Proximal Sensing Technology to Predict the Quality of Forage Sorghum for Optimal Time of Harvest.

Proximal sensing technology is a promising tool for predicting end of season yield using Normalized Difference Vegetation Index (NDVI) for variable rate nitrogen (N) applications in field crops. In forage production, mid-season prediction of forage quality using crop sensing may assist farmers in determining the optimal time of harvest. This study investigated the relationship between NDVI and forage quality of brachytic dwarf brown midrib (BMR) forage sorghum (Sorghum bicolor (L.) Moench) at different growth stages. Trials were established at two sites in upstate New York in 2015. Sensor scans were performed twice per week starting 19 days after planting (DAP) until 69 DAP. Sorghum was weekly harvested for 9 times from 63 DAP until 2 weeks after the soft dough stage. Forage quality indicators including crude protein (CP), acid detergent fiber (ADF), neutral detergent fiber (NDF), in vitro fiber digestibility at 30 hr fermentation (NDFD30), lignin, and starch were analyzed. Over the 9 weeks, on average, CP decreased from 10.8% to 8.2% of dry matter (DM), ADF from 29.6% to 28.4% of DM, NDF from 53.7% to 48.9% of DM, and NDFD30 from 37.9% to 28.7%, while lignin increased from 2.8% to 3.6% and starch from 8.5% to 20.2%. Yield increased by 6-8 Mg DM per hectare. Data analyses for determining the optimal timing of harvest for both yield and quality as well as the relationship between NDVI and forage sorghum quality are currently ongoing. Results and the viability of quality prediction using NDVI will also be presented.