Back

Characterization of the pearl millet cultivation environments in India: status and perspectives enabled by expanded data analytics and digital tools

Garin, V.; Choudhary, S.; Murugesan, T.; Kaliamoorthy, S.; Diancumba, M.; Hajjarpoor, A.; Satyavathi, T.; Gupta, S.; Kholova, J.

2023-02-21 plant biology
10.1101/2023.02.18.529051 bioRxiv
Show abstract

The cultivation of pearl millet in India is experiencing important transformations due to changes in weather, socio-economic trends, and technological progress. In this scope, we propose a new characterization of the pearl millet production environment in India using the latest available data and methodology. For that, we constructed a database incorporating data on various aspects of pearl millet cultivation at the district level from 1998 to 2017. We complemented this analysis using extensive pearl millet agri-system simulations to evaluate crop models abilities to reconstruct and analyse the system at an unprecedented scale. We also proposed a new method to infer system parameters from crop model data. Our results show important differences compared to the characterization currently used. The East part of the pearl millet tract (East Rajasthan, Haryana, Uttar Pradesh, and Madhya Pradesh) emerges as the only region where pearl millet cultivation has grown with potential surplus that is likely exported. Important reductions of pearl millet cultivated area in Gujarat, Maharashtra and Karnataka are potentially due to economy-driven transition to other more pro table crops like cotton, maize, or castor bean. The data used also point toward a constant increase of the rain during the growing season which could have major consequences on the future of this crop, with potential positive effects like extra yield but also negative like extra pressure due to more intense and erratic rainfall or transition to more pro table crops requiring more water. Despite difficulties to predict pearl millet yield in rapidly changing environments, the tested crop models reflected reasonably well the pearl millet production system, thus, setting the base for effective system design in future climatic scenarios. Our data and results have been gathered in an open-source interactive online application.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.