Invited research article
Global physical controls on estuarine habitat distribution during sea level change: Consequences for genetic diversification through time

https://doi.org/10.1016/j.gloplacha.2020.103128Get rights and content

Highlights

  • There is less estuarine habitat on narrow continental margins than wide margins

  • Fore-arc settings are narrower than passive margins

  • Extant fishes have a negative species-area relationship on continental margins

  • Narrow continental margins can facilitate diversification or species accumulation

  • The nuance of Pleistocene sea-level changes should be considered in biological studies

Abstract

Determining the extrinsic (physical) factors controlling speciation and diversification of species through time is of key interest in paleontology and evolutionary biology. The role of sea-level change in shaping species richness patterns of marginal marine species has received much attention, but with variable conclusions. Recent work combining genetic data and Geographical Information Systems (GIS)-based habitat modeling yielded a framework for how geomorphology of continental margins mediates genetic connectivity of populations during sea-level change. This approach may ultimately yield insights on how distinct lineages, species, and biodiversity accumulate in coastal settings. Here, we expand this GIS work globally to different geomorphic settings to model estuarine habitat in a larger geographic framework and test how tectonic setting, oceanographic setting, climate, and margin age affect habitat distribution during sea-level change. In addition, independent of estuaries we explore paleobiologic (e.g. Olsson, 1961) and neontolologic effects of sea-level change on evolution, and test the relation between overall shelf area and species richness using data of 1721 fish species. We find 82% global reduction of estuarine habitat abundance at lowstand relative to highstand, and find large habitats change in size much more than small habitats. Consistent with prior work, narrow continental margins have significantly less habitat at highstand and lowstand than wide margins, and narrow margins significantly associate with fore-arc settings, effectively linking tectonic setting to habitat abundance. Surprisingly, narrow margins host greater species richness, a finding which violates the canonical species-area relation. This finding can be explained if: 1) the physical isolation imposed by narrow margins facilitates the formation of new species over time; 2) the size-stability of small habitats, which disproportionately occur on narrow margins, accumulate and retain species extirpated in the more variable habitats on wide margins; or 3) the smaller habitats on narrow margins facilitate greater species richness through greater habitat heterogeneity. These results are generally at odds with prior interpretations, but the combination of richness data and population genetic principles offer a different perspective on these long-studied questions. Finally, we emphasize that the nuance of Pleistocene-Holocene sea level oscillations should be more explicitly considered in genetic studies.

Introduction

How Earth's climate and surface processes influence genetic and morphological change of species over time is a leading question within evolutionary biology, paleontology, and geobiology. Earth processes that physically isolate populations promote allopatric genetic divergence—the genetic relatedness of the isolated populations drift apart over time in the absence of gene flow (the exchange of genetic material). In addition, change in the environment can result in biological adaptations as individuals with qualities that maximize survival and reproduction on the new landscape increase in number through disproportionate reproductive success. This can lead to ecologically differentiated populations (Berner et al., 2010; Crespi and Nosil, 2013; Dettman et al., 2008; Gray and Goddard, 2012; Klosterman et al., 2011; Roesti et al., 2012). If divergence is protracted or selection is strong enough, then isolated or differentially adapted populations may become separate species, increasing biodiversity (Coyne and Orr, 2004). Based on genetic evidence, processes known to yield such biological effects include Northern Hemisphere glaciations (Hewitt, 2000, Hewitt, 2004; Jansson and Dynesius, 2002), mountain building (Antonelli et al., 2018; Craw et al., 2015; Hoorn et al., 2010), geographic rainfall asynchrony/variability (Quintero et al., 2014; Thomassen et al., 2013), tectonic rifting (Clark, 2012; Lieberman, 1997), and reorganization of river drainages (Dias et al., 2014; Dolby et al., 2019; Goodier et al., 2011; Hershler and Liu, 2008; Hershler et al., 1999), among others. Recent work also recognized how changes in land configuration during sea level oscillations has affected population connectivity of land-dwelling species (Papadopoulou and Knowles, 2017; Sawyer et al., 2019). These types of studies contribute to our understanding of the cause-effect relation between physical processes and evolutionary responses. Such cause-effect relations are important because they ultimately reveal the ways in which Earth shapes life and how much of biological evolution is extrinsically (versus intrinsically) forced. When knowledge from such relations is detailed and mechanistic, it can be used to make predictions about the biological effects of such processes over deeper timescales and in new geologic settings.

