Comparing mechanical harvest with alternative ground‐based methods for estimating forage yields in cool‐season annual grasses

Abstract

Abstract To accurately estimate forage yield of experimental lines and cultivars without the time and expense of harvesting with and maintaining a plot harvester would aid cool‐season forage grass researchers and breeders in cultivarselection and advancement. A cool‐season annual grass forage trial was conducted at the Texas A&M Research Farm during the 2015–2016 and 2016–2017 growing seasons. The trial included commercial and experimental small grains and ryegrass cultivars that include wheat (Triticum aestivum L.), rye (Secale cereale L.), oat (Avena sativa L.), triticale (Triticum secale), barley (Hordeum vulgare L.), and annual ryegrass (Lolium multiflorum Lam.). The trial was set up in a randomized complete block design (RCBD) with four replications of 37 entries in 2016 and 38 in 2017. Visual rating (VR), plant height (HT), and sample weights were collected. The objective of this research was to compare mechanically harvested aboveground biomass yield of a multiple species cool‐season annual grasses with limited or nondestructive measurements to estimate plot yield. The VR evaluation was most successful in the fall (r = .84, p < .001) compared to other measurement dates. Plant height had the highest correlation to yield in the fall (r = .81, p < .001) across species, but when combined across species and measurement date it decreased to r = .15 (p < .001). Subsampling would be the only acceptable method for comparisons of entries across measurement dates, species, and years with an r = .83 (p < .001). While subsampling provided a reliable correlation, combining all measurements into one model improved forage yield prediction as an alternative to mechanical whole plot forage harvest.