Chang HR (2005) Changes in tropical cyclone in the last 2000 years. Earths Futur 3(2):49–65. number, duration, and intensity in a warming environment. Science 309(5742): 10. Denommee KC, Bentley SJ, Droxler AW (2014) Climatic controls on hurricane patterns: 1844–1846. A 1200-y near-annual record from Lighthouse Reef, . Sci Rep 4:3876. 33. Vecchi GA, Swanson KL, Soden BJ (2008) Climate change. Whither hurricane activity? 11. Lane P, Donnelly JP, Woodruff JD, Hawkes AD (2011) A decadally-resolved paleo- Science 322(5902):687–689. hurricane record archived in the late sediments of a Florida . Mar 34. Black DE, et al. (2007) An 8-century tropical Atlantic SST record from the Cariaco Geol 287(1-4):14–30. Basin: baseline variability, twentieth-century warming, and Atlantic hurricane fre- 12. Frappier AB, et al. (2014) Two millennia of tropical cyclone-induced mud layers in a quency. Paleoceanography 22(4):PA4204. northern Yucatan : Multiple overlapping climatic hazards during the Maya 35. Kilbourne KH, Alexander MA, Nye JA (2014) A low latitude paleoclimate perspective Terminal Classic “”. Geophys Res Lett 41(14):5148–5157. on Atlantic multidecadal variability. J Mar Syst 133:4–13. 13. Chenoweth M, Divine D (2008) A document-based 318-year record of tropical cy- 36. Mann ME, et al. (2009) Global signatures and dynamical origins of the Little Ice Age clones in the Lesser Antilles, 1690-2007. Geochem Geophys Geosyst 9(8):Q08013. and Medieval Climate Anomaly. Science 326(5957):1256–1260. 14. Garcia-Herrera R, Gimeno L, Ribera P, Hernandez E (2005) New records of Atlantic 37. Tierney JE, et al. (2015) Tropical sea surface temperatures for the past four centuries hurricanes from Spanish documentary sources. J Geophys Res 110(D3):D03109. reconstructed from archives. Paleoceanography 30(3):226–252. 15. Rappaport EN, Fernandez-Partagas JJ (1997) History of the deadliest Atlantic tropical 38. Trouet V, et al. (2009) Persistent positive North Atlantic oscillation mode dominated cyclones since the discovery of the New World. Hurricanes: Climate and Socioeconomic the Medieval Climate Anomaly. Science 324(5923):78–80. Impacts, eds Diaz HF, Pulwarty RS (Springer-Verlag, Berlin), pp 93–108. 39. Haug GH, Hughen KA, Sigman DM, Peterson LC, Röhl U (2001) Southward migration 16. Marx RF (1983) Shipwrecks in the Americas (Crown Publishers, ), p 482. of the intertropical convergence zone through the Holocene. Science 293(5533): 17. Landsea CW, Franklin JL (2013) Atlantic hurricane database uncertainty and pre- 1304–1308. sentation of a new database format. Mon Weather Rev 141(10):3576–3592. 40. Hodges RE, Elsner JB (2012) The spatial pattern of the sun-hurricane connection across 18. Kozar ME, Mann ME, Emanuel KA, Evans JL (2013) Long-term variations of North the North Atlantic. ISRN Meteorol 2012:1–9. Atlantic tropical cyclone activity downscaled from a coupled model simulation of the 41. Hodges RE, Elsner JB (2011) Evidence linking solar variability with US hurricanes. Int J last millennium. J Geophys Res Atmos 118(24):13383–13392. Climatol 31(13):1897–1907. 19. Reed AJ, Mann ME, Emanuel KA, Titley DW (2015) An analysis of long-term rela- 42. Singer SD (1992) Shipwrecks of Florida (Pineapple Press, Sarasota, FL). tionships among count statistics and metrics of synthetic tropical cyclones downscaled 43. Ross MS, Obrien JJ, Sternberg LDL (1994) Sea-level rise and the reduction in pine from CMIP5 models. J Geophys Res Atmos 120(15):7506–7519. forests in the Florida Keys. Ecol Appl 4(1):144–156. 20. Harley GL, Grissino-Mayer HD, Franklin JA, Anderson C, Kose N (2012) Cambial activity 44. Stokes MA, Smiley TL (1968) An Introduction to Tree-Ring Dating (Univ of Chicago of Pinus elliottii var. densa reveals influence of seasonal insolation on growth dy- Press, Chicago). namics in the Florida Keys. Trees Struct Funct 26(5):1449–1459. 45. Harley GL, Grissino-Mayer HD, Horn SP (2011) The dendrochronology of Pinus elliottii 21. Crowley TJ (2000) Causes of climate change over the past 1000 years. Science in the lower Florida Keys: Chronology development and climate response. Tree Ring 289(5477):270–277. Res 67(1):39–50. 22. Burn MJ, Palmer SE (2015) Atlantic hurricane activity during the last millennium. Sci 46. Hart JL, Grissino-Mayer HD (2008) Vegetation patterns and dendroecology of a mixed Rep 5:12838. hardwood forest on the Cumberland Plateau: Implications for stand development. 23. Brandon CM, Woodruff JD, Lane DP, Donnelly JP (2013) Tropical cyclone wind speed For Ecol Manage 255(5-6):1960–1975. constraints from resultant storm surge deposition: A 2500 year reconstruction of 47. Saha AK, et al. (2011) and South Florida coastal forests. Clim Change hurricane activity from St. Marks, FL. Geochem Geophys 14(8):2993–3008. 107(1-2):81–108. 24. Ercolani C, Muller J, Collins J, Savarese M, Squiccimara L (2015) Intense southwest 48. Haurwitz MW, Brier GW (1981) A critique of the Superposed Epoch Analysis method - Florida hurricane landfalls over the past 1000 years. Quat Sci Rev 126:17–25. Its application to solar weather relations. Mon Weather Rev 109(10):2074–2079.

3174 | www.pnas.org/cgi/doi/10.1073/pnas.1519566113 Trouet et al. Downloaded by guest on September 26, 2021