Processes thought to be important in shaping marginal marine ecosystems include the formation of new, isolated marine embayments or habitats such as the Gulf of California and Red Sea (DiBattista et al., 2016; Dolby et al., 2015; Lau and Jacobs, 2017), currents that facilitate or limit dispersal of individuals such as in the Indo-Pacific Coral Triangle (Barber et al., 2006; Davies et al., 2014; Kool et al., 2011), and formation of physical barriers (Hobbs et al., 2009) such as the Isthmus of Panama that not only bisected and isolated marine populations, but also initiated the large-scale reorganization of marine currents (O'Dea et al., 2016; Schneider and Schmittner, 2006). Often the effects of these processes are mixed because marine species vary greatly in dispersal capacity and their response to external influences (Bernardi, 2013; Kelly and Palumbi, 2010; Marko, 2004).

From the paleontological literature, Dall (1890) and Olsson (1961) proposed that lowered sea level stands (eustatic regressions) reduced the habitat footprint of marginal marine species and led to local or regional extinctions. This idea was based on the greater proportion of shallow shelf area in the Caribbean that hosted lower molluscan diversity relative to the Panamanian region that has less shallow shelf area and higher molluscan diversity even though they originated from the same set of ancestors before the isthmus closed. The argument here is that shallow-shelf Caribbean areas were vulnerable to extirpation via lowered Pleistocene sea levels (Olsson, 1961). Stanley, 1986, Stanley, 1984 contradicted this hypothesis in favor of temperature change as the causal explanation using the rationale that regression-based extinctions should be geographically global while temperature-based extinctions should be regional.

Using a neontological approach, several different studies recently evaluated how late Pleistocene sea-level change affected marginal marine populations (Neiva et al., 2018; Waltari and Hickerson, 2013). Using genetic data and habitat modeling of coastal estuaries, Dolby et al., 2018, Dolby et al., 2016 proposed a mechanism that relates shelf morphology to species diversification. The framework proposed that tectonic and sedimentary processes control the geomorphic properties (e.g., slope) and geologic substrate (e.g., unconsolidated sediments, bedrock, etc.) of continental shelves (Algeo and Wilkinson, 1991), and those properties in turn control where many types of coastal habitats can form (e.g., estuaries, mangroves, beaches). When sea-level oscillates against the heterogeneous shelf area it changes the distribution of habitats, which directly controls genetic connectivity (i.e. whether individuals can move between neighboring populations to reproduce). This either results in habitat connectivity and genetic similarity, or in habitat isolation and genetic divergence, depending on the geomorphic setting and the dispersal capacity of the species. Fewer, smaller habitats occur along tectonically active (steep) coastlines than broad coastlines and because dispersal is negatively correlated with geographic distance, they found more genetically differentiated populations on steep coasts (Dolby et al., 2018). Overall, this framework suggests that large-scale geologic processes exert top-down control on the connectivity and genetic evolution of coastal species and could possibly produce new species when extrapolated over deeper geologic time.

However, the previous work was done over a regional setting (western coast of North America). Here, we test the effects of tectonic setting, oceanographic setting, climate, and margin age on habitat distribution during sea-level change at a global scale. Second, we test Olsson's areal restriction hypothesis by comparing overall shelf area to species richness data and use population genetic and speciation theory to integrate these finding with the prior results (Dolby et al., 2018; Dolby et al., 2016). To do so, we calculate estuarine habitat abundance during Pleistocene sea level oscillations along unglaciated coastlines globally (Fig. 1). We integrate these results with existing datasets and perform a suite of statistical tests to understand the relation of tectonic, sedimentary, and oceanographic properties on habitat abundance and species richness patterns of 1721 fish species.

