Elsevier

Quaternary Science Reviews

Volume 80, 15 November 2013, Pages 129-142
Quaternary Science Reviews

Moisture and temperature changes associated with the mid-Holocene Tsuga decline in the northeastern United States

https://doi.org/10.1016/j.quascirev.2013.09.001Get rights and content

Highlights

  • We evaluate the most abrupt tree species decline in N. America over the Holocene.

  • During the decline, vegetation histories differed at inland and coastal sites.

  • Lake-level histories are, however, consistent between inland-coastal New England.

  • Changes in temperature preference gradients imply abrupt climate change ca 5.5 ka.

  • Abrupt atmosphere–ocean circulation changes may explain abrupt regional changes.

Abstract

A decline of hemlock (Tsuga) populations at ca 5.5 ka (thousands of calibrated radiocarbon years before 1950 AD) stands out as the most abrupt vegetation change of the Holocene in North America, but remains poorly understood after decades of study. Recent analyses of fossil pollen have revealed a concurrent, abrupt oak (Quercus) decline and increases in the abundance of beech (Fagus) and pine (Pinus) on Cape Cod in eastern Massachusetts, but the replacement of drought-tolerant oaks by moisture-sensitive beeches appears inconsistent with low lake levels in the region at the same time. The oak and beech changes are also limited to coastal areas, and the coastal-inland differences require an explanation. Here, we develop a new lake-level reconstruction from Deep Pond, Cape Cod by using a transect of sediment cores and ground-penetrating radar (GPR) profiles to constrain the past elevations of the sandy, littoral zone of the pond. The reconstruction shows that a series of multi-century episodes of low water coincide with the abrupt hemlock and oak declines, and interrupt subsequent phases of hemlock recovery. The lake-level variations equal precipitation deficits of ∼100 mm superimposed on a Holocene long moisture increase of >400 mm. However, because moisture deficits do not easily explain the oak and beech changes, we also evaluate how the climate preferences of the regional vegetation changed over time by matching the fossil pollen assemblages from Deep Pond with their modern equivalents. Reconstructions of the precipitation requirements of the vegetation correlate well even in detail with the lake-level record (r = 0.88 at Deep Pond), and indicate close tracking of effective moisture (precipitation minus evapotranspiration) by the vegetation despite the abrupt species declines, which could have decoupled climate and vegetation trends. Reconstructions of the temperature preferences of the vegetation indicate that coastal sites may have cooled by 0.5–2.5 °C after ca 5.5 ka, while inland sites warmed by 0.5–1 °C. The change in coastal temperature preferences agrees with sea surface cooling in the western Atlantic Ocean of >1 °C. Consequently, the persistence of low hemlock abundance after 5.5 ka in the northeast U.S. may have resulted from oceanic changes that produced multi-century droughts and thus delayed the post-decline recovery of hemlock populations.

Introduction

Davis (1981) described an abrupt hemlock (Tsuga spp.) decline across eastern North America at ca 5.5 ka (thousands of calibrated radiocarbon years before 1950 AD), noting sharp decreases in hemlock pollen abundances from Michigan to Maine. The decline was initially attributed to a pathogen or pest outbreak because of 1) the range-wide abrupt nature of the decline, 2) the apparent change in only one taxon, 3) the similarity of the event's palynological signature to recent forest disease outbreaks, and 4) a lack of evidence for concurrent climate change or other drivers (Allison et al., 1986). The change, therefore, appeared to represent an abrupt ecological event independent of progressive Holocene climate changes. Since then, additional work has linked the decline to insect outbreaks (Anderson et al., 1986, Bhiry and Filion, 1996), dated the mid-point of the regional decline to 5.5 ± 0.38 ka (Bennett and Fuller, 2002), and documented the decline as one of the most abrupt vegetation changes of the past 8 ka in North America (Shuman et al., 2005, Shuman et al., 2009).

