Evolutionary biogeography of Aotearoa

When one of us (GPW) arrived in Aotearoa in 1988, Carolyn Burns recommended Prehistoric New Zealand (Stevens et al. 1988) as an introduction to the biological history of Aotearoa. And what a read that was. GPW learned about a 13-kg eagle that dive-bombed megapodes, giant wētā living on forest giants, marine inundation, glaciers calving to the sea: all on a mammal-free landmass, torn apart by earthquakes, buried by slips and pierced by volcanoes, sitting on the Pacific rim of fire. This was quite a contrast to the relatively inert, species-depauperate, glacier-scoured northern Europe (Wallis and Arntzen 1989; Hewitt 1996) with which he was familiar. This was biogeography on steroids, reinforced by the civil defence emergency tsunami warning signs at his daughter’s first primary school: the land of Orcs waiting to be cast. When Allan Wilson had asked GPW in 1985 how his findings on European Triturus newts reflected the biogeography of Europe, he had to admit to not having thought much about it. This was certainly not going to be the case in Aotearoa. A population geneticist was bound to ask how the history of the land mass had affected genetic structuring of species, and a phylogeneticist how it was reflected in species relationships.

The other Hitchhikers’ Guide (New Zealand Journal of Zoology 16(4)) to this new sea land, although rather more sedate on the surface, was equally audacious in its prose. This Panbiogeography Special Issue spawned an early phylogeographical PhD project with an hypothesis (Patrick 1989) on the early Tertiary stranding of coastal forms in current Central Otago salt pans (Emerson and Wallis 1995; Emerson et al. 1997). The panbiogeographic method, however, had gained little traction internationally: was this really mainstream biogeography of Aotearoa? Subsequent reading of Fleming’s work suggested not. Panbiogeography has been widely critiqued (Cox 1998; McDowall 2004; McGlone 2005, 2016; Briggs 2007; Waters et al. 2013) and has since rather fallen by the wayside.

TRB, on the other hand, has deeper roots in Aotearoa. Bemused by the ‘complete vicariance’ versus ‘everything dispersed’ extremes between which biogeography in Aotearoa seemed to oscillate, he too was drawn to the more balanced views of Fleming and later Gibbs (2016). These authors demonstrated that while each taxon has its own story, defying simplistic or ideological approaches, it is possible to make general inferences with phylogenies from large numbers of taxa spanning the biota. TRB read Fleming’s seminal 1979 book as a student and its pluralistic nature had an enduring influence on him. Fleming’s views were grounded in decades of observations from natural history and palaeontology: he was probably the first to produce a comprehensive overview of the Aotearoa situation. His series of increasingly detailed papers (Fleming 1962) and volumes (Fleming 1979) also had a strong historical, evolutionary framework. Perhaps still the most comprehensive and definitive review of Aotearoa biogeography appeared in this era (Kuschel 1975), involving 19 authors, all leaders in their fields. This 689-page tome was, however, typical of the period in having taxonomically arranged subject material, rather than addressing particular issues or questions, making true synthesis difficult.

The evolutionary biogeography of Aotearoa was thrust into the limelight on the global stage in a Special Feature (Trends in Ecology of Evolution 8(12):429–457), comprising six papers on biogeography, evolution and conservation of the flora and fauna. Each of these papers became a citation classic and prompted renewed focus on the biogeography of Aotearoa. Over the last decade or so, several notable reviews have appeared in local and international journals (Trewick et al. 2011; Trewick and Bland 2012; Buckley et al. 2015; Lee et al. 2016; Cole and Wood 2018; Heenan and McGlone 2019; Waters et al. 2020), taking on specific topics on the historical biogeography of Aotearoa. A (zoologically-biased) bibliography of the area has been compiled by Spencer and Rawlence (2020).

The breadth of biogeography and the increasing application of specialist tools to its questions makes it a challenging area to review and synthesise. This current volume attempts to bring together a diverse set of topics and authors that reflect the current state of the biogeography of Aotearoa as a whole. It does not seek to be exhaustive, since major recent reviews already exist on some topics, such as the Oligocene Marine Transgression, for example (Special Issue of New Zealand Journal of Geology and Geophysics 57(2), Wallis and Jorge 2018). We hoped to include papers with a focus on palaeontology, as well as the marine and freshwater habitats, but these were not forthcoming. Authors were asked to “synthesise and review” in preference to presenting new empirical studies on a specific taxon or region. The papers are intended to be both reflective and prospective.

