Studying animal seed dispersal and forest regeneration: where cutting-edge modelling techniques meet intensive field-based research

Behind every large-scale ecological model is the indispensable fieldwork of numerous experts on the ground. Here, we dive into the collective research effort behind our latest paper – highlighting four datasets on frugivory, animal behaviour, gut passage time of seeds, and forest restoration.

Share this post

Choose a social network to share with, or copy the shortened URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

It’s 3.00 am, and it's still dark outside. A young ornithologist, dressed in a variety of earthy green shades, leaves her accommodation, heading to a study site. The first sun rays stream through the canopy when the researcher finds a tree full of fruits. She sits down, binoculars on hand, notebook on her lap, ready to note down all the birds that will come to eat the fruits during the next 3 hours. Neither the mosquitoes nor the heat nor the fatigue will be able to distract her attention. A few kilometres away, a PhD student is walking in complete silence and concentration. He hears a repetitive trill and quickly identifies that the bird must be less than 30 meters away. It’s an unusual song, but he understands that it’s a familiar bird species that makes a slightly different sound in that region. These might not be the first images that come to the mind of readers of the article 'Frugivores enhance potential carbon recovery in fragmented landscapes', recently published in Nature Climate Change (NCC). However, the foundation of this important scientific piece rests on the precise and thorough contributions of bird specialists and field ecologists.

In our work, we combine empirical datasets and modelling techniques to assess the potential contribution of diverse frugivorous birds to seed dispersal and carbon storage potential across landscapes within the Atlantic Forest of Brazil. We find that natural forest regeneration can be effective in landscapes with > 40% forest cover and < 133 meters between forest fragments. In these areas, diverse bird communities can generate seed rain that facilitates the recovery of late succession forests with maximum carbon sequestration potential. The free movement of seed-dispersing birds through landscapes can increase the carbon storage of regenerating tropical forests by up to 38%. In contrast, the movement of larger frugivores tends to be disproportionately limited in more fragmented landscapes, leading to a reduced dispersal of larger seeds.

But how did we arrive at these insights? In this post, we would like to take you behind the science and explain four ground-sourced datasets that fed our models, as well as acknowledge the collective effort employed to build them.

The datasets

The first data source is the ‘Atlantic frugivory dataset,’ published by the leading author of our NCC paper and her collaborators (Bello et al. 2017). This work compiles 8320 frugivory interactions reported for the Atlantic Forest, extracted from 166 published and unpublished sources from 1961 to 2016. The dataset encompasses interactions between 331 vertebrate species and 788 plant species. These impressive numbers reflect the hard work of countless professionals involved in the campaigns, including field technicians, students, volunteers, researchers, and PIs. Each interaction recorded in the dataset represents the diligent efforts of patient observers stationed beneath fruiting trees or the meticulous care given to germinating seedlings sourced from seeds found in animal droppings. Moreover, these interaction data are complemented by measurements of bird and seed dispersal characteristics, which have been gathered over the years through field collections or during visits to herbaria and museums. All these efforts sum up innumerable hours of fieldwork censusing animal feeding behaviour and laboratory work.

In our second dataset describing bird movement in the Atlantic Forest, we were able to quantify the fieldwork invested (Ramos et al. 2020): more than 25 people, among them bird specialists and volunteers, completed around 430 hours of observation of bird movements in 6 fragmented landscapes (72 hours per landscape). Each morning, starting at sunrise, the observers worked in pairs from vantage points, looking for birds for four and a half hours. Once a bird was spotted, they identified the species and tracked its movement, recording the time spent flying and perching (in seconds) with the help of binoculars and chronometers. The observations resulted in a set of short-distance movement tracks of 73 bird species. 

The third dataset represents the efforts of both students and professionals who spent countless hours observing and recording the time it took captive animals to digest and defecate fruits provided to them. Finally, the fourth dataset stems from extensive fieldwork to mark, measure, and collect trees of the native forests in this region. So, after thorough work in the herbaria, the identification of the species allows for an understanding of the composition, structure, and carbon stock potential of the forests in the area. All these datasets and the ground-based knowledge they created were essential to build meaningful models in our paper and to discuss the implications of our results properly.

Improving impact

Studies based on intense fieldwork often contrast with large-scale scientific and conservation projects. The pressure to increase impact indicators in scientific groups usually leads to a replacement of data sampling and experimentation with desk-based analysis of large and numerous datasets (see Ríos-Saldaña et al. 2018). Likewise, government and private entities are increasing the size of conservation and restoration endeavours. But to meet their ambitions, costs for implementing and monitoring projects need to be reduced, driving the automation of data sampling and homogenization of methods, and shorter time frames for projects. Global and large-scale initiatives have provided valuable guidance to tackle the biodiversity and climate change crises, particularly to address international problems requiring cross-border cooperation and the mobilization of large funds. Nevertheless, neglecting social-environmental specificities of local systems handicaps conservation decision-making and real-world application of scientific knowledge. Field-based research is essential to validate and refine global-scale products and can be more effective in guiding environmental policy. 

In our study 'Frugivores enhance potential carbon recovery in fragmented landscapes', we make the most of the well-trained eyes and ears of field experts working in south-eastern Brazil to provide a clear tipping point of long-term carbon storage potential in deforested landscapes. This work can have a big impact on restoration planning as we provide evidence for effective decision-making. Considering the available degraded area, long-term restoration of the Atlantic Forest can contribute between 0.1 Pg and 0.5 Pg to carbon sequestration (Barros et al. 2023). In fact, this region houses some of the biggest restoration initiatives in Brazil. However, these initiatives struggle to overcome the intense threats to biodiversity, and the costs of restoring the Atlantic Forest remain high due to the price of land, topography, the degree of soil degradation, and the presence of invasive grasses. Our results can support practitioners to prioritize areas for restoration and identify cost-effective methods. By quantifying the critical role of animals in facilitating forest regeneration, our research highlights how forest management practices can focus more explicitly on enhancing animal movement to achieve landscape-scale restoration and carbon targets.

 An ecologist monitors short-distance movements of birds in a fragmented landscape of the Brazilian Atlantic Forest.

An ecologist monitors short-distance movements of birds in a fragmented landscape of the Brazilian Atlantic Forest.


Barros, F. de V., et al., Cost-effective restoration for carbon sequestration across Brazil's biomes. Science of The Total Environment, 2023. 876: 162600.

Bello, C., et al., Atlantic frugivory: a plant–frugivore interaction data set for the Atlantic Forest. Ecology, 2017. 98(6): p. 1729-1729.

Ramos, D.L., et al., Forest and connectivity loss drive changes in movement behavior of bird species. Ecography, 2020. 43(8): p. 1203-1214.

Ríos-Saldaña, C.A., Delibes-Mateos, M., Ferreira, C.C., Are fieldwork studies being relegated to second place in conservation science? Global Ecology and Conservation, 2018. 14: e00389.

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Subscribe to the Topic

Life Sciences > Biological Sciences > Ecology
Ecosystem Services
Life Sciences > Biological Sciences > Ecology > Ecosystems > Ecosystem Services
Climate Change Ecology
Life Sciences > Biological Sciences > Ecology > Climate Change Ecology
Restoration Ecology
Physical Sciences > Earth and Environmental Sciences > Earth Sciences > Biogeosciences > Restoration Ecology
Life Sciences > Biological Sciences > Ecology > Biodiversity