New evidence, however, challenges the hypothesis that the period of low hemlock abundance resulted from biotic interactions independent of climate. Evidence for regional to hemispheric climate change at ca 5.5 ka now includes reduced lake levels and other evidence of drought from Ontario to New Hampshire (Yu et al., 1997, Haas and McAndrews, 1999, Lavoie and Richard, 2000, Newby et al., 2000, Newby et al., 2009, Newby et al., 2011, Shuman et al., 2001, Shuman et al., 2004, Shuman et al., 2005, Shuman et al., 2009, Muller et al., 2003, Booth et al., 2012), changes in atmospheric circulation (Yu et al., 1997, Kirby et al., 2002), and shifts in sea-surface temperatures (SSTs) (deMenocal et al., 2000, Sachs, 2007). Pollen records from areas near the range limits of hemlock also indicate climate change (Calcote, 2003, Foster et al., 2006), and importantly, show that hemlock did not change independently of other taxa.

The hemlock decline was concurrent with increases in Fagus (beech) and Pinus (pine) and decreases in Quercus (oak) populations in coastal Massachusetts (Tzedakis, 1992, Foster et al., 2006), as well as potentially synchronous (within dating uncertainties) beech declines (Williams, 1974, Shane and Anderson, 1993), Ulmus (elm) declines (Williams et al., 2004, Nelson et al., 2006, Grimm et al., 2009), and prairie expansion (Nelson et al., 2006) in the Midwest. However, the relative timing of the various climate and ecological changes has only been examined in detail at a few locations where simple cause-and-effect relationships were not apparent (Booth et al., 2012). The detailed site-specific analyses raise questions about the relative rates and timing of vegetation and moisture changes at local spatial scales and fine time scales (Booth et al., 2012), but the range-wide maintenance of low hemlock abundance and other contemporaneous features of the regional vegetation (e.g., high coastal beech abundance) for millennia after the decline could be consistent with the persistence of a new climatic regime, which also contributed to low lake levels and other changes (Shuman et al., 2004, Foster et al., 2006).

The role of temperature during these mid-Holocene changes has not been examined and may be essential to understanding the associated coastal oak-beech dynamics, in particular. In fact, pollen data from New England show different vegetation changes at inland and coastal sites (Foster et al., 2006), and suggest a persistent change in climate gradients across southern New England at the time of the hemlock decline. Decreases in oak and increases in beech populations might indicate that an increase in fog or other forms of effective moisture buffered the coast from drought experienced inland (e.g., Yu et al., 1997, Newby et al., 2011, Booth et al., 2012). Alternatively, the temperature gradient between coastal and inland sites across southern New England may have changed; a decrease in temperatures at the coast and an increase in temperatures at inland sites (ca 5.5 ka BP) could be consistent with the observed changes. Similar shifts in moisture and temperature gradients have been observed historically in association with atmospheric and ocean circulation changes in the AD 1960s when both temperatures and precipitation in New England were below normal during the most severe historic drought (Namias, 1966).

Testing these hypotheses requires investigations of both inland and coastal effective moisture and temperature histories. Effective moisture, discussed here as the balance of precipitation (P) and evapotranspiration (E), can be reconstructed by evaluating lake sediments to locate the position of lake shorelines through time using ground-penetrating radar (GPR) and various sedimentary analyses (Pribyl and Shuman, Digerfeldt, 1986, Dearing, 1997, Shuman, 2003, Shuman et al., 2005). For this study, we provide the record of lake-level changes at Deep Pond, a kettle lake located on Cape Cod, MA (Table 1, Fig. 1A) where the mid-Holocene oak decline and beech rise were first documented (Foster et al., 2006), and compare the lake-level record with effective moisture (PE) data from inland sites to address the hypothesis that varying moisture levels during the mid-Holocene contributed to or sustained the hemlock decline.

We use regional sea-surface temperature (SST) reconstructions (Sachs, 2007) to help evaluate possible coastal temperature trends, but calibrate the pollen data themselves to quantify what temperature changes (if any) would be consistent with the abrupt vegetation changes. By doing so, we build on previous efforts to use pollen data to reconstruct regional climate gradients (Webb et al., 1993), but simply use the modern analog technique (MAT) (Overpeck et al., 1985, Guiot et al., 1993) to evaluate how the temperature and precipitation preferences of the regional vegetation changed. Because inferring a climatic cause of the vegetation changes from the pollen data would be circular, we use the approach to ask what magnitudes and directions of climate change could be consistent with the pollen record. Comparisons with the lake-level and SST records then enable us to evaluate whether the vegetation changes are consistent with the independently constrained climatic history.

