Suboptimal water and P availability are primary limitations to grain legume production. Root architecture influences water and P acquisition, but tradeoffs need to be better understood and mitigated. We hypothesized that tradeoffs in root class investment and resource acquisition strategy would be observable in a variety of grain legumes. Diversity panels of common bean (Phaseolus vulgaris L.), tepary bean (Phaseolus acutifolius A. Gray), cowpea [Vigna unguiculata (L.) Walp.], soybean [Glycine max (L.) Merr.], chickpea (Cicer arietinum L.), groundnut (Arachis hypogaea L.), and single accessions of other legumes were phenotyped in the field. We identified inverse relationships among investments in different root classes in most species, and between indicators of deep and shallow exploration in all species. Bean and tepary bean showed particularly strong tradeoffs in investment patterns among root classes, whereas chickpea and groundnut show less pronounced tradeoffs. We found that legume root architectural phenotypes can be placed on a root system architecture (RSA) spectrum, and that root phenotypes of epigeal and hypogeal taxa present distinct adaptive mechanisms. These life strategies integrating resource acquisition, use, and phenology are exemplified by contrasting chickpea, with many root axes, to tepary bean with few root axes and a contrasting water use strategy. We propose several RSA ideotypes and highlight how dimorphic root architecture may co‐optimize resource acquisition.

Selection criteria that co-optimize water use efficiency and yield are needed to promote plant productivity in increasingly challenging and variable drought scenarios, particularly dryland cereals in the semi-arid tropics. Optimizing water use efficiency and yield fundamentally involves transpiration dynamics, where restriction of maximum transpiration rate helps to avoid early crop failure, while maximizing grain filling. Transpiration restriction can be regulated by multiple mechanisms and involves cross-organ coordination. This coordination involves complex feedbacks and feedforwards over time scales ranging from minutes to weeks, and from spatial scales ranging from cell membrane to crop canopy. Aquaporins have direct effect but various compensation and coordination pathways involve phenology, relative root and shoot growth, shoot architecture, root length distribution profile, as well as other architectural and anatomical aspects of plant form and function. We propose gravimetric phenotyping as an integrative, cross-scale solution to understand the dynamic, interwoven, and context-dependent coordination of transpiration regulation. The most fruitful breeding strategy is likely to be that which maintains focus on the phene of interest, namely, daily and season level transpiration dynamics. This direct selection approach is more precise than yield-based selection but sufficiently integrative to capture attenuating and complementary factors.

Species-level crop and weed semantic segmentation in agricultural field images enables plant identification and enhanced precision weed management. However, the scarcity of labeled data poses significant challenges for model development. Here, we report a patch-level synthetic data generation pipeline that improves semantic segmentation performance in natural agriculture scenes by creating realistic training samples, achieved by pasting patches of segmented plants onto soil backgrounds. This pipeline effectively preserves foreground context and ensures diverse and accurate samples, thereby enhancing model generalization. The semantic segmentation performance of the baseline model was higher when trained solely on data synthesized by our proposed method compared to training solely on real data, with an approximate increase in the mean intersection over union (mIoU) by approximately 1.1% (from 0.626 to 0.633). Building on this, we created hybrid datasets by combining synthetic and real data and investigated the impact of synthetic data volume. By increasing the number of synthetic images in these hybrid datasets from 1× to 20×, we observed a substantially performance improvement, with mIoU increasing by 15% at 15×. However, the gains diminish beyond this point, with the optimal balance between accuracy and efficiency achieved at 10×. These findings highlight synthetic data as a scalable and effective augmentation strategy for addressing the challenges of limited labeled data in agriculture.

Summary: Pearl millet is a key cereal for food security in drylands but its yield is strongly impacted by drought. We investigated how root anatomical traits contribute to mitigating the effects of vegetative drought stress in pearl millet. We examined associations between root anatomical traits and agronomical performance in a pearl millet diversity panel under irrigated and vegetative drought stress treatments in field trials. The impact of associated anatomical traits on transpiration was assessed using subpanels grown in different soil within a greenhouse. In the field, total metaxylem area was positively correlated with grain weight and its maintenance under drought. In the greenhouse, genotypes with larger metaxylem area grown in sandy soil exhibited a consumerist water use strategy under irrigation, which shifted to a conservative strategy under drought. Water savings was mediated by transpiration restriction under high evaporative demand. This mechanism was dependent on soil hydraulics as it was not observed in peat soil with higher hydraulic conductivity upon soil drying. We propose that water savings under drought, mediated by large metaxylem area and its interaction with soil hydraulics, help mitigate vegetative drought stress. Our findings highlight the role of soil hydraulic properties in shaping plant hydraulics and drought tolerance.

Plant roots exhibit distinct architectural organization and overall shape. Current concepts to quantify architectural variation assume a homogeneous phenotype for a given genotype. However, this assumption neglects the observable variation in root architecture for two reasons: (i) sampling strategies are designed to capture architectural variation only for the most common phenotype, and (ii) traits are often measured locally within a root system and ignore the architectural organization. Here, we introduce a new concept: the phenotypic spectrum of crop roots to quantify architectural variation as the number of architecture types for one genotype in a specific environment. We use the shape descriptor DS-curve to characterize the whole root system architecture. Using DS curves as a core, we developed a computing pipeline that combines Kmeans++ clustering, outlier filtering and the Fréchet distance as a similarity metric to classify types of root architectures. Subsequently, we applied this pipeline to analyze a field dataset including three common bean (Phaseolus vulgaris) genotypes DOR364 (n=797), L88_57 (n=1772), and SEQ7 (n=768) under non-limiting and water-stressed conditions in 2015 and 2016. We found DOR364 showed five different root architecture types across environments, while L88_57 and SEQ7 showed four. The total variation within classified root architecture types of DOR364, L88_57, and SEQ reduced by 58.59%, 50.19% and 53.01%, compared to the variation of the complete data sets. DOR364 had stable fractions of root architecture types across environments. In contrast, L88_57 and SEQ7 showed more variation in their fractions. There was no significant biomass difference among root architecture types for all studied genotypes within each environment. As such, we hypothesize that the phenotypic spectrum might buffer the impact of environmental stresses as an acclimatization strategy by changing the composition of root architecture types at the population level.