Although legumes are essential to meet the nutritional needs of Afghanistan’s growing population, the productivity of these vital crops remain low under farmer conditions. In response, ICARDA and its national partners are introducing improved varieties of chickpea (Cicer arietinum) and mung bean (Vigna radiata), and promoting optimal agronomic practices through demonstration trials in three provinces. The study uses a Bayesian approach (BA) for analyzing future on-farm trials.
To reverse the low productivity of locally grown legumes, and enhance the adoption of improved varieties, ICARDA and Afghanistan’s Ministry of Agriculture, Irrigation and Livestock initiated demonstration trials in 2009 with improved varieties of chickpea and mung bean.
Statistical analysis of data from the on-farm trials is normally based on the estimation and comparison of means using two-sample t-test, paired t-test or analysis of variance. However, an approach based on stochastic dominance of the improved varieties appears more appropriate in the context of on-farm trials when the farmer has to make a decision on preference of a crop production package/technology.
In the given project, BA appeared appropriate to weigh in the information from past trials. Given that past data from on-farm trials was gathered in a broadly similar biophysical environment, the objective of this study was to present a BA for analysis of on-farm trials data and apply the method and computing codes for an analysis of chickpea and mung bean trials in Afghanistan in 2012, with prior information extracted from 2009 to 2011 data. Mean productivity of the improved varieties has been estimated, as well as the probability that the variety under evaluation exceeds the set target level.
The study used datasets generated under ICARDA-Afghanistan project from farmer participatory demonstration trials in eight districts of Baghlan, Mazar and Uruzgan provinces during 2009-2012. Farmers’ participatory demonstrations were laid out in order to popularize improved varieties of chickpea (Sehat and Madad) and mung bean (Mai-2008 and Mash 2008) along with their associated best practices.
Each demonstration was laid out in an area of 1,000 sq. m. Besides, best agronomic practices such as use of seed rate (100 and 50 kg per ha for chickpea and mungbean, respectively), optimum fertilizer (50 kg urea and 100 kg diammonium phosphate (DAP) for chickpea and 50 kg urea and 120 kg DAP for mungbean per ha) and applying weed control methods were included in the demonstrations.
The yields obtained in the demonstrations were compared with the yields obtained by farmers growing local varieties with local agronomic practices. The datasets for 2012 were evaluated to compare the packages for productivity and risk, while the data from 2009 to 2011 were used for prior information on the variance parameters in the Bayesian analysis.
Author: Murari Singh, Senior Biometrician (ICARDA)