Remote Sensing Leaf Area Index of Winter Wheat from Unmanned Airborne Vehicles (UAVs).
E. Raymond Hunt Jr.1, W. Dean Hively1, Stephen Fujikawa2, T. L. Ng2, Mike Tranchitella2, Woody Raszula2, David Yoel2, Craig Daughtry1, and Gregory McCarty1. (1) Hydrology and Remote Sensing Laboratory, USDA-ARS, Bldg. 007, Room 104, 10300 Baltimore Ave, Beltsville, MD 20705, (2) IntelliTech Microsystems, Inc., 4931 Telsa Drive, Suite B, Bowie, MD 20715
Remote sensing with unmanned airborne vehicles (UAVs) has more potential for within-season crop management than conventional satellite imagery because: (1) pixels have very high resolution, (2) cloud cover would not prevent acquisition during critical periods of growth, and (3) quick delivery of information to the user. Winter wheat was planted early (October 2006) and late (November 2006) in a field on the Eastern Shore of Maryland (39° 2’ 2” N lat, 76° 10’ 36” W long). Each planting was divided into 6 north-south strips, each with various levels of initial nitrogen fertilizer, which caused large variations in leaf area index, biomass, and yield. The Vector P from Intellitech Microsystems (BowieMaryland) was flown on three dates in late April/early May 2007 at two elevations. A color-infrared digital camera was mounted in the Vector P and the pixel sizes were 6 cm at 210 m elevation and 3 cm at 115 m elevation. Inspection of the color-infrared photographs revealed large spatial variation in biomass and leaf area index within each strip. Because pixel size was much smaller than the position error of the airborne global positioning system, field plots (20 cm by 50 cm) were located using visual features. As with most photography, there were problems in the imagery with vignetting and anisotropy. The green normalized difference vegetation index [GNDVI = (NIR - green)/(NIR + green)] reduced the image problems and was linearly correlated with leaf area index and biomass. There were no significant differences in the regressions of GNDVI and leaf area index based on pixel size. Variation in biomass was highly correlated to yields, so the GNDVI was also a good predictor of the spatial variability of yields.