The application of genetics and phylogenetics to examining fine population structure in a geographical and geological context has allowed the testing of many long-standing biogeographic hypotheses in Aotearoa. Marske and Boyer (2024) focus on how phylogeography has helped us to understand the effects of the Last Glacial Maximum (LGM) evidenced by structuring in our flora and fauna. They provide a comprehensive review of studies since the first major review 15 years ago (Wallis and Trewick 2009), using a framework of several geographical biotic boundaries/breaks in Aotearoa: Northland line, kauri line, Taupō line, Cockayne’s line, beech gap, Waitaki River and Clutha River. The complex geological and climatic history of the country, and its manifold interactions with species with quite different habitat requirements, has led to a great variety of patterns. Furthermore, some related species that one might expect to respond in similar ways show disparate patterns, possibly reflecting threshold effects based on subtle differences, or simply the natural stochasticity of biological systems.

Alpine zones are renowned for the diversity that they house and generate (Spehn et al. 2011), particularly notable in the flora and insect fauna. Compared to most other mountain systems around the world, the Southern Alps formed quite recently, so give us a chance to analyse the recent formation and radiation of alpine forms. Buckley et al. (2024) review our knowledge of this fascinating biota, which generally shows recent origins from lowland sister groups, in keeping with major Miocene-Pleistocene uplift. Melanism, freeze-avoidance/tolerance, host specificity and brachyptery are widespread adaptive responses to the alpine environment, the latter two of which can promote further speciation. The extreme and often fragile nature of alpine ecosystems, particularly with respect to mammalian grazing, introduced pests, erosion and climate change, poses a threat to this significant part of our biodiversity.

Hybrid zones (geographic regions where genetically distinct forms/taxa meet, mate and produce at least some individuals of mixed ancestry) can provide important information to biogeographers, as these zones typically owe their very existence to secondary contact of closely related parental forms (Harrison 1993). Their positioning can thus reflect change in distribution and abundance relating to alteration of many biogeographic factors, including climate, ecology, sea-level and tectonic processes. These regions have been the subject of genetic analysis in many different groups and Shepherd et al. (2024) summarise the conclusions of these studies. Although geographically widespread, they include the Taupō volcanic zone (extinction through vulcanism), southern North Island and northern South Island (distribution shifts during Pleistocene climate cycles), central South Island (extirpation during Pleistocene glaciation), Banks Peninsula (volcanic island formation and subsequent connection to mainland) and southern South Island (refugia during Pleistocene glaciation and climate cycling). The biotic effects of changing landscape depend very much on the biology of the taxa involved. For example, the repeated opening of Cook Strait has given rise to genetically distinct terrestrial forms to the north and south, whereas its repeated closure has led to genetically distinct marine forms to the east and west. Likewise, while glaciation can cause wholesale extinction of lowland forms, it can be a creative force leading to speciation of alpines (Wallis et al. 2016). As with Marske and Boyer (2024), diversity is the order of the day for hybrid zones.

Just as genetic analysis of hybrid zones can help us determine the nature and timing of events on a scale of 10⁵–10⁶+ years ago, ancient DNA technology allows us to access information held in subfossil remains of extinct taxa to broaden our understanding of their origins and relationships in deep time, but also to help reconstruct the more recent make-up of pre-human communities and habitats. Verry et al. (2024) summarise the impact of this approach to our understanding of NZ faunal origins, which has in particular been applied to birds, and precipitated a big re-think regarding ratite evolution. Detailed sampling through time has brought light to bear on patterns of species (birds, marine mammals) decimation and recolonisation after human arrival, implying spatial and temporal shifts in human diet. Improving technical sensitivity and accessibility should allow interrogation of a wider array of sample types, such as coprolites and soils, providing “palaeoenvironmental DNA” to help reconstruct the plant, invertebrate, fungal and even microbial diversity of pre-human Aotearoa.

With reference to the diversity of patterns seen, which to some extent reflect ecological and physical trait differences among taxa (Zamudio et al. 2016), one can also ask: How does a changing environment sort traits through space and time? Dale et al. (2024) approach this question in three genera of woody plants that occur across forest, open and alpine biomes. Radiations extend back to the mid-Cenozoic, with forest assumed to be the ancestral habitat. Trait values are indeed filtered by biomes, but there is little evidence for trait evolution in response to a shift in biome. Where this change happens, a shift to extreme (alpine) habitats is usually involved. One can predict allocation of species to biomes using just five traits. This is highly ambitious research, as it attempts to combine pattern and process through time, uniting historical and functional biogeography. As phylogenetic and trait data accumulate for taxa outside of plants and vertebrates, we hope to see more such applications.

