Can Rubber Crop Systems Recover Termite Diversity in Previously Degraded Pastures in the Colombian Amazon Region?

Livestock production extension in Amazon has caused deforestation and soil degradation, with negative consequences on diversity and environmental services. Recently, rubber crops have been established in deteriorated soils of the Colombian Amazon as an option to restore hectares of unproductive degraded pastures. Bioindicator insects, such as termites, have been used to assess soil quality and fertility restoration. This study evaluated differences in termite abundance, species richness, and community composition in three different rubber crop systems as an indirect way of evaluating soil diversity restoring. Three rubber crop systems were sampled: clonal fields (rubber monocultures with different rubber clones), traditional commercial rubber plantations (rubber monocultures with just one rubber clone), and mixed plantations (rubber fields intercropped with copoazú fruit trees). Additionally, pastures in use for livestock production and natural forest relicts were compared to rubber crop systems, to serve as reference habitats. Termites were sampled using a 105-m transect method. Alpha diversity and beta diversity were estimated and compared between rubber crops and reference habitats. A total of 80 termite species belonging to two families were collected. Mixed plantations and pastures presented the lowest diversity rates. Species richness in rubber crop systems was 39% higher than that in pastures and included 72% of the termite species found in natural forests. Indicator species analysis associated soil-feeding termites with less diverse habitats and wood-feeding termites with high diverse habitats. Our results demonstrate that termite recovery will depend on the farming system selected and the agricultural practices implemented in the field, with some rubber crop systems, like commercial rubber plantations and clonal fields, recovering termite diversity better than others, such as mixed plantations.