Skip to main content
ARS Home » Midwest Area » Madison, Wisconsin » U.S. Dairy Forage Research Center » Dairy Forage Research » Research » Publications at this Location » Publication #436637
Research Project: Developing Sustainable Forage and Cover Crop Systems for Dairy Farms

Location: Dairy Forage Research

Title: Insights on Forage Variety Testing from a Multi-Decade, Multi-Species Dataset

Author
item Tilhou, Neal
item Riday, Heathcliffe

Submitted to: North American Alfalfa Improvement Conference
Publication Type: Abstract Only
Publication Acceptance Date: 5/4/2026
Publication Date: 7/15/2026
Citation: Tilhou, N.W., Riday, H. 2026. Insights on Forage Variety Testing from a Multi-Decade, Multi-Species Dataset. North American Alfalfa Improvement Conference. Presented at conference.

Interpretive Summary:

Technical Abstract: Public forage variety trials are the only objective source of information for forage producers when purchasing seed. However, there are unresolved questions about how to interpret trial results from different years or regions. We have carried out a retrospective analysis of hundreds of alfalfa, red clover, tall fescue, and orchardgrass trials from across the greater northeastern U.S. In this study, individual trial results have limited reliability within a broader set of trials. Instead, producers should trust aggregated performance data from multiple trials. For example, alfalfa yield standard errors are 4.1%, 3.3%, and 2.9% (mean deviation from trial yield) for varieties evaluated in two, three, and four trials, respectively. Location-specific (genotype-by-environment or local adaptation) effects were unexpectedly small, indicating that information can be shared among states. In alfalfa, there was clear progress for biomass yield (+0.63% per year between 1988-2017; std. err.=0.029%; p=9.0x10-78). In red clover, biomass yield improvement has been slower (0.12% per year; std. err.=0.04%; p=0.007). Curiously, there is no evidence of orchardgrass yield improvement (0.03% per year, std. err.=0.04%, p=0.41), but there has been a large shift to later flowering time (1.5 days per decade; p=9.2x10-6) and this appears to artificially reduce trial yield. Tall fescue also has limited evidence of yield progress (<0.01% per year; std. err.=0.06%; p=0.34), but breeding has been focused on eliminating or replacing toxic endophytes.