We will evaluate three aspects of environmental change at the time of the hemlock decline: 1) past hydroclimatic trends at Deep Pond for comparison with inland water-level declines; 2) past changes in the temperature preferences of the coastal and inland vegetation; and 3) past agreement among vegetation climate preferences and independent climate indicators. We use our sediment-based lake-level reconstruction approach to pursue the hydroclimate changes, and then apply the MAT to pollen data to reconstruct the gradients and consistency of the vegetation climate preferences. We first apply the MAT to well-dated, detailed pollen data from a pair of inland and coastal sites in Massachusetts, Little Pond-Royalston (LPR) and Deep Pond, respectively (Table 1, Fig. 1A), and then use a wide array of available data (e.g., SST, lake level, and temperature preference data) from southern New England to examine the spatial patterns and potential drivers of mid-Holocene climate changes during the hemlock decline.

Section snippets

Primary study site

We reconstruct the lake-level history of Deep Pond (41.581°N, −71.640°W, 23 m above sea level), a 0.89 ha, ∼3.5-m-deep, closed-basin kettle lake at the base of Cape Cod near Falmouth, MA on the Buzzards Bay Moraine (Fig. 1). The pond lies within glacial till and outwash on a perched aquifer underlain by fine-grained sediments. The fine sediments isolate the pond from interactions with deep groundwater and sea-level fluctuations (Oldale, 1992), and allow lake-level fluctuations to result

Lake-level changes at Deep Pond

Stratigraphic changes at Deep Pond show evidence of a long-term rise in water levels based on shoreward expansion of deep-water sediments (Figs. 6 and 7). The rise culminated in high water after ca 3 ka consistent with other regional reconstructions (e.g., Shuman et al., 2001). Additionally, 17 stratigraphic units, including 8 high LOI units (even numbered events) and 9 low LOI units (odd numbered events), interrupt the long-term shoreward expansion of sediments (Figs. 6 and 7). The low LOI

Reconstruction similarities and their implications

Many of the lake-level changes interpreted from stratigraphic units at Deep Pond correspond with similar evidence of lake-level changes at Davis and Crooked Ponds, MA (Newby et al., 2009, Newby et al., 2011, Shuman et al., 2001, Shuman et al., 2009) as well as other sites in the region (see discussion of the comparison of New Long Pond, MA and White Pond, NJ with Davis Pond in Newby et al., 2011). All of these sites experienced brief episodes of low water recorded by sand layers in intermediate

Conclusions

The hemlock decline represents the most abrupt vegetation change of the Holocene in eastern North America, and although extensively studied, it remains poorly understood. Evidence of a synchronous oak decline, and beech and pine increases at coastal Massachusetts, raise questions about potential climatic drivers of these extensive vegetation changes. Even if climate did not cause the declines, our results provide support for the interpretation that drought helped to maintain the subsequent

Acknowledgments

This work was supported by National Science Foundation grants to Shuman (DEB-0816731) and Foster and Oswald (DEB-0815036). We also thank S.T. Jackson, P. Heller, T.A. Minckley, B.R. Marsicek, J.L. Gill, and three anonymous reviewers for helpful comments that greatly improved the manuscript; G.E. Carter, J. Calder, J. Donnelly, P. Pribyl, T. Miller, and D. Colvard for field and lab assistance; D.J. Marsicek and L.A. Marsicek for moral support.

References (69)

  • A.C. Parnell et al.

    A flexible approach to assessing synchroneity of past events using Bayesian reconstructions of sedimentation history

    Quat. Sci. Rev.

    (2008)
  • L.C.K. Shane et al.

    Intensity, gradients and reversals in late glacial environmental change in east-central North America

    Quat. Sci. Rev.

    (1993)
  • B. Shuman et al.

    Late- Quaternary water-level variations and vegetation history at Crooked Pond, southeastern Massachusetts

    Quat. Res.

