B060-0012
Integrating earth observations and biodiversity data to predict nature-based across Costa Rica

Friday, 11 December 2020
Poster
Jeffrey R Smith1, Alejandra Echeverri2, Dylan MacArthur-Waltz3, Katherine Lauck4, Christopher Anderson1, Rafael Monge5, Irene Alvarado-Quesada6, Gretchen Daily7 and Rebecca Chaplin-Kramer1,8, (1)Stanford University, Stanford, CA, United States, (2)Natural Capital Project, Stanford, United States, (3)Stanford University, Stanford, United States, (4)University of California Davis, Davis, United States, (5)Ministry of Environment and Energy, San Jose, Costa Rica, (6)Banco Central, San Jose, Costa Rica, (7)Stanford University, Department of Biology, Stanford, CA, United States, (8)Natural Capital Project, Stanford University, Woods Institute for the Environment, Stanford, CA, United States
Abstract:
Nature and biodiversity provide immense benefits to humanity by improving our economies, physical and mental health, and overall well-being. These ecosystem services [ES] are vast in scope, and the relative importance of each service depends on a complex suite of biophysical and socioeconomic factors. In Costa Rica, for example, nature-based tourism is hugely important in supporting the national economy, driving up to 20% of annual GDP through direct and indirect avenues. To date, however, the underlying variables that influence nature-based tourism are poorly understood. In this study we investigate the roles of environmental conditions, human-built infrastructure, and biodiversity in driving tourism across Costa Rica.

In this project we aim to combine earth observations [EOs] of climate and land cover with biodiversity data to generate species distribution models [SDMs] for 682 vertebrate species. These SDMs, which are corrected for sampling bias across the country, show the relative occurrence of each species across the Costa Rican landscape. From these maps we generated 2 summary biodiversity layers, total species richness and bird species richness. We combine these biodiversity layers with environmental conditions and infrastructure to predict tourism across the country using 3 separate datasets: visitation rates to Protected Areas (provided by SINAC), locations of Flickr photo uploads (to investigate general tourism patterns), and locations of eBird checklist uploads (to investigate bird tourism specifically).

By using multi-model averaging and spatial regressions, we show that nature-based tourism is positively correlated with both biodiversity and human-made infrastructure. Importantly, our results show that none of these variables alone are sufficient to predict nature-based tourism. This is true both in protected areas (where biodiversity is the second most important variable in models - behind human footprint, Fig. 1) and across the entire country (where biodiversity is the fourth most important variable - behind human footprint, distance to roads, and distance to protected areas). This shows the immense importance of biodiversity to generating tourism revenues and that smart economic planning must incorporate consideration of key benefits provided by nature.


<< Previous Abstract | Next Abstract >>