Brazil is an agricultural country, with 190 Mha of pastures sustaining 209 million cattle. Fewer than 10% of the cattle are fattened in feedlots, whereas cattle reared on pastures have a competitive advantage for export, eliminating the risks presented by the mad cow disease (bovine spongiform encephalopathy) and considerations related to animal welfare. Brazil has been the world’s largest exporter of beef since 2004 and has the largest commercial herd in the world. In 2011, 16.5% of its production was exported, and the livestock sector contributed 30.4% of the gross national product from agribusiness and 6.73% of the total GNP. Many forage breeding programs, mainly at Embrapa, have contributed to the development of improved pastures, and cultivars of Brachiaria brizantha, B. decumbens, B. humidicola and Panicum maximum are the main pastures used in the country. All have apomictic reproduction, which means there are few cultivars occupying very large, continuous areas, thus suggesting a risk to the productive system. Such is the case of B. brizantha cv. Marandu, which occupies around 50 Mha. The Brazilian tropical forage seed industry is also important, and Brazil is the main seed exporter, supplying all Latin American countries. Due to pasture degradation, around 8 Mha is renovated or recovered each year. Forages are also used and planted each year in integrated crop–livestock and integrated crop–livestock–forest systems. Nowadays, these systems occupy 4 Mha. Improved pastures are thus a major asset in Brazil not only for the beef production chain but also for the dairy industry.

The main benefits expected from genomic selection in forage grasses and legumes are to increase selection accuracy, reduce cycle time, and potentially reduce evaluation costs per genotype. Aiming at designing a training population and first generation of selection, deterministic equations were used to compare the gain and accuracy of six genomic selection methods implemented within the context of traditional experimental plot designs used in forage breeding. Combined use of both phenotypic and genotypic information was superior to other methods using low density markers (approximately three per cM) and for heritability lower than 0.6 but resulted in lower selection gain per year in relation to individual genomic selection using higher marker density. Initial accuracies were increased by a selection index method proposed as a procedure to improve long‐term rates of gain for advanced generations using genomic selection. Application of genomic selection methods to forage breeding is expected to be of greatest value under the following circumstances: (i) when phenotypic evaluation of individual plants is incapable of predicting performance under sward conditions, (ii) when it is difficult or impossible to apply meaningful selection pressure within families, or (iii) when time‐intensive phenotypic evaluations necessitate long cycle times, for example, 4 to 5 yr.

Pasture is the main food source for more than 200 million cattle heads in Brazil. Although Brazilian forage breeding programs have successfully released well-adapted, high-yielding cultivars over the years, the use of genomic tools in these programs is currently limited. These tools are required to tackle the main challenges for tropical forage breeding in Brazil. In this context, this note lists the main research priorities raised at the workshop "Breeding Forages in the Genomic Era", which are necessary to accelerate the use of genomic tools for next-generation breeding of tropical forages and allow breeders to increase genetic gains. Additionally, an online discussion forum (hosted at http://www. cnpgl.embrapa.br/genfor) has been launched to strengthen collaborations among research groups. The research priorities and more synergistic collaborations will assist researchers and decision-makers in delivering a sustainable increase in production of animal products, especially beef and milk, which are required to feed a rising world population.

Monitoring biomass of forages in experimental plots and livestock farms is a time-consuming, expensive, and biased task. Thus, non-destructive, accurate, precise, and quick phenotyping strategies for biomass yield are needed. To promote high-throughput phenotyping in forages, we propose and evaluate the use of deep learning-based methods and UAV (Unmanned Aerial Vehicle)-based RGB images to estimate the value of biomass yield by different genotypes of the forage grass species Panicum maximum Jacq. Experiments were conducted in the Brazilian Cerrado with 110 genotypes with three replications, totaling 330 plots. Two regression models based on Convolutional Neural Networks (CNNs) named AlexNet and ResNet18 were evaluated, and compared to VGGNet—adopted in previous work in the same thematic for other grass species. The predictions returned by the models reached a correlation of 0.88 and a mean absolute error of 12.98% using AlexNet considering pre-training and data augmentation. This proposal may contribute to forage biomass estimation in breeding populations and livestock areas, as well as to reduce the labor in the field.

The world population is expected to be larger and wealthier over the next few decades and will require more animal products, such as milk and beef. Tropical regions have great potential to meet this growing global demand, where pasturelands play a major role in supporting increased animal production. Better forage is required in consonance with improved sustainability as the planted area should not increase and larger areas cultivated with one or a few forage species should be avoided. Although, conventional tropical forage breeding has successfully released well-adapted and high-yielding cultivars over the last few decades, genetic gains from these programs have been low in view of the growing food demand worldwide. To guarantee their future impact on livestock production, breeding programs should leverage genotyping, phenotyping, and envirotyping strategies to increase genetic gains. Genomic selection (GS) and genome-wide association studies play a primary role in this process, with the advantage of increasing genetic gain due to greater selection accuracy, reduced cycle time, and increased number of individuals that can be evaluated. This strategy provides solutions to bottlenecks faced by conventional breeding methods, including long breeding cycles and difficulties to evaluate complex traits. Initial results from implementing GS in tropical forage grasses (TFGs) are promising with notable improvements over phenotypic selection alone. However, the practical impact of GS in TFG breeding programs remains unclear. The development of appropriately sized training populations is essential for the evaluation and validation of selection markers based on estimated breeding values. Large panels of single-nucleotide polymorphism markers in different tropical forage species are required for multiple application targets at a reduced cost. In this context, this review highlights the current challenges, achievements, availability, and development of genomic resources and statistical methods for the implementation of GS in TFGs. Additionally, the prediction accuracies from recent experiments and the potential to harness diversity from genebanks are discussed. Although, GS in TFGs is still incipient, the advances in genomic tools and statistical models will speed up its implementation in the foreseeable future. All TFG breeding programs should be prepared for these changes.