Dataset associated with publication "Are agricultural researchers working on the right crops to enable food and nutrition security under future climates?"

A simple crop modelling approach was applied to derive expected future changes in regional climate suitability for crops. To determine if different starch-rich and pulse crops are currently underresearched or overresearched, a global relation between crop-specific research output (number of publications) and the total nutrient output available for human consumption was calculated. The study tallied the number of scientific publications produced by crop and world region and compared this with their production statistics. Benchmarks were developed based upon the number of publications on a given crop against the expected number of publications based on the relative share of nutrient output from that crop, benchmarks were made regionally and globally. To set benchmarks, a zero-inflated Poisson generalised linear model with publications as the response variable and ln(total nutritional values +1) and region (a categorical variable which modifies the intercept) as the explanatory variables, while we predicted zero-inflation using the energy score. The regression line from this model was taken as the benchmark. To set global benchmarks, the model was applied but without the region variable in the linear regression. For each crop-region combination the difference between the actual number of publications and the regional and global benchmarks were calculated. We then summed the research intensity for each of the under researched regions.