Simulating common bean performance under selected environmental stresses for improving index-based insurance model and its uptake among farmers in three districts of Rwanda

Author: Ntukamazina, Nepomuscene

Supervisors: Richard N.; John Mburu and Rolf Sommer

Index based crop insurance products such as Area Yield Index Insurance (AYII) are widely promoted as a means of addressing climate related constraints for bean crop production. However, in Rwanda, farmers are reluctant to subscribe to the full season cover contract of the AYII product arguing that the product is expensive. This research thus aimed at enhancing AYII to better respond to environmental stresses at different growth and developmental stages of common bean. Greenhouse and field experiment were conducted for two growing periods each during the period from September 2015 to February 2017. In the greenhouse experiment (Sep 2015 – Feb 2016 and Mar 2016 – Jul 2016), the response of bush and climbing bean to excessive and minimal soil moisture at five plant growth stages (emergence, vegetative, flowering, pod setting and seed filling) was investigated. Two bean genotypes (RWR2245 for bush type and MAC44 for climbing type) were used in a Completely Randomized Design with four replicates. For the field experiment (Mar 2016 – Jul 2016 and Sep 2016 – Feb 2017), four bean genotypes (Akararakagenda & RWR2245 for bush type and MAC44 & RWV1129 for climbing type) were used to assess the effect of natural bean disease pressure on bean yield losses in low, medium and high altitude of Rwanda. The field experiment was laid out in a split-split-plot design where the bean genotypes were assigned to the whole plots, plant growth stages with four levels (vegetative, flowering, pod setting and seed filling) to sub-plots and pesticide application with two levels (application and no application) to the sub-sub-plots. Data collected on MAC44 and RWR2245 from both greenhouse and field experiments were modelled with GROPGRO-Dry bean model of DSSAT for simulating bean yield losses due to drought, waterlogging and natural bean disease pressure at the various plant growth stages of common bean. For each treatment, both simulated grain yield and yield reduction rate were fitted in the area yield index insurance model to estimate the subsequent expected premium rate. Seed filling stage was severely affected by waterlogging stress with a yield reduction of 28%. Drought stress significantly affected bean production during seed filling stage with an estimated yield reduction of 23%. Pod setting stage was the most sensitive to natural bean disease pressure with an estimated yield loss of 30%. The corresponding expected premium rates were estimated at 429 kg ha-1 for waterlogging stress at seed filling stage, 257 kg ha-1 for drought stress at seed filling and 467 kg ha-1 for natural bean disease pressure at pod setting. As the AYII product does not covering, in separate contracts, weather stresses that could happen during different plant growth stages, the product was considered inadequate to the needs of resource limited farmers. This study has suggested an anticipated claim formula that insurers can use for diversifying the area yield index insurance product into sub-products from which farmers can select insurance coverage based on their experiences in bean production and their income level (financial means). The formula predicts both actual area yield and corresponding premium rates for drought, waterlogging and natural disease pressure at vegetative, flowering, pod setting and seed filling growth and developmental stages of common bean. Keywords: bean disease, drought, index-based insurance, Phaseolus vulgaris, premium rate, yield loss and waterlogging.