High-throughput data collection techniques and large-scale computing are transforming our understanding of ecosystems, making convergent scientific frameworks a research priority [1]. As human activities increasingly impact ecosystem processes, we need new approaches that focus on how whole communities of organisms interact with people and the physical environment at the scale of landscapes or catchments [2]. This requires an e-infrastructure for data intensive science that enables the integration of computational physics, chemistry, biology, ecology, economics and other social sciences. Such an advance would allow researchers to (1) characterize the multidisciplinary functional attributes of social-ecological systems; (2) quantify the relationships between those functional attributes under historic and current conditions; and (3) model the trajectories of goods and services under a range of policy-driven scenarios and future environmental conditions. The resulting knowledge would improve our ability to predict human and natural change at scales relevant to management/conservation actions.

Unlike some aspects of climate change, processes related to biodiversity and ecosystem services are “typically place-based and many of the effects are seen at sub-global scales” [3]. Inspired by successes in modeling complex systems at other scales of organization, notably the cell [4], the Island Digital Ecosystem Avatars (IDEA) Consortium aims to build computer simulations (‘avatars’) to the scale of whole social-ecological systems. With a common boundary constraining their physical, ecological, and social networks, islands have long been recognized as model systems for ecology and evolution [5]. Their geography sets clear limits on the species to inventory, space holders (ground cover) to measure, organisms to count, physical–chemical contexts to characterize, and natural–human interactions to consider. The knowledge and cyberinfrastructure developed for island avatars, and complementary efforts targeting major cities like Singapore [6] and New York [7], will eventually scale to countries and regions, including their associated coastal waters. Island avatars are built from the genome up, while at the same time downscaling regional models to establish boundary conditions. They address many of the challenges faced by macrosystems ecology [8], and general ecosystem models [9]. Indeed, an Earth Avatar would converge on the Global Earth Observation System of Systems (GEOSS) [10] and Future Earth [11]. The IDEA approach, however, avoids overwhelming complexity, instead concentrating effort on the simplest social–ecological systems that include most of the data types covered by GEOSS, and many of the processes found globally.

Moorea IDEA

The small island of Moorea (134 km²) in French Polynesia is well placed for a proof-of-concept study. About 15 km northwest of Tahiti with a population of ~17,000, Moorea is perhaps the best studied island in the world [12] thanks to several decades of activity at its two research stations (CNRS-EPHE CRIOBE [13], and the University of California (UC) Berkeley Gump Station [14]), which respectively house France’s Center of Excellence for Coral Reef Research (LabEx CORAIL [15]), and the US National Science Foundation’s only coral reef Long-Term Ecological Research (MCR LTER) site [16], which is administered by UC Santa Barbara. Additionally, the Moorea Biocode Project [17] has characterized every species (>1 mm) on the island, including genetic sequences, museum specimens, and digital photographs [18]. While there is still much more to learn, especially concerning the vastly diverse microbes [19] and human systems, the existing physical and biotic databases provide a powerful foundation for whole system ecological modeling. Combined with a wealth of data on the resilience of Moorea’s ecosystems, including the response of its coral reefs to large-scale perturbations [20] and the evolution of Polynesian society [21], Moorea has many of the characteristics needed to advance systems ecology and sustainability science [22].

The Moorea IDEA aims to understand how biodiversity, ecosystem services, and society will co-evolve over the next several decades depending upon what actions are taken. Specifically, we ask: (1) what is the physical, biological, and social state of the island system today? (2) How did it get to this point? (3) What is its future under alternative scenarios of environmental change and human activity, including conservation efforts? These questions are addressed through a place-based data science infrastructure and computational platform (Fig. 1). The Moorea Avatar is the best digital representation of the island; a three-dimensional visualization of Moorea that looks similar to the one on Google Earth, but which includes the dimension of time and enables researchers to zoom into a location, access data, and run simulations. Today’s island represents the key baseline because the majority of the modeling data needed are not available for historic time periods. The Avatar computational platform allows other versions of Moorea to be generated and visualized in silico for a range of purposes. This involves the integration of physical, biological, and social data [12–21], and drawing on best-available scientific knowledge to show what the island looked like in the past (using historic baselines to explore particular issues), and to predict how it might look in the future. Unlike video games, our projections are constrained by reality and are intended not only for research and education, but also to support scenario-based planning; helping local communities adapt to environmental change and maximize ecological resilience. Our goal is to emulate P4 Medicine, extending to social–ecological–physical systems the Predictive, Preventive, Personalized, and Participatory approach that promises to revolutionize the biomedical field [23].

Working with the local population to co-develop island avatars as “boundary objects” [24] is critical to ensuring that the simulations are useful, credible and legitimate. Non-scientist stakeholders take joint ownership of their avatar by prioritizing the policies to simulate (e.g., conservation plans), and by contributing traditional/local knowledge and data. The participatory approach includes citizen science, taking advantage of newly affordable technologies and helping reconnect people with natural processes through observation and experiential learning.

The new Intergovernmental Platform on Biodiversity and Ecosystem Services (IPBES) has prioritized an assessment on the “Modeling of Biodiversity and Ecosystem Services” [25]. Addressing this grand challenge will require computational models of place that are able to simulate alternative scenarios and visualize likely outcomes for scientists, policymakers, and the public [26]. Big data, computational ecology, and sophisticated simulation platforms cannot solve all of the world’s problems, but harnessing scalable technology addresses the lack of capacity in local knowledge management systems, and can help illuminate pathways to sustainability. In an era in which society is seeking to transition to clean energy and sustainable economic growth, knowledge of the pathways to these futures is needed, along with showcases demonstrating that such change is possible. At least initially, examples are more likely to come from islands and cities than large regions and countries. Islands are disproportionately affected by global change and epitomize the coastal zones where most of humanity lives. They serve as models for continental regions and, ultimately, for our common island home: planet Earth.