Restoring Gorongosa’s Ecosystem Functions - Research Update

April 10, 2017

by Marta Correia, PhD student, Centre for Functional Ecology, University of Coimbra and Sérgio Timóteo, Post-doc Researcher, Centre for Functional Ecology, University of Coimbra

Legal protection and an ambitious restoration project have been contributing to the recovery of many of the most charismatic species to Gorongosa National Park. A major component of this was the construction in 2006 of a 62-km2 fenced wildlife sanctuary designed to protect relocated wildlife, predominantly buffalo and wildebeest, until the populations were large enough to be released into the wild. Within a few years, animal density and diversity inside the sanctuary increased rapidly. Some species’ populations more than doubled (e.g., wildebeest and waterbuck). The reintroduction of some large grazers, like buffalo, also gave smaller grazer populations the chance to recover naturally (e.g., oribi, impala, and bushbuck). However, an effective protection of natural ecosystems requires more than restoring species; it also requires the recovery of natural processes and functions that allow the long-term sustainability of ecosystems. 
Our research team, from University of Coimbra, Portugal, in collaboration with Gorongosa National Park and the Gorongosa Restoration Project, has been evaluating the recovery of Gorongosa’s vital signs in recent years. We aimed to determine whether recovery of the animal populations may effectively restore the seed-dispersal service provided by animals to plants. 
Field research
To explore this question, we applied network theory to understand how species interact with each other. Species do not exist by themselves; they are truly embedded in a web of connected species. A network approach allows us to look beyond what individual species are doing, but also take into account the interactions among different sets of species at the level of a whole community. By studying the pattern of those interactions, we can understand the overall functioning of a given ecosystem process, and understand the potential consequences of changes, or disturbances affecting a community. 
This has proved a valuable tool in studies of seed dispersal at the community level, and particularly in evaluating the effectiveness of ecological restoration. To identify seed dispersal interactions, we walked along transects sampling dung piles of all possible animals. We then screened these samples for undamaged seeds, which we later identified either against a reference collection of seeds collected in the field or by molecular techniques. 
Marta Correia and Sérgio Timóteo at the EO Wilson Lab
This study was recently published in Conservation Biology under the title “The refaunation and the reinstatement of the seed-dispersal function in Gorongosa National Park”. We showed that the recovery of animal populations is translating into an increase in the service they provide to the plants as active dispersers of their seeds to new locations where new plants can grow. This results from an increase in the complexity of the network of interactions between plants and their dispersers in the wildlife sanctuary (Figure 1). This work shows that by putting back the "pieces" that disappeared with the war (i.e. the animals), the "machine" that maintains the diversity of Gorongosa ecosystems can start afresh. The signs for the recovery and conservation of this system are therefore encouraging not only for life in Gorongosa but also for other places that have experienced similar loss of wildlife.
See the full text of the paper here
Figure 1. Quantitative seed dispersal networks inside the wildlife sanctuary (a) and outside the sanctuary (b). Both networks are based in the same sampling effort and represented on the same scale. The bottom boxes represent plant species dispersed, top boxes the dispersers (highlighted boxes represent the exclusive dispersers in the wildlife sanctuary), and the grey links represent the interactions. The width of each link is proportional to the frequency of occurrence of intact seeds on animal droppings. Dispersers are (common name): a. Impala, b. Duiker, c. Vervet monkey, d. Civet, f. Sable antelope, g. Porcupine, h. Waterbuck, r. Elephant, i. Honey badger, j. Oribi, k. Baboon, l. Warthog, m. Reedbuck, n. Eland, o. Nyala, p. Kudu and q. Bushbuck. Plant species are: 1. Acacia nilotica, 2. Amaranthus dubius, 3. Bobgunnia madagascariensis, 4. Bridelia mollis, 5. Cassia abreviatta, 6. Catunaregam sp., 7. Centaurea praecox, 8. Cucumis africanus 9. Cyperus sp., 10. Dicrostachys cinerea, 11. Diospyros mespiliformis, 12. Diospyros senensis, 13. Diospyros squarrosa, 14. Eriochloa meyeriana, 15. Grewia caffra, 16. Grewia inequilatera, 19. Grewia sp., 20. Hyphaene natalensis, 21. Indigofera sp., 22. Luffa cylindrica, 23. Mimusops obtusifolia, 24. Opuntia sp., 25. Panicum coloratum, 26. Pistia stratiotes, 27. Salicaceae, 28. Sclerocarya birrea, 29. Solanum incanum, 31. Sporobolus panicoides, 32. Tabernaemontana elegans, 33. Tamarindus indica, 34-48 unidentified seed species, 45. Xanthocercis zambesiaca and 46. Ziziphus mucronata