%0 Journal Article
%9 ACL : Articles dans des revues avec comité de lecture répertoriées par l'AERES
%A Hajjarpoor, A.
%A Pavlík, J.
%A Hora, J.
%A Konopásek, J.
%A Pusupuleti, J.
%A Vadez, Vincent
%A Soltani, A.
%A Feike, T.
%A Stoces, M.
%A Jarolimek, J.
%A Kholová, J.
%T In-silico optimization of peanut production in India through envirotyping and ideotyping
%D 2025
%L fdi:010093479
%G ENG
%J Computers and Electronics in Agriculture
%@ 0168-1699
%K Crop Modeling ; High Performance Computing (HPC) ; Genotype x Environment ; x Management ; (GxExM) interaction ; Index of Goodness (IoG) ; Environmental Clustering (EC)
%K INDE
%M ISI:001480962000001
%P 110383 [16 ]
%R 10.1016/j.compag.2025.110383
%U https://www.documentation.ird.fr/hor/fdi:010093479
%> https://horizon.documentation.ird.fr/exl-doc/pleins_textes/2025-06/010093479.pdf
%V 235
%W Horizon (IRD)
%X Peanut (Arachis hypogaea L.) is an important cash crop with significant yield gaps, especially in developing countries. Optimizing peanut production could foster economic growth for a significant number of smallholder farmers across the globe. In this study, we used an in-silico cropping system model to simulate and optimize genotype x crop management (G x M) across India that would narrow the existing peanut yield gaps. For that, we simulated diverse G x M combinations across range of environments (E) in India, considering three irrigation regimes typical for managing peanut production systems. Covering whole India in a 0.5 degrees x0.5 degrees resolution, we simulated 60,480 G x M combinations for each grid, summing up to a total of 2.3 billion simulations and 1.02 TB output data. This required well-structured high-performance computing (HPC) approaches, data management, and analytical capacities. For this, we present the concept of a re-usable HPC system with interoperable modules, which can be readily adapted for different simulation setups. We introduced the novel way of analyzing simulation outputs - "Index of Goodness" (IoG) - that aggregates key peanut production characteristics (grain and haulm production) and production risk failure. IoG is a simple way to evaluate the suitability of simulated GxM options from the perspective of end-users, including primary producers and crop improvement programs. The generated output was used to identify the geographic regions (environmental clusters, EC) with high degree of similarities within each of the tested irrigation regimes. For each cluster, we identified a specific suite of GxM to benefit peanut production and prioritize G targets for breeding. In principle, irrigated cropping systems would benefit from high planting densities, long duration and vigorous crop types. With diminishing water availability (particularly in the Thar Desert and SE India), the optimal production included shorter duration crop types which could quickly respond to drought stimuli (i.e. close stomata and conserve soil water upon soil and atmospheric drought exposure). These traits should also be considered in phenotyping strategies to support context-specific breeding.
%$ 076 ; 020 ; 122