Effects of Climate Variability on Queen Production and Pollen Preferences of Neotropical Bumblebee Bombus atratus in a High Andean Suburban Condition.

Bombus atratus Franklin is a widely distributed bumblebee of South America. In Colombia, this species is recognized for its ability to adapt to highly disturbed habitats. However, knowledge of its ecology is poorly known, in particular conditions to ensure the long-term conservation of its populations. Identification of pollen resources is an important issue that could be used as a tool to manage and conserve bumblebees. In tropical areas, rainfall patterns could affect floral phenology and therefore the availability of pollen resources. Considering this, the present work aimed to establish the effect of extreme weather conditions (El Niño) in pollen availability, use of pollinic sources, and gyne production in B. atratus colonies. We reared and located 14 B. atratus colonies in a suburban area during a dry season (ENSO "El Niño") and a rainy season (ENSO "La Niña"). We registered time to gyne production and numbers of gynes produced per colony. We extracted pollen samples to establish both its floral origin and its relative abundance. We measured floral offer for each season. The data of pollen use per colony were utilized to perform Bipartite networks. We analyzed the production of gynes and pollen use per season with correlation models and generalized linear models. Colonies of the rainy season produced more gynes and faster. The floral diversity and offer were higher during the rainy season. Successful colonies used specific pollen sources in two seasons, independently of the floral offer. Extreme dry season affected development of B. atratus colonies.

Diversity and Phenology of Wild Bees in a Highly Disturbed Tropical Dry Forest "Desierto de la Tatacoa", Huila-Colombia.

Colombian tropical dry forest is considered the most endangered tropical biome due to anthropic activities. Desierto de la Tatacoa (DsT) is an example of high disturbed tropical dry forest which still maintains a high biodiversity. The objective of the study was to record the diversity and phenology of wild bees in this place by monthly sampling between December 2014 and December 2016 in a 9-km2 area. During the study, there was a prolonged El Niño-Southern Oscillation period. Bees were collected by entomological nets, malaise traps, eugenol scent trapping, and nest traps. Shannon index was calculated to estimate diversity and Simpson index to determine dominance of a species. The effect of environmental conditions (wet and dry season) in richness and abundance was analyzed by paired T tests. A total of 3004 bee specimens were collected, belonging to 80 species from Apidae, Megachilidae, Halictidae, and Colletidae. Apidae was the most diverse. Shannon index value was 2.973 (discarding Apis mellifera Linnaeus 1758 data); thus, DsT can be considered as a zone of high wild bee diversity. Dry and rainy season showed differences in diversity (p < 0.05). Rainy season showed larger blooming periods and higher bee diversity than dry season. In both seasons, social species were dominant (e.g., A. mellifera or Trigona fulviventris Guérin 1844). Although DsT is a highly disturbed ecosystem, this study found it has the second highest number of genera and the fourth highest number of species reported in Colombia.

A Simulation Model of the Mass Rearing of Tetranychus urticae Koch (Acari: Tetranychidae) on Beans.

The supply of predatory mites as natural enemies is a key component to guarantee the success of biological pest control programs as alternatives to chemical control in commercial crops. To meet the demand for a supply of biologicals, the mass rearing of natural enemies is an option, and the first step must be to develop a standardized system that maximizes the production of prey. One choice for this first step is to use simulation models that can evaluate scenarios that are difficult or complex to address experimentally. In this work, a model was developed to evaluate the current management conditions for the mass rearing of the pest mite Tetranychus urticae Koch. Our aim was to identify alternative scenarios to maximize mite production through mass rearing that could be evaluated in real systems. We assumed that populations of T. urticae were regulated by the conditions of supply-demand theory and modeled the age structure, temperature effects, and individual phenology of T. urticae. The supply-demand theory of resources was used to regulate populations, which involved structured ages and temperature effects for the different stages in the development of individuals. We used the functional response and the paradigm of metabolic pool models to describe resource acquisition and allocation. We demonstrated that 7- to 14-day-old plants infested with 45 or 62 T. urticae/plant could reach 25,000 individuals/plant, being 50% of these preys at the preferred stages by the predator Phytoseiulus persimilis Athias-Henriot. Our theoretical model requires validation in experimental/real systems of mass rearing to better verify the validity of all of the parameters and predictions before commercial implementation.