Labelling data (.json files) can be opened using labelme in python. R scripts (.r files) can be opened in R or as a text file. Python scripts (in script_7_for_Dataset_8_Fig_S29) can be opened in Python or as a text file.Funding provided by: Japan Science and Technology AgencyCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100002241Award Number: JPMJCR16O3Funding provided by: Japan Science and Technology AgencyCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100002241Award Number: JPMJCR15O1Funding provided by: Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen ForschungCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001711Award Number: 31003A_182318Funding provided by: Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen ForschungCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001711Award Number: CRSII5_183578Funding provided by: Japan Society for the Promotion of ScienceCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001691Award Number: 22H02316Funding provided by: Japan Society for the Promotion of ScienceCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001691Award Number: 17H06990Funding provided by: Japan Society for the Promotion of ScienceCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001691Award Number: 21H04977Funding provided by: Universität ZürichCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100006447Award Number: University Research Priority Program in Evolution in ActionFunding provided by: Japan Society for the Promotion of ScienceCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001691Award Number: 22H05179Funding provided by: Kyoto UniversityCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100005683Award Number: 2017jurc-cer05Funding provided by: Universität ZürichCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100006447Award Number: Global Affairs of the University of ZurichFunding provided by: Japan Society for the Promotion of ScienceCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001691Award Number: 22K21352Funding provided by: Kyoto UniversityCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100005683Award Number: 2018jurc-cer09Funding provided by: Kyoto UniversityCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100005683Award Number: 2019jurc-cer02Funding provided by: Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen ForschungCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001711Award Number: 310030_212551Funding provided by: Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen ForschungCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001711Award Number: 310030_212674

Long-term field monitoring of leaf pigment content is informative for understanding plant responses to environments distinct from regulated chambers but is impractical by conventional destructive measurements. We developed PlantServation, a method incorporating robust image-acquisition hardware and deep learning-based software that extracts leaf color by detecting plant individuals automatically. As a case study, we applied PlantServation to examine environmental and genotypic effects on the pigment anthocyanin content estimated from leaf color. We analyzed >4 million images of small individuals of four Arabidopsis species in the field, where the plant shape, color, and background vary over months. Past radiation, coldness, and precipitation significantly affected the anthocyanin content. The synthetic allopolyploid A. kamchatica recapitulated the fluctuations of natural polyploids by integrating diploid responses. The data support a long-standing hypothesis stating that allopolyploids can inherit and combine the traits of progenitors. PlantServation facilitates the study of plant responses to complex environments termed "in natura".

Top-view images of Arabidopsis were collected multiple times per day using commercial RGB cameras in a common garden in Japan for three seasons. Labelling data were generated by manually marking the areas of target plants in images. Scripts were produced for analyzing the images and color and size information of the target plants extracted from the images.