For millions of years, a diverse menagerie of large herbivores, or megafauna, influenced terrestrial ecosystems. In Australia, these included giant wombats and hoofed 3 m tall kangaroos, in North America there were ground sloths and mammoths, and in Europe some 130,000 years ago hippos would have lined Danube, Volga, Rhine and Thames while elephants and giant deer roamed around. These animals experienced a wave of extinctions corresponding to the worldwide expansion of human beings, leading to dramatic but still mysterious changes in ecosystems. 

Also the survivors of these so-called Late Pleistocene extinctions are still in trouble. For example 60% of the remaining 75 large mammalian herbivores are threatened with extinction and so are even 100% of the 12 remaining herbivores > 1000 kg. These large animals have unique impacts on ecosystems due to their massive body sizes. Elephants, for example, are reportedly able to knock down more than 1000 trees a year. That's something smaller herbivores like springboks or roe deer would have a hard time even trying.

However, even though there’s quite a large number of case studies and literature reviews about the effects of megafauna on ecosystems - a quantitative test for the generality of those effects has been lacking. We aimed to change that with our new study in Nature Ecology & Evolution. We conducted a systematic literature search to collect all suitable studies about the effects of wild mammalian herbivores ≥ 45 kg that we could find. Subsequently, we extracted the data from the studies and re-analyzed them to investigate the effects of megafauna on the diversity and abundance of different taxa (plants, birds, small mammals, invertebrates), vegetation structure and nutrient concentrations, soil nutrients and properties and biogeochemical processes. Overall, we included 5990 data points (each representing for example a comparison between an exclosure plot and a control) from 297 studies across six continents (although biased towards Europe, North America, South Africa, and Australia).

Despite a striking variability in the effects of megafauna on ecosystems, we found that megafauna can drive shifts in soil and plant nutrient concentrations and can promote open and semi-open habitats by reducing vegetation biomass and cover. We also found that megafauna tend to reduce the abundance of small mammals. Notably, and contrary to frequent claims, we did not find a statistically significant overall effect of megafauna on soil carbon or local plant diversity. 

The megafauna in our dataset spanned from 45 kg (e.g., the size of a small impala) to elephants (about 4500 kg), which allowed us to test how body mass shapes megafauna impacts on ecosystems We found that larger-bodied megafauna communities had more positive effects on available soil nitrogen while smaller-bodied megafauna communities tended to  have negative effects on plant diversity. Indeed, a thousand springbok cannot do what an elephant does. 

Finally, we tested one of the key hypothesized impacts of megafauna: the promotion of ecosystem heterogeneity. This is something we could confirm by showing for example an increase in the variability in vegetation structure in presence of megafauna. That’s a particularly interesting finding given that environmental heterogeneity is a universal driver of species richness across spatial scales, biomes and species groups. Environmental heterogeneity may for example prevent one or a few species from dominating a system as well as help species survive climatic stress by providing local refugia, and via both effects likely supporting the resilience of ecosystems to climate change pressures.

Our findings imply that ecosystems which lost their large animals during the late Quaternary extinctions (which is the vast majority of ecosystems) are likely missing key processes. This also suggests that most current conservation efforts are targeted towards such incomplete systems which may have severe implications for their success. An increasingly popular strategy to overcome this problem is trophic rewilding which aims to restore lost functionality via the (re-)introduction or reestablishment of large animals. While there are concerns about potentially unwanted effects due to the introduction of non-native animals as functional substitutes for extinct species, recent work - using the same dataset as this study - suggests that the effects of megafauna - at least on plants - are independent of their nativeness but rather explained by their functional traits (see here for the study: https://www.science.org/doi/10.1126/science.adh2616). Together, this advocates for the consequent protection of the few remaining megafauna species and the carefully planned and implemented reestablishment of megafauna populations where their functionality is lost in order to achieve the most positive outcome for the Earth's ecosystems under the increasingly unprecedented global environmental conditions.