New Zealand’s Forgotten Grazing Guild, and How to Rewild Without Moa?

           While the most famous large herbivores of New Zealand were undoubtedly the various species of extinct moa, they were not the only ones. Moa occupied many different habitats, but they seem to have been rather consistent in their niche, with most species showing adaptations towards browsing on twigs and leaves, or the occasional fruit. Browsers and frugivores are still represented by parrots (Cyanoramphus, Nestor, Strigops), pigeons (Hemiphaga), and kokako (Callaeas), but these are all primarily forest-dwelling birds. New Zealand did have native grassland herbivores as well, but they were represented by very different taxonomic groups, namely the Phasianidae, Rallidae, and Anatidae.

           The Phasianid Coturnix was the only native genus of Galliform (or chicken-like) bird native to New Zealand. The native New Zealand quail or koreke (?Coturnix novaezelandiae) went extinct in the late nineteenth century due to hunting by humans, introduced predators, and possibly avian diseases carried by chickens and other alien gamebirds. One of these introduced gamebirds was the related brown quail (Coturnix ypsilophora), which still survives in parts of the North Island. Some authors have proposed that the brown quail might be fulfilling a similar ecological role to its extinct congener, but there are a few points to dispute this. First, the brown quail is not the closest living relative of the koreke, which is instead closest to (and formerly considered conspecific with) the stubble quail, (Coturnix pectoralis). Both the brown quail and stubble quail are native to Australia and as sympatric taxa, their habitat preferences and ecological roles necessarily differ. Both eat primarily seeds, foliage, and invertebrates, but stubble quails are adapted primarily to grasslands, like the koreke was, whereas brown quails are found primarily in shrublands. Stubble quails are also closer in size (though still smaller) to the koreke, and better suited to more temperate climates and high altitudes. I don’t think it’s necessary to remove brown quail from New Zealand, but the introduction of stubble quail would perhaps more adequately fill the koreke’s niche, and the two quail species have already proven capable of coexistence in their natural range. Both taxa are potentially important as seed disturbers, invertebrate predators, and grassland/undergrowth engineers.

           While primarily omnivorous in most species, several of the grazing genera of New Zealand were members of the rail family, Rallidae, including coots (Fulica), swamphens (Porphyrio), and waterhens (Tribonyx). The New Zealand coot (?Fulica prisca) was flightless and weighed almost 2 kg, over three times larger than the extant Eurasian coot (Fulica atra) which colonized New Zealand centuries after the former went extinct. While the extinct species likely descended from a similar colonization event, they are thought to have been far more terrestrial in habit while the living coots, even the large South American species, are always tied to water, limiting their potential effectiveness as herbivores. The New Zealand waterhen (?Tribonyx hodgenorum) was similarly flightless and terrestrial when compared to its likely ancestor, the wetland-based Australian waterhen (Tribonyx ventralis) but was of a similar size. The swamphens were represented in New Zealand by three species, two of which, the takahe and the moho, were large, flightless, and adapted towards grazing in open habitats. The moho, or North Island takahe (?Porphyrio mantelli) is extinct, but might be replaced ecologically by its South Island relatives, (Porphyrio hochstetteri), which have already been translocated to several North Island sites as a conservation measure. The third swamphen, the purple swamphen or pukeko (Porphyrio melanotus) is also a recent colonist, like the coot, and is similarly more confined to wetland habitats. So, out of the four native herbivorous, flightless rails, ?Fulica prisca, ?Tribonyx hochstetteri, Porphyrio hochstetteri, and ?Porphyrio mantelli, three are extinct, and only one has an obvious surrogate. Coots and translocated populations of waterhen might eventually take on more terrestrial characteristics, at least in areas free of introduced predators, but in the meantime would be largely relegated to wetland environments. It might instead be possible to investigate the possibility of introducing flightless taxa from other islands. Aside from the takahe, there are only three flightless, extant, mostly herbivorous rails adapted to temperate climates: the giant coot (Fulica gigantea), Gough moorhen (Gallinula comeri), and the Tasmanian nativehen (Tribonyx mortierii). The giant coot is of a similar size, or even a bit larger, than the New Zealand coot, but lacks the same terrestrial adaptations and is confined to wetlands. The Gough moorhen is well adapted to grassland and shrubland habitats, and Gallinula is the most closely related genus to Fulica, but at 0.5 kg the Gough moorhen is about a quarter the size of the New Zealand coot. The Tasmanian nativehen is geographically closest to New Zealand, and is far more terrestrially inclined than the congeneric Australian waterhen, but weighs roughly 1.25 kg, five times as much as the New Zealand waterhen. The correct course of action for these species is unclear. I think the best way forward, for now, is to employ takahe (by far the largest extant rail) as native grazers on both main islands.