Discussion of the factors important in ecological succession, and specifically the importance of incumbency and priority effects, has been debated for over a century in ecology (Clements 1916). Using the entire conifer and angiosperm flora to address its significance in the ontogeny of plant communities in Aotearoa, McGlone et al. (2024) find little evidence for eco-evolutionary priority as evidenced by either increased adaptation or speciation rates in lineages of early arrivals in an ecosystem. Instead, the most speciose groups tend to be newer arrivals, and traits such as tree height and biotic dispersal of seed are better predictors of range size. Other studies finding priority effects appear to be based on smaller regions with limited numbers of taxa. Perhaps priority effects are short-lived, eventually erased by newly-arriving taxa over time, an idea that should be testable.

Finally, Costello (2024) takes the broadest approach to a study of total species diversity. Aotearoa is highly unusual in that at least 50% of its named species are endemics. Six predictions are made with respect to predicted levels of endemism across groups, specifically that it should be lower for: mobile, marine, microscopic, megafauna, flighted and pelagic groups. The same follows for overall amount of diversity, with the exception of mobility, which, it is argued, should promote diversity. Although local species richness across different groups is highly correlated with global richness in those groups, the amount of endemism shown is not. In general, the six predictions hold, with one significant exception: mobile/flighted groups showed over 90% endemism, against predictions. This is likely explained by the widespread phenomenon of flight-loss in our birds and insects; i.e. flighted groups have a better chance of finding and radiating into new niches than sessile groups, provided that gene flow is severed by a reduction in mobility once there. This argument can be generalised to any dispersal trait, for example loss of diadromy in fishes (Allibone and Wallis 1993). As with McGlone et al. (2024), intrinsic (dispersal) traits are better biogeographical predictors than neutral or phylogenetic (as reflected by taxonomy) effects. Stephen Gould once cautioned that: “the more you encompass the more formless you become” (Gould 1981). The findings of this paper are likely to be controversial and to invite alternative explanation.

If one were forced to summarise our current understanding of the history of the biota of Aotearoa, the popular aphorism “it’s complicated” might spring to mind. Given the complexity of the landscape and the processes that actively mould it, together with the manifold ways populations of different taxa in different habitats are likely to respond to these processes, this should be no surprise. The way forward to a deeper understanding is to focus region by region using a multi-taxon approach. Just as a systematist needs many characters for a more reliable phylogeny, a biogeographer needs to encompass as many groups as possible for a fuller understanding. In phylogeography, it is the assimilation of information from many single-taxon studies over the last 30 or so years that have given our present depth of knowledge.

New techniques are important to advancing all fields: of particular relevance here are next generation sequencing (NGS; Campbell et al. 2022; Vaux et al. 2023) and the genomic tools to analyse these data, and new methods of data analysis (Ronquist and Sanmartín 2011; Klaus and Matzke 2020; Hackel and Sanmartín 2021). The application of these to the biogeography of Aotearoa context is still in its infancy. These approaches will give us better acuity and resolution, but may not significantly change the phylogeographic pattern that we already observe. Genomics, however, offers an opportunity to determine the genetic basis of some phenotypic traits and adaptations (Dennis et al. 2015; McCulloch et al. 2019; Guhlin et al. 2023). When combined with ancient DNA analysis and the types of approaches used by Dale et al. (2024) and McGlone et al. (2024) above, a much richer picture of the construction of our flora and fauna through time should emerge.

We also call for an intense focus on the natural history of the species of Aotearoa. Most of the studies presented here are based on deep understanding of the biology of native species, where they are found, their fine-grained habitat preferences, life cycle, morphology, physiology and ecological relationships. Without this knowledge it is impossible to make sense of the genomic data, which are now so easily obtained, or geographic distributions, which have accumulated for so many species. Synthesising this natural history knowledge with ‘big-data’ and new analytical tools, will lead to the greatest advances in Aotearoa biogeography.

Another important component of the success and impact of evolutionary biogeography in Aotearoa has been a multi-disciplinary approach, particularly biologists and geologists working together (e.g. Waters et al. 2001; Trewick et al. 2007). In this way, the hypotheses of each discipline are reciprocally tested with data from the other, rather than in the more usual one-way direction that tends to happen in a single discipline. This process leads to a richer mining and mutual support of ideas, befitting the field whose origins date to the great polymathic natural scientists of their day (Darwin 1859; Wallace 1876; Hutton 1904).