Crop phenotyping in a context of global change: What to measure and how to do it.

High-throughput crop phenotyping, particularly under field conditions, is nowadays perceived as a key factor limiting crop genetic advance. Phenotyping not only facilitates conventional breeding, but it is necessary to fully exploit the capabilities of molecular breeding, and it can be exploited to predict breeding targets for the years ahead at the regional level through more advanced simulation models and decision support systems. In terms of phenotyping, it is necessary to determined which selection traits are relevant in each situation, and which phenotyping tools/methods are available to assess such traits. Remote sensing methodologies are currently the most popular approaches, even when lab-based analyses are still relevant in many circumstances. On top of that, data processing and automation, together with machine learning/deep learning are contributing to the wide range of applications for phenotyping. This review addresses spectral and red-green-blue sensing as the most popular remote sensing approaches, alongside stable isotope composition as an example of a lab-based tool, and root phenotyping, which represents one of the frontiers for field phenotyping. Further, we consider the two most promising forms of aerial platforms (unmanned aerial vehicle and satellites) and some of the emerging data-processing techniques. The review includes three Boxes that examine specific case studies.

Preharvest phenotypic prediction of grain quality and yield of durum wheat using multispectral imaging.

Durum wheat is an important cereal that is widely grown in the Mediterranean basin. In addition to high yield, grain quality traits are of high importance for farmers. The strong influence of climatic conditions makes the improvement of grain quality traits, like protein content, vitreousness, and test weight, a challenging task. Evaluation of quality traits post-harvest is time- and labor-intensive and requires expensive equipment, such as near-infrared spectroscopes or hyperspectral imagers. Predicting not only yield but also important quality traits in the field before harvest is of high value for breeders aiming to optimize resource allocation. Implementation of efficient approaches for trait prediction, such as the use of high-resolution spectral data acquired by a multispectral camera mounted on unmanned aerial vehicles (UAVs), needs to be explored. In this study, we have acquired multispectral image data with an 11-band multispectral camera mounted on a UAV and analyzed the data with machine learning (ML) models to predict grain yield and important quality traits in breeding micro-plots. Combining 11-band multispectral data for 34 cultivars and 16 environments allowed to develop ML models with good prediction capability. Applying the trained models to test sets explained a considerable degree of phenotypic variance with good accuracy showing r squared values of 0.84, 0.69, 0.64, and 0.61 and normalized root mean squared errors of 0.17, 0.07, 0.14, and 0.03 for grain yield, protein content, vitreousness, and test weight, respectively.

Bridging the genotype-phenotype gap for a Mediterranean pine by semi-automatic crown identification and multispectral imagery.

Progress in high-throughput phenotyping and genomics provides the potential to understand the genetic basis of plant functional differentiation. We developed a semi-automatic methodology based on unmanned aerial vehicle (UAV) imagery for deriving tree-level phenotypes followed by genome-wide association study (GWAS). An RGB-based point cloud was used for tree crown identification in a common garden of Pinus halepensis in Spain. Crowns were combined with multispectral and thermal orthomosaics to retrieve growth traits, vegetation indices and canopy temperature. Thereafter, GWAS was performed to analyse the association between phenotypes and genomic variation at 235 single nucleotide polymorphisms (SNPs). Growth traits were associated with 12 SNPs involved in cellulose and carbohydrate metabolism. Indices related to transpiration and leaf water content were associated with six SNPs involved in stomata dynamics. Indices related to leaf pigments and leaf area were associated with 11 SNPs involved in signalling and peroxisome metabolism. About 16-20% of trait variance was explained by combinations of several SNPs, indicating polygenic control of morpho-physiological traits. Despite a limited availability of markers and individuals, this study is provides a successful proof-of-concept for the combination of high-throughput UAV-based phenotyping with cost-effective genotyping to disentangle the genetic architecture of phenotypic variation in a widespread conifer.

Assessing durum wheat ear and leaf metabolomes in the field through hyperspectral data.

Hyperspectral techniques are currently used to retrieve information concerning plant biophysical traits, predominantly targeting pigments, water, and nitrogen-protein contents, structural elements, and the leaf area index. Even so, hyperspectral data could be more extensively exploited to overcome the breeding challenges being faced under global climate change by advancing high-throughput field phenotyping. In this study, we explore the potential of field spectroscopy to predict the metabolite profiles in flag leaves and ear bracts in durum wheat. The full-range reflectance spectra (visible (VIS)-near-infrared (NIR)-short wave infrared (SWIR)) of flag leaves, ears and canopies were recorded in a collection of contrasting genotypes grown in four environments under different water regimes. GC-MS metabolite profiles were analyzed in the flag leaves, ear bracts, glumes, and lemmas. The results from regression models exceeded 50% of the explained variation (adj-R2 in the validation sets) for at least 15 metabolites in each plant organ, whereas their errors were considerably low. The best regressions were obtained for malate (82%), glycerate and serine (63%) in leaves; myo-inositol (81%) in lemmas; glycolate (80%) in glumes; sucrose in leaves and glumes (68%); γ-aminobutyric acid (GABA) in leaves and glumes (61% and 71%, respectively); proline and glucose in lemmas (74% and 71%, respectively) and glumes (72% and 69%, respectively). The selection of wavebands in the models and the performance of the models based on canopy and VIS organ spectra and yield prediction are discussed. We feel that this technique will likely to be of interest due to its broad applicability in ecophysiology research, plant breeding programmes, and the agri-food industry.