    (2001)
  • B.N. Shuman et al.

    Abrupt climate change as an important agent of ecological change in the Northeast U.S. throughout the past 15,000 years

    Quat. Sci. Rev.

    (2009)
  • R.J. Telford et al.

    The secret assumption of transfer functions: problems with spatial autocorrelation in evaluation model performance

    Quat. Sci. Rev.

    (2005)
  • R.S. Webb et al.

    Pollen response-surface estimates of late-Quaternary changes in the moisture balance of the northeastern United States

    Quat. Res.

    (1993)
  • D.R. Whitehead

    Late-glacial and post-glacial vegetational history of the Berkshires, western Massachusetts

    Quat. Res.

    (1979)
  • J. Whitmore et al.

    Modern pollen data from North America and Greenland for multi-scale paleoenvironmental applications

    Quat. Sci. Rev.

    (2005)
  • J.W. Williams et al.

    Obtaining accurate and precise environmental reconstructions from the modern analog technique and North American surface pollen dataset

    Quat. Sci. Rev.

    (2008)
  • M.G. Winkler

    A 12,000-year history of vegetation and climate for Cape Cod, Massachusetts

    Quat. Res.

    (1985)
  • T.D. Allison et al.

    Pollen in laminated sediments provides evidence for a mid-Holocene forest pathogen outbreak

    Ecology

    (1986)
  • R.S. Anderson et al.

    History of late-and post-glacial vegetation and disturbance around Upper South Branch Pond, northern Maine

    Can. J. Bot.

    (1986)
  • G.S. Baker et al.

    An Introduction to Ground Penetrating Radar (GPR)

    (2007)
  • P.J. Bartlein et al.

    Pollen-based continental climate reconstructions at 6 and 21 ka: a global synthesis

    Clim. Dyn.

    (2010)
  • K.D. Bennett et al.

    Determining the age of the mid-Holocene Tsuga canadensis (hemlock) decline, eastern North America

    Holocene

    (2002)
  • R.K. Booth et al.

    Decomposing the mid-Holocene Tsuga decline in eastern North America

    Ecology

    (2012)
  • P.E. Busby et al.

    Forest response to chronic hurricane disturbance in coastal New England

    J. Veg. Sci.

    (2009)
  • R. Calcote

    Mid-Holocene climate and the hemlock decline: the range limit of Tsuga canadensis in the western Great Lakes region, USA

    Holocene

    (2003)
  • M.B. Davis

    Climatic changes in southern Connecticut recorded by pollen deposition at Rogers Lake

    Ecology

    (1969)
  • M.B. Davis

    Outbreaks of forest pathogens in Quaternary history

  • M.B. Davis et al.

    Sediment focusing in Mirror Lake, New Hampshire

    Limnol. Oceanogr.

    (1982)
  • J.A. Dearing

    Sedimentary indicators of lake-level changes in the humid temperate zone: a critical review

    J. Paleolimnol.

    (1997)
  • G. Digerfeldt

    Studies on past lake-level fluctuations

  • P.W. Dunwiddie

    Postglacial vegetation history of coastal islands in southeastern New England

    Natl. Geogr. Res.

    (1990)
  • Cited by (50)

    • Reconstructing the groundwater recharge history for the Plymouth-Carver Aquifer Massachusetts, USA

      2020, Quaternary International
      Citation Excerpt :

      To our knowledge, this is the first comparison of the TraCE-21ka precipitation model results with hydrologic reconstructions in the Northeastern USA. While reconstructed water levels provide indirect evidence of changing hydrologic conditions, which could be affected by other factors, mainly sea level change, pollen-based moisture reconstructions indicate a significant increase in effective moisture affecting inland and coastal water levels over the whole Holocene [Webb et al., 1993; Marsicek et al., 2013]. The long-term increase in effective moisture is also the dominant hydroclimate trend of mid-latitude North America (Fig. 3a) [Shuman and Marsicek, 2016].

    • Approaches to Paleoclimate Reconstruction

      2020, Encyclopedia of Geology: Volume 1-6, Second Edition
    View all citing articles on Scopus
    View full text