Ecosystems, health, and big data

Background/Question/Methods

Ecosystems worldwide are undergoing rapid change, and the effects of these changes on both biological communities and ecosystem services are increasingly well known.  Ecosystems also support human health by regulating disease vectors, improving water quality, contributing to mental well-being, etc.  While these ecosystem-health relationships are intuitively appealing and often claimed, they remain poorly understood.  Field studies and analytical models have tested some of these relationships and illuminated their dynamics.  Increasingly available global data, however, along with modeling techniques to analyze them, are beginning to allow a more synthetic approach to the linkages among nature, human health, and the consequences of global change.  As part of a SESYNC working group, we compiled a global database of household surveys, covering millions of individuals in more than 50 countries over 30 years.  The Demographic and Health Surveys contain geo-referenced information on economic (e.g., wealth, assets), demographic (e.g., education, age, gender), and health (e.g., diet, anemia, malaria, diarrhea, stunting) variables.  We added information on land cover, protected areas, climate, watershed conditions, and other environmental variables.  Using this database and model selection and averaging techniques, our working group is asking the general question: What is the relationship among ecosystems, conservation, and human health?

Results/Conclusions

We investigated the effects of watershed condition on waterborne disease among children downstream.  We found that the incidence of diarrhea increases with the amount of human-dominated land, and decreases with coverage of protected areas, upstream.  Attributes of water supply and sanitation were not important in our models as simple effects, but did show some interacting effects with watershed condition.  We also tested the effects of Marine Protected Areas (MPAs) on children’s health in coastal communities.  Preliminary results suggest that childhood stunting z-scores decrease (i.e., more stunting) as distance to MPAs increases.  Additional analyses on the multi-dimensional health impacts of both terrestrial and marine protected areas are underway, and the database is a unique resource that can help address many additional questions.  Syntheses of global observations and relationships can complement and broaden the insights gained through local field studies and analytical models.  Through all of these approaches, we can better understand whether, and under what conditions, nature conservation can be viewed as an investment in public health.