Open Collections

Abstract

Forests support food security and nutrition worldwide, especially so for rural households who collect a variety of food products from nearby forests. While the importance of forest cover to local diets has been well-researched, little is known regarding the role of more specific forest characteristics. Further insight on the role of specific forest characteristics would be valuable for better identifying the types of landscapes that support a nutritious and diverse diet, in the short and long-term. To address this research gap, I linked household dietary data to remotely-sensed geospatial indicators of surrounding forest cover characteristics – using more nuance than is typically undertaken – by examining forest age, tree density, and forest fragmentation in Kenya’s East African Montane Forests. First, I analyzed Demographic and Health Survey results from 338 rural households to identify dietary trends of the oldest child in each household (within the preceding 24 hours), as well as all for the entirety of the household (over the preceding 7 days). Using contemporary landcover data, I then calculated the cover of three different forest tree density classes (open, moderate, dense), as well as mean forest fragmentation within a 10 km radius of all households. Finally, using a sequence of landcover data spanning 24 years, I determined the age of each forest. Interestingly, diet diversity of children and households demonstrated no or relatively weak associations with forest characteristics. However, by parsing out individual food groups, I exposed the nuance and complexities associated with the forest-diet relationship. Open forests were positively associated with vegetable/fruit consumption, but negatively associated with meat consumption. Younger forests were positively associated with consumption of meat and vitamin A-rich fruit, but negatively associated with consumption of dark green leafy vegetables. While forest fragmentation was negatively associated with vegetable/fruit consumption, it was positively associated with meat consumption. My findings provide clear evidence that there is no single ‘ideal’ type of forest for supporting food security and nutrition – rather, different types of forests are associated with different dietary benefits. Taken together, these results indicate the need for more in-depth research that accounts for factors beyond the amount of generic forest cover.