675-10 Exploring Remote Sensing Applications in Perennial Ryegrass Seed Production.

Tuesday, 7 October 2008
George R. Brown Convention Center, Exhibit Hall E
Michael Flowers1, John Hart2, William Young2, Neil Christensen2, Carol Garbacik3, Mark Mellbye4, Tom Silberstein2 and Gail Gingrich2, (1)Oregon State University - Crop & Soil Sciences, Albany, OR
(2)Oregon State University, Corvallis, OR
(3)Crop and Soil Science, Oregon State University, Corvallis, OR
(4)Oregon State University, Albany, OR
Perennial ryegrass seed is an import crop in the Willamette Valley of Oregon.  The winter rainfall pattern of the region limits the amount of fall nitrogen (N) applied to perennial ryegrass seed crops.  Therefore, spring applications are recommended to meet crop N requirements.  While soil N measurements accurately predict spring N for cereal crops in the region, perennial ryegrass seed growers still rely on yield goal estimates and experience to formulate spring N rates.  Remote sensing has the potential to assist growers in optimizing spring N rates.  A research project was initiated in 2006 to study the use of remote sensing to estimate spring N rate and predict clean seed yield prior to harvest.  The study was conducted at two locations near Corvallis, Oregon.  A randomized complete block design with 21 N treatments and four replications was used.  Nitrogen treatments were arranged in a factorial design with three fall N rates (0, 45, and 90 kg N ha-1) and seven spring N rates (0, 45, 90, 135, 180, 225, and 270 kg N ha-1). Treatments were sampled in mid-February and late-March to determine in-season plant N status. Aerial images were obtained within one week of sampling.  Additionally, aerial images were obtained in late-May.  Treatments were harvested in mid-July.  Strong linear relationships were found between plant N status and aerial images in mid-February.  Similarly, strong curvilinear relationships between plant N status and aerial images were found in late-March. Clean seed yield was also strongly related to aerial images obtained in late-May.  These results indicate that remote sensing has the potential to predict spring N rates and clean seed yield in perennial ryegrass seed production.  However, further research is required to develop a robust model across environments, tillage systems, and varieties.