The Global Wheat Head Detection (GWHD) dataset was created in 2020 and has assembled 193,634 labelled wheat heads from 4700 RGB images acquired from various acquisition platforms and 7 countries/institutions. With an associated competition hosted in Kaggle, GWHD_2020 has successfully attracted attention from both the computer vision and agricultural science communities. From this first experience, a few avenues for improvements have been identified regarding data size, head diversity, and label reliability. To address these issues, the 2020 dataset has been reexamined, relabeled, and complemented by adding 1722 images from 5 additional countries, allowing for 81,553 additional wheat heads. We now release in 2021 a new version of the Global Wheat Head Detection dataset, which is bigger, more diverse, and less noisy than the GWHD_2020 version.

To accelerate genetic gains in breeding, physiological trait (PT) characterization of candidate parents can help make more strategic crosses, increasing the probability of accumulating favorable alleles compared to crossing relatively uncharacterized lines. In this study, crosses were designed to complement “source” with “sink” traits, where at least one parent was selected for favorable expression of biomass and/or radiation use efficiency—source—and the other for sink-related traits like harvest-index, kernel weight and grains per spike. Female parents were selected from among genetic resources—including landraces and products of wide-crossing (i.e. synthetic wheat)—that had been evaluated in Mexico at high yield potential or under heat stress, while elite lines were used as males. Progeny of crosses were advanced to the F₄ generation within Mexico, and F₄-derived F₅ and F₆ generations were yield tested to populate four international nurseries, targeted to high yield environments (2nd and 3rd WYCYT) for yield potential, and heat stressed environments (2nd and 4th SATYN) for climate resilience, respectively. Each nursery was grown as multi-location yield trials. Genetic gains were achieved in both temperate and hot environments, with most new PT-derived lines expressing superior yield and biomass compared to local checks at almost all international sites. Furthermore, the tendency across all four nurseries indicated either the superiority of the best new PT lines compared with the CIMMYT elite checks, or the superiority of all new PT lines as a group compared with all checks, and in some cases, both. Results support—in a realistic breeding context—the hypothesis that yield and radiation use efficiency can be increased by improving source:sink balance, and validate the feasibility of incorporating exotic germplasm into mainstream breeding efforts to accelerate genetic gains for yield potential and climate resilience.

The productivity of durum wheat [Triticum turgidum subsp. durum (Desf.) van Slageren] is affected by drought and/or high temperatures, challenges to be amplified by climate change. Pre‐breeding using wild relatives can supply useful traits for durum wheat improvement to adapt to major abiotic and biotic stresses. Sixty‐seven lines issued from backcrosses of Cham5 and Haurani durum wheat varieties with accessions of Triticum aegilopoides (Link) Bal. ex Koern., T. dicoccoides Koern. ex Schweinf., T. urartu Thumanian ex Gandilyan, and Aegilops speltoides Tausch were evaluated for drought and heat tolerance. The trials were conducted during two seasons (2016−2017 and 2017–2018) at Tessaout, Morocco, under full irrigation (optimal conditions) and rainfed conditions (drought stressed) and at Wed Medani, Sudan, under full irrigation combined with heat stress. The recurrent parents, along with eight best cultivars and elite breeding lines, were used as checks. Drought reduced the grain yield by 62%. Grain yield and drought tolerance index were used to identify lines to be used by breeding programs to enhance drought and heat tolerance. The derivatives lines 142014 (Cham5*3/T. aegilopoides), 142074 (Cham5*3/T. dicoccoides), and 142015 along with the checks Icarachaz and Gidara 2 ranked among the best under heat stress. Under drought stress, the lines 141972 (Haurani*2/T. urartu) and 141973 (Cham5*2/T. dicoccoides) yielded 196 and 142% of their recurrent parents’ yield, respectively. High variation was found for agronomic and phenology traits, with heading time explaining 16% of grain yield under drought, while thousand kernel weight accounted for 18% of the yield under heat. We conclude that gene introgression from wild relatives pays off and can increase wheat resilience to cope with climate change effects.