Section snippets

Estimating habitat abundance

To estimate putative estuarine habitat we used the SRTM30_PLUS digital elevation model (DEM; Becker et al., 2009) because it integrates topographic and bathymetric data worldwide at relatively high resolution (30 arc-second, about 1-km resolution), making it a good resource for global analysis of marginal marine areas. It also allows for consistency with previous work (Dolby et al., 2018; Dolby et al., 2016). In a geographic information system (GIS) we calculated putative estuarine habitat area

Statistical relations

Our GIS analysis produced habitat abundance estimates for 14 depth bins in each of 277 manually curated coastal regions (Fig. 3). The habitat data were overdispersed based on a Pearson Chi Square test that yielded a value of 2.3; overdispersion was therefore accounted for in the GLMs. Broadly, we found several significant correlations between habitat abundance and coastal characteristics (summarized in Table 2). MWU tests revealed significantly greater habitat abundance on wide shelves (≥

Discussion

We performed a GIS analysis to estimate putative estuarine habitat abundance through time as a function of sea level change and continental shelf geomorphology across continents with different tectonic, sediment, climatic, and oceanographic properties (North America, South America, Africa, India, Australia, and Japan; Fig. 9, Fig. 1). Results show on average there is 82% less estuarine habitat at sea-level lowstand (140 mbsl; 19.5 ka) relative to present-day sea level (0–6.5 ka; Fig. 3). Recent

Conclusions

Nearly 60 years after Olsson (1961) proposed the areal restriction hypotheses to explain regional extinctions in the fossil record, we provide global evidence drawn from geological and biological records to tentatively suggest that tectonic setting directly controls the physical size and distribution of marginal marine habitats on coastlines worldwide, and that habitat abundance decreases during sea level regression.

We find fore-arc settings produce narrow continental shelves that limit

Author contributions

GAD and DKJ co-conceived of this project. AMB and GAD performed GIS analyses, GAD performed statistics and drafted the paper. DKJ and SEKB contributed to the paper and all authors gave final approval.

Data availability & appendices

Third part data sources are cited in the methods. All other products associated with this study are bundled in a Harvard's Dataverse: https://dataverse.harvard.edu/dataverse/DolbyGPC. Products include: spreadsheet of all habitat data with spatial join data, KML layers, html archiving of spatial autocorrelation and JMP statistical results.

Funding

GAD was supported by NSF-EAR-1925535 and the College of Liberal Arts and Sciences at Arizona State University. DKj was supported by NASA Astrobiology (NNA13AA90A) Foundations of Complex Life

Declaration of Competing Interest

We have no conflicts of interest to declare.

Acknowledgements

We thank P Barber, R Hechinger, A Munguia Vega, and B Dorsey for helpful conversations and discussion of this study, as well as BB Nyberg and D Muller for data support. We thank K Kusumi for providing additional support and Jonathan Richmond and Geerat Vermeij for helpful reviews that improved the quality of this manuscript. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

References (127)

  • R. Moucha et al.

    Interplay between dynamic topography and flexure along the U.S. Atlantic passive margin: Insights from landscape evolution modeling

    Glob. Planet. Chang.

    (2017)
  • B. Nyberg et al.

    Global distribution of modern shallow marine shorelines. Implications for exploration and reservoir analogue studies

    Mar. Pet. Geol.

    (2016)
  • B. Nyberg et al.

    The distribution of rivers to terrestrial sinks: Implications for sediment routing systems

    Geomorphology

    (2018)
  • K. Pedoja et al.

    Quaternary coastal uplift along the Talara Arc (Ecuador, Northern Peru) from new marine terrace data

    Mar. Geol.

    (2006)
  • A.C. Algar et al.