           Before human arrival, there were 12 genera of waterfowl native to New Zealand, more than half of which are no longer present. Of these, three are considered to be/have been terrestrial herbivores, the rest being primarily aquatic herbivores/omnivores. The only remaining representative of the New Zealand grazing waterfowl is the Paradise shelduck, or putangitangi (Tadorna variegata), which remains common today. Its ability to fly may have made it more resilient to predation by humans and introduced predators than the other two species. Finsch’s Duck (?Chenonetta finschi), a large, flightless bird related to the Australian wood duck, (Chenonetta jubata) went extinct shortly after the arrival of the Polynesians (though a possible record exists from shortly after European arrival), and is known to have been hunted from the archaeological record. Based on its fossil distribution and its morphological similarity to the wood duck, Finsch’s duck was likely a herbivorous resident of open habitats, eating primarily grass and the occasional fruit. Interestingly, the Australian species has become a semi-common vagrant to New Zealand in the past century, and in 2015 and 2016 a pair successfully bred there. By encouraging further colonization and possibly translocating animals directly, the genus could become re-established. The last remaining group of herbivorous waterfowl to be discussed were also the largest native herbivores aside from the moa and the largest grazers of any New Zealand taxa. The New Zealand geese (Cnemiornis sp) were giant, weighing around 15 kg. Their closest relatives are the Cape Barren (Cereopsis novaehollandiae) geese of southern Australia, which weigh about a third as much. Interestingly, Cape Barren geese have become semi-established in New Zealand after being introduced, and may potentially be providing many of the same ecosystem services. Herbivorous waterfowl serve important ecological functions, especially in systems where there are no herbivorous mammals. Through their foraging activity and disturbance they can create landscape heterogeneity, transport seeds and nutrients, and potentially maintain open spaces.

           Unfortunately, it was likely the moa birds that did a lot of the legwork in that regard. Without any large, terrestrial browsers, encroachment by trees and shrubs will eventually make grasslands into forests, except in coastal, wetland, and alpine situations where the climatic factors prevent it. The presence of the aforementioned grazing herbivores, along with other species adapted to open landscapes, in lowland and inland habitats suggest that there was greater heterogeneity in the past. Moa birds likely cleared the understory and slowed encroachment in clearings made by fire, weather, or other factors. This would theoretically have allowed grazing species to more easily exploit mixed habitats. Grassland species that survived the first colonization event would have been able to exploit open habitats created by the burning and clearing of woodlands. Those species that also survived the second colonization could instead live on agricultural land or livestock pasture, possibly becoming a pest in the case of the shelduck. But unfortunately, the succession of woodlands can still be a problem when we talk about rewilding in New Zealand. It isn’t like Eurasia or the Americas, where we can substitute extinct herbivores with close relatives or ecological analogues, because there really aren’t any for the moa. Mammalian herbivores have entirely different effects on the vegetation, and the living large ratites like emus and ostriches are all grazers. Probably the closest equivalents niche-wise are giant tortoises, which have been used to replace extinct waterfowl in Hawaii, but that just isn’t an option in New Zealand’s cooler climate. The result is that New Zealand’s offshore islands and fenced reserves, currently the last refuge for much of its native biodiversity, are mostly closed-canopy forests or in the gradual process of becoming so. For many native taxa this is ideal, but not for all. Many sanctuaries that keep and breed Takahe will eventually have to get rid of them, for the simple reason that there won’t be enough grassland to sustain them. The same may eventually be true for several native birds and reptiles that prefer open or semi-open habitats. Even species that primarily live in forested environments will take advantage of pastures and clearings when they get the chance.

