Developing Novel Non-Destructive Sampling Techniques for Assessing Botanical Composition in Grass-Legume Pastures.

To better manage forage productivity and quality, it is important to accurately and objectively monitor legume concentrations in grass-legume pastures. Currently researchers rely on subjective visual assessments or labor-intensive botanical hand separations to determine legume content in mixed pastures. Research published over the past decade has led to developments in photospectroscopy and remote sensing techniques useful in pasture research and management. Four non-destructive sampling techniques were compared to traditional botanical hand separations to determine their ability to make site-specific assessments of botanical composition in tall wheatgrass-alfalfa [Thinopyrum ponticum (Podp.); Medicago sativa L.] and Old World bluestem-alfalfa-yellow sweetclover pastures [OWB, Bothriochloa bladhii (Retz) S.T. Blake; Melilotus officinalis L.]. The analyses were low to moderately successful when used in tall wheatgrass-alfalfa pasture (R² = 0.27 to 0.67); visual assessment generated the highest coefficient of determination. While the visual and PowerPoint analyses (R² = 0.82) most accurately predicted legume content in OWB-legume pasture, ImageJ® and in-field canopy reflectance certainly warrant more attention in future trials (R² = 0.67). The somewhat low R² values may be partly explained by the fact that hand separation measures dry weight contribution whereas the alternative methods measure legume ground cover. Another season of data collection is planned for testing another environment. The ultimate goal is to apply these findings to automated scanners which offer researchers and producers rapid measurements of legume content in mixed species pastures.