    Evolutionary assembly of island faunas reverses the classic island–mainland richness difference in Anolis lizards

    J. Biogeogr.

    (2011)
  • T.J. Algeo et al.

    Modern and ancient continental hypsometries

    J. Geol. Soc. Lond.

    (1991)
  • L. Anselin et al.

    GeoDa: an introduction to spatial data analysis

    Geogr. Anal.

    (2006)
  • A. Antonelli et al.

    Geological and climatic influences on mountain biodiversity

    Nat. Geosci.

    (2018)
  • P.H. Barber et al.

    Comparative phylogeography of three codistributed stomatopods: origins and timing of regional lineage diversification in the coral triangle

    Evolution

    (2006)
  • A.C. Bastos et al.

    SHELF morphology as an indicator of sedimentary regimes: a SYNTHESIS from a mixed siliciclastic-carbonate shelf on the eastern Brazilian margin

    Journal of South American Earth Sciences

    (2015)
  • J.J. Becker et al.

    Global bathymetry and elevation data at 30 arc seconds resolution: SRTM30_plus

    Mar. Geod.

    (2009)
  • J. Belmaker et al.

    Determinants of the steep species–area relationship of coral reef fishes

    Coral Reefs

    (2006)
  • S.E. Bennett et al.

    Oblique rifting ruptures continents: example from the Gulf of California shear zone

    Geol

    (2014)
  • S.E.K. Bennett et al.

    Transtensional rifting in the proto-Gulf of California near Bahia Kino, Sonora, Mexico

    Geol Soc America Bull

    (2013)
  • G. Bernardi

    Speciation in fishes

    Mol. Ecol.

    (2013)
  • D. Berner et al.

    Constraints on speciation suggested by comparing lake-stream stickleback divergence across two continents

    Mol. Ecol.

    (2010)
  • R.W. Bialas et al.

    How sediment promotes narrow rifting: application to the Gulf of California

    Tectonics

    (2009)
  • A. Birand et al.

    Patterns of species ranges, speciation, and extinction

    Am. Nat.

    (2012)
  • R.S. Bivand et al.

    Comparing implementations of global and local indicators of spatial association

    TEST

    (2018)
  • R.S. Bivand et al.

    Applied Spatial Data Analysis with R

    (2013)
  • R. Bivand et al.

    Package “rgdal.”

  • B.W. Blackwelder

    Late cenozoic stages and molluscan zones of the U.S. Middle Atlantic Coastal Plain

    Memoir, the Paleontological Society

    (1981)
  • C. Heine et al.

    Rift migration explains continental margin asymmetry and crustal hyper-extension

    Nat. Commun.

    (2014)
  • W.R. Buck

    Modes of continental lithospheric extension

    J. Geophys. Res.

    (1991)
  • C.A. Burk et al.

    The Geology of Continental Margins

    (2013)
  • T.B. Chalk et al.

    Causes of ice age intensification across the Mid-Pleistocene transition

    Proc. Natl. Acad. Sci. U. S. A.

    (2017)
  • J.D. Chaytor et al.

    Measuring vertical tectonic motion at the intersection of the Santa Cruz-Catalina Ridge and Northern Channel Islands platform, California Continental Borderland, using submerged paleoshorelines

    Geol Soc America Bull

    (2008)
  • P.M. Chittaro

    Species-area relationships for coral reef fish assemblages of St. Croix, US Virgin Islands

    Mar. Ecol. Prog. Ser.

    (2002)
  • N. Clark

    The effects of the spreading of the Central Atlantic during the Middle Jurassic on dinosaur faunas

    OUGS Journal

    (2012)
  • P. Clift et al.

    Controls on tectonic accretion versus erosion in subduction zones: Implications for the origin and recycling of the continental crust

    Rev. Geophys.

    (2004)
  • E.F. Connor et al.

    The Statistics and biology of the species-area relationship

    Am. Nat.

    (1979)
  • J.A. Coyne et al.

    Speciation

    (2004)
  • D. Craw et al.

