Skip to main content
ARS Home » Northeast Area » University Park, Pennsylvania » Pasture Systems & Watershed Management Research » Research » Publications at this Location » Publication #328335

Title: Application of a rising plate meter for estimation of forage yield in multi-species swards

item Dillard, Sandra
item AIMEE, HAFLA - Agri-King, Inc
item Rubano, Melissa
item Stout, Robert
item Soder, Kathy

Submitted to: Third Grazing Livestock Nutrition Conference Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 6/25/2016
Publication Date: 6/25/2016
Citation: Dillard, S.L., Hafa, A., Rubano, M.D., Stout, R.C., Soder, K.J. 2016. Application of a rising plate meter for estimation of forage yield in multi-species swards.Proceedings of the 2016 Grazing Livestock Nutrition Conference,July 17-19,2016,Park City,Utah.P.1.

Interpretive Summary: No Interpretive Summary is needed for this Abstract Only.

Technical Abstract: The rising plate meter (RPM) provides rapid estimates of pasture forage yield. Accurate calibration of the RPM is difficult due to variability in forage management and growth. The RPM is typically calibrated by linear regression of actual yield and RPM height; however, the r2 is usually low. We hypothesized that curvilinear regression would provide a more robust calibration equation and decrease the inherent variability in RPM estimates. A study was conducted on 3 rotationally-grazed, dairy farms in PA with multi-species pastures. Actual and estimated forage yield was determined on each farm one day prior to grazing using 30 - 45 measurements from a FILIPS RPM (NZ Agriworks, Feilding, NZ) and destructive samples using a 1 m × 10 cm quadrat at the same location. Botanical composition was determined by visual assessment. Calibration equations, coefficients of determination, and standard errors of prediction were determined using PROC REG (SAS Institute Inc., Cary, NC) and regressing clipped forage yield against RPM height. First-order and second-order equations were developed and evaluated based on data pooled: 1) across all farms and seasons, 2) by farm, 3) by season, and 4) by farm ad season. Pasture composition was 23, 44, and 63% grasses, 48, 25, and 20% legumes, and 21, 21, and 19% weeds on Farm 1, 2, and 3, respectively. Standard error of prediction as a percent of measured yield across all farms was between 38.58 and 64.30% for all calibration equations. Quadratic farm × season calibration equations provided the best estimate of forage yield on Farm 2 and Farm 3 (r2 = 0.55 and 0.32, respectively); the linear farm × season calibration equation provided the best fit on Farm 1 (r2 = 0.48). Using the equation with the best r2 for each farm and season would have lead to ± 5.7, 6.6, and 4.9% error in estimated yield during the fall grazing season on Farm 1, 2, and 3, respectively. The error in estimated forage yield was higher in the summer grazing season than the fall on all farms (± 9.2, 7.1, and 10.6% on Farm 1, 2, and 3, respectively). Our results indicate that on-farm, seasonal calibrations are necessary to estimate forage yield on farms in PA with multi-species pastures. Additionally during the summer, when forage species tend to have a faster growth rate, more frequent calibration of the RPM is necessary to prevent large errors in yield estimations.