The rate of improvement in genetic yield potential (YP) of wheat has stagnated and is now much too slow to keep pace with anticipated future demand. There is an urgent need to develop more-efficient breeding and selection tools to accelerate YP gains. Radiation-use efficiency (RUE = biomass/intercepted radiation) is proposed as one focus-point for future YP improvement. However, RUE is conventionally very tedious to measure. Here we report on work to develop and evaluate several new, higher-throughput techniques for assessing genotypic variation in RUE and associated traits in modern wheats.

Field trials involving multiple germplasm sets (RIL populations, advanced breeding lines, released varieties) and many hundreds of plots were grown under irrigated, high-input conditions over multiple seasons in SE Australia. Grain yield (GY), yield components, harvest index (HI) and crop height were measured at season end in all trials. Flowering date was also determined. For some trials, ground-level cuts were taken to obtain biomass at anthesis. These biomass measurements were combined with periodic pre-anthesis measurements of canopy light interception and the time-integral of incoming radiation to calculate RUE to anthesis.

A range of additional techniques was used to collect very large amounts of phenotypic information on RUE traits of canopies and of leaves. These techniques included: LiDAR, multi-spectral reflectance (NDVI) and RGB imagery using instruments mounted on a motorised ground-based buggy; thermal imagery for canopy temperature (CT) using both an aerial, helicopter-based platform and ground-based, logging, in-plot sensors; hyperspectral reflectance and chlorophyll fluorescence at leaf-level using hand-held instruments; light interception using in-plot, logging radiation sensors. Several phenotypic indicators of RUE were then extracted. Some data types (e.g. CT) were used effectively unmodified, whereas other data types were used to derive additional, newly-developed or well-established indices. The bulk of these indices were measures or indicators of variation in RUE.

These studies revealed substantial genotypic variation in RUE within the wheat germplasm grown. A substantial proportion of this variation was associated with variation in canopy architecture or other traits related to carbon gain. The studies also revealed strong potential for the future use of several of the higher-throughput techniques evaluated here at the plot scale in germplasm screening and/or selection for greater RUE in breeding. These include leaf hyperspectral reflectance, CT and NDVI used either singly or in combination, LiDAR-based estimation of anthesis biomass and light interception to anthesis and the combination of these to calculate a LiDAR-based estimate of RUE. 018971