PSXI-12 Non-destructive assessment of green biomass in natural grasslands for precision livestock in the Pampa Biome

• Published in: Journal of animal science 2018-12, Vol.96 (suppl_3), p.221-221
• Main Authors: Kuhn, J, Bremm, C, Quevedo da Rosa, F, Díaz, S, Jaurena, M, Lattanzi, F, de Faccio Carvalho, P
• Format: Article
• Language: English
• Subjects: Abstracts
• ISSN: 0021-8812; 1525-3163
• DOI: 10.1093/jas/sky404.481

Abstract Maximizing forage intake rate is one strategy to improve the performance of cattle grazing natural grasslands of the Pampa Biome. This rests on understanding animal responses to variation in both quantity and nutritive value of these heterogeneous pastures. The aim was to assess the ability of data on sward structure, botanical composition, and the Normalized Difference Vegetation Index (NDVI) to predict the percentage of green material (% GM) in total biomass. For this, we evaluated five 0.25 m2 quadrats of different percentages of visual green material in total biomass (55%, 60%, 65%, 70%, 75%, 80%, 85% and 90%) by experimental unit of an area of natural grassland at the Experimental Station of the Federal University of Rio Grande do Sul, in which forage allowance (FA) is controlled at 4, 8, 12 or 16 kg dry matter/ 100 kg live weight since 1986, using the ‘put- and- take’ method with crossbred steers. Data used in the models were collected in 11/23/2017. In these, sward height (SH) was measured with a sward stick, and botanical composition using a modified Braun-Blanquet abundance scale in which all plant species present in the quadrat are recorded and their coverage estimated visually when > 5%. The NDVI) was measured with the handheld crop sensor ‘Greenseeker’, positioned 85 to 115 cm above the canopy surface. Finally, the forage in the quadrat was harvested at ground level, manually separated into green) and dead material, and then each fraction was oven dried at 55 °C for 72 hours, and then weighed. Dominant botanical families were Poaceae (64 %), Asteraceae (14 %) and Cyperaceae (11 %). The best adjusted multiple regression model for % GM was: y = 39.2 + 70.2*NDVI - 2.86*SH - 2.93*FA (16&12 - 8& 4) - 3.20*FA (8 - 4), (R2=0.79, p