Creating the data basis to adapt agricultural decision support tools to new environments, land management and climate change—A case study of the RiceAdvice App.

Increasing demand for land to ensure human food security in the future has already impelled agricultural production into marginal areas. The environmental conditions found there have a more pronounced impact on agricultural productivity than in the systems used so far under favourable conditions. In addition to this challenge, climate change is expected to increase the unreliability of weather conditions (through increased variability and occurrence of extremes) for farmers considerably. This unreliability is even more serious in developing countries' farming system where food security is vulnerable. Current efforts in digitalization offer great possibilities to improve farmers' decision‐making processes. A wide range of online tools and smartphone applications is available to support both agricultural extension services and smallholder farmers alike. These apps are often parameterized and validated to certain environments and are troubled when applied to new geographical locations and different environmental conditions. We have conducted field trials to demonstrate potential methods to close knowledge gaps in the data background for one of these apps, RiceAdvice, concerning three key aspects: shifting of cropping calendar, adjustment of fertilizer management and genotype selection. Sites in Ethiopia, Madagascar and Rwanda were selected to represent altitudinal gradients, with overlapping elevations reflecting differences in temperature to enable cross‐country comparisons. Planting dates were distributed throughout three calendar years, with continuous iterative planting dates taking place in Madagascar, in‐ and off‐season planting dates in Rwanda with different fertilizer applications, and one planting date during each rainy season in Ethiopia with different management options. With these trials, we have been able to identify key data sets needed for the adaptation of agricultural decision support tools to new environments. These include the assessment of climatic constraints on innovations to cropping calendars (e.g. double cropping), informed selection of alternative varieties able to complete crucial parts of their phenological development to avoid temperature‐related stress inducing, for example spikelet sterility in rice in late development stages and the effectivity of potential innovations in fertilizer management strategies. [ABSTRACT FROM AUTHOR]