    Rapid biological speciation driven by tectonic evolution in New Zealand

    Nat. Geosci.

    (2015)
  • T.M. Cronin

    Speciation and Stasis in Marine Ostracoda - Climatic Modulation of Evolution

    Science

    (1985)
  • W.H. Dall

    Contributions to the Tertiary Fauna of Florida: With Especial Reference to the Miocene Silex-Beds of Tampa and the Pliocene Beds of the Caloosahatchie River

    (1890)
  • S.W. Davies et al.

    Exploring the role of Micronesian islands in the maintenance of coral genetic diversity in the Pacific Ocean

    Mol. Ecol.

    (2014)
  • J.R. Dettman et al.

    Divergent adaptation promotes reproductive isolation among experimental populations of the filamentous fungus Neurospora

    BMC Evol. Biol.

    (2008)
  • M.S. Dias et al.

    Global imprint of historical connectivity on freshwater fish biodiversity

    Ecol. Lett.

    (2014)
  • J.D. DiBattista et al.

    A review of contemporary patterns of endemism for shallow water reef fauna in the Red Sea

    J. Biogeogr.

    (2016)
  • W.R. Dickinson

    Clastic sedimentary sequences deposited in shelf, slope and trough between magmatic arcs and associated trenches

    Pac. Geol.

    (1971)
  • Cited by (10)

    • Integrating Earth–life systems: a geogenomic approach

      2022, Trends in Ecology and Evolution
      Citation Excerpt :

      Regional rainfall variations may be influenced by orographic effects produced by interactions of water-bearing air masses and topography, with rain shadows commonly forming on the lee side of large mountain ranges. Biologically relevant examples of climate change include variations in annual or seasonal temperature and precipitation, exposure or flooding of continental shelves due to glacio-eustatic sea-level changes [37–39], and changes in ocean circulation that shift the location, timing, and seasonality of monsoonal precipitation [40–42]. The formation and maintenance of climate dipoles, or two contrasting but adjacent regional climatic conditions, are particularly important because they exert different selection pressures in proximity that can lead to differential adaptation (e.g., through selective sweeps) or temporal isolation of species or populations [43,44].

    • Pleistocene expansion, anthropogenic pressure and ocean currents: Disentangling the past and ongoing evolutionary history of Patella aspera Röding, 1798 in the archipelago of Madeira

      2021, Marine Environmental Research
      Citation Excerpt :

      During the Quaternary, more specifically during the Last Glacial Maximum (LGM; 26500–19000 Kyr BP), the sea retreated over 100 meters below current sea level, connecting previously separated islands and exposing large coastal areas (Pirazzoli 1997; Clark and Mix 2002; Clark 2009). With the melting of the ice and rising sea levels during the Holocene (11500 Kyr BP up the present), previously connected landmasses became isolated, affecting the connectivity of species (Mairal et al., 2015; Patiño et al., 2015; Fernández-Palacios et al., 2016; Weigelt et al., 2016; Dolby et al., 2020). As consequence of these changes, populations became fragmented, as occurred with some of the islands in the archipelago of Madeira (Rijsdijk et al., 2014; Fernández-Palacios et al., 2016).

    • Multiple-scale processes shape the population genetics of Tehuelche scallop, Aequipecten tehuelchus, in Northern Patagonia

      2021, Fisheries Research
      Citation Excerpt :

      Hence, a possibly explanation for the genetic signals might be that the population expansion may have followed colonization of newly inundated coastal zones after the LGM (9,000–12,000 ya; Ponce et al., 2011). Such timing fits with the sea-level stabilization at the present-day height about 6500 ya (Dolby et al., 2020). In the Southwestern Atlantic Ocean, a similar genetic signal of population expansion was found for a scorched mussel (Trovant et al., 2015), a nemertean (Alfaya et al., 2012), and a limpet (de Aranzamendi et al., 2011), attributed to demographic process after the LGM.

    View all citing articles on Scopus
    View full text