Environmental analyses to inform transitions to sustainable diets in developing countries: case studies for Vietnam and Kenya

Abstract

Purpose: Sustainable diets are an environmental, economic, and public health imperative, but identifying clear intervention points is challenging. Decision-making will require descriptive analyses from a variety of perspectives, even under the inevitable uncertainty introduced by limited data. This study uses existing data to provide a diet-level perspective on environmental impact from food production in the case study countries of Vietnam and Kenya. Methods: FAO food supply data at decadal time steps were used as a proxy for national average diets in Vietnam and Kenya. We combined these data with estimates of the greenhouse gas emissions (GHGE) and water use impact associated with producing food commodities. Generic GHGE factors were derived from a survey of the life cycle assessment literature. Country- and commodity-specific blue water use estimates were used, reflecting country-of-origin for import-dominated commodities. The AWARE characterization model was used to offer a diet-associated water scarcity footprint. Trends in diet-associated environmental impacts were interpreted in light of diet shifts, economic development trends, and other factors. Results and discussion: Increasing per capita food supply in Vietnam, and in particular increases in meat, have led to rising diet-associated per capita GHGE. While supply of beef remains 5.2 times smaller than pork—the dominant meat—increases in beef demand in the past decade have resulted in it becoming second only to rice in contribution to diet GHGE. The water use and water scarcity footprint in Vietnam follow an increasing trend comparable to food supply. On the other hand, historically consistent levels of dairy and beef in Kenya dominate diet-level GHGE. Water use associated with the Kenyan diet shows marked increases between the 1990s and 2000s due to imports of wheat and rice from water-stressed regions. Environmental performance data for characteristic food production systems in these and other developing countries are needed to improve the representativeness and reliability of such assessments. Conclusions: Despite data limitations, the methods and results presented here may offer a fresh perspective in sustainable development policy deliberations, as they offer an entry point to linking environmental impact and consumption behaviors and can elucidate otherwise obscure or unexpected outcomes. A clear need emerges for further environmental analysis of dominant production systems within both Vietnam and Kenya.