Factors driving tree community structure in traditional home gardens in the Mayan forest

Introduction

'Peasant populations in tropical areas use Traditional Home Gardens (THG) as a key resource for their homes. THG supports significant plant biodiversity, most directly related to asset production, including locally rare species related to food, fuelwood, cultural, ornamental, shade, and wind barriers. Previous studies have considered factors driving tree diversity in THG; however, few have evaluated the biological, environmental, and management factors that shape tree diversity. Our objective in the present study was to identify the drivers of the tree community in THG across a precipitation gradient, evaluating biological, environmental and management factors.

Methods

We visited forty-eight THG distributed across the precipitation gradient spanning around 300 km The tropical Mayan Forest in the Yucatan Peninsula, Mexico., where tree inventory and identification were obtained from specimens with a diameter at breast height > 5 cm. Interviews were conducted with owners about the history of THG. Metadata were captured and analyzed for 19 putative determinants of tree population structure. Differences in tree diversity by individual THG and region were determined statistically. Modeled interaction networks were determined based on a co-occurrence matrix.

Results

THG tree abundances are mainly explained by the relative abundance of not deliberately established individuals, the organic carbon content in the soil, and surrounding forest integrity. Management history of THG by peasants, including factors such as irrigation, TGH area, and fertilization, showed smaller but significant differences. Tree distribution is at a lower degree influenced by the precipitation gradient in the Yucatan Peninsula, which shows different abundances even for species established by THG owners.

Conclusions

Our results suggest that biological and environmental factors may shape tree abundance and distribution over anthropic activities in THG. We identified rare, region-specific, and indicator species by region and modeled interaction networks that show coincidences with the empirical knowledge of THG owners.