           The question that then needs to be asked is whether it is desirable to maintain open habitats in New Zealand Rewilding areas, and if so, how would we go about doing so? Open areas could be created or maintained manually, but this would be antithetical to the basic premise of Rewilding, i.e. minimum intervention and maximum natural process. It would certainly be interesting to do some grazing projects with native grazers and proxies, i.e. Coturnix pectoralis, Chenonetta jubata, Tadorna variegata, Porphyrio hochstetteri, and Cereopsis novaehollandiae, to measure what, if any, trophic effects they have on the habitat. I think it is very unlikely that they could successfully maintain open areas long-term, but they might be able to slow succession to some degree. More likely, and just as importantly, they might be able to improve the productivity and diversity of existing grassland habitats. Like in the rest of the world, the extinction of browsing megafauna increased the importance of fire in creating open habitats in New Zealand. Unlike on other landmasses, however, there are far fewer potential candidates for restoring the browsing niche, and even the responsible use of anthropogenic fire is A) increasingly risky in an era of climate change and B) not really rewilding as it relies on constant human action and vigilance. Experiments with emus and ostriches have been proposed, but as I mentioned earlier, neither could really be classified as a browser. At least three ratite species: the emu (Dromaius novaehollandiae), greater rhea (Rhea americana), and common ostrich (Struthio camelus) can be farmed as livestock, and knowing this, it might be interesting to do controlled studies determining exactly what effects they do have on New Zealand vegetation, but I doubt they would be directly equivalent. Somali ostrich (Struthio molybdophanes) and lesser rhea (Rhea pennata) potentially incorporate more browse in their diets but would be far more difficult to obtain, while cassowaries (Casuarius sp) are almost purely frugivorous. Moas are thought to have engaged in occasional frugivory, but the main disperser of large-fruited taxa in New Zealand is the kereru or New Zealand wood pigeon (Hemiphaga novaeseelandiae). There do not appear to be any ratite-specialized, anachronistic fruits as there are in Madagascar or Eastern Australia. Most living ratites seem able to live in temperate conditions, although only the lesser rhea and emu do so naturally, but even the species that are considered to be occasional browsers are really just folivores, eating leaves but not twigs or bark. This seems also to have been the case for a few of the smaller moa species (ex. Emeus, Euryapteryx), but that simply means that some moa species had different trophic effects than others, and it doesn’t help us with the problem of woody overgrowth.

           While genetic technologies may eventually allow us to recreate the moa or some facsimile thereof, that will likely be a long way off. Only the one genome has so far been sequenced, from the little bush moa, ?Anomalopteryx didiformis, a small representative of the group which was nevertheless very widespread and ecologically important as an understory browser. In the meantime, the correct course of action is unclear. Taxon substitution, both purposeful and accidental, is proving to be an important part of rewilding and nature conservation in New Zealand, and the concept is regularly included in proposals for new sanctuaries and reserves. North Island Kokako (Callaeas wilsoni) have been sent to Secretary island as a substitute for their South island relatives (?Callaeas cinereus) and Snares Island snipe (Coenocorypha huegelli) have been sent to multiple islands previously occupied by the South Island snipe (?Coenocoryha iredelai). Similar proposals exist for Chatham snipe (Coenocorypha pusilla) to be sent to Little Barrier Island as a replacement for North Island snipe (?Coenocorypha barrierensis), and for rock wren (Xenicus gilviventris) to be acclimated to lowland habitats in order to fill the niche of the related bush wrens and stout-legged wrens (?Xenicus longipes, ?Pachyplicas sp). The practice has its limitations, however, and in this case, it is likely unable to restore the function of the moa. As a consequence, certain reserves may prove unable to support certain species after reaching a certain level of succession, unless open areas are artificially maintained. The alternative is that certain grassland species, including threatened taxa like takahe, will be increasingly relegated to refuge habitats representing only a fraction of their Holocene distribution, unless able to acclimate to agricultural habitats that are likely still inhabited by introduced predators.