Skip to main content
ARS Home » Plains Area » Brookings, South Dakota » Integrated Cropping Systems Research » Research » Publications at this Location » Publication #146360
Title: USING MULTI-SPECTRAL IMAGERY TO EVALUATE CORN GROWN UNDER NITROGEN AND DROUGHT STRESSED CONDITIONS

Author
item Osborne, Shannon
item Schepers, James
item Schlemmer, Michael

Submitted to: Journal of Plant Nutrition
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/12/2003
Publication Date: 11/11/2004
Citation: Osborne, S.L., Schepers, J.S., Schlemmer, M.R. 2004. Using multi-spectral imagery to evaluate corn grown under nitrogen and drought stressed conditions. Journal of Plant Nutrition.

Interpretive Summary: Estimating in-season crop streses which have the potential of adversely affecting crop yield could assist producers in making in-season management decisions to correct for the particular stress. A field study was conducted to evaluate the use of multi-spectral imagery for estimating corn grain yield, in-season biomass and nitrogen (N) concentration under varying N and drought stresses. The experiment was conducted in a continuous corn system utilizing linear drive irrigation. Three irrigation treatments including dry land, irrigation based on 0.5 and full evapotranspiration (ET) and five N rates (0, 40, 80, 120, 240 lb N ac-1) were included in the experiment. Multi-spectral imagery consisted of four wavebands (blue, green, red, and near-infrared (NIR)). Imagery was collected on various dates throughout the growing season; biomass sampling was performed within two days of image collection. Grain yield, in-season biomass and N concentration increased with increasing N rate regardless of sampling date or year. Yield was only affected by irrigation during 1997 due to a significant difference in rainfall between 1997 and 1998, 247 versus 457 mm, respectively. Regression correlation coefficients for the 1998 values were generally higher compared to 1997 values across years for imagery collected at similar growth stages, possibly due to differences in sensor sensitivity or the increased plant response to applied N. Regression correlation coefficients increased as the growing season progressed. The green waveband and normalized difference greenness vegetation index (GNDVI) had the greatest ability to estimate grain yield in the presence of varying N and/or drought stresses. This study demonstrates the ability of multi-spectral imagery analysis to estimate grain production in the presence of N and/or drought stresses.

Technical Abstract: The in-season estimation of crop stresses which have the potential of adversely affecting crop yield and/or quality could allow producers to make in-season management decisions to correct for the particular stress. A field study was conducted to evaluate the use of multi-spectral imagery for estimating corn (Zea mays L.) grain yield, in-season biomass and nitrogen (N) concentration under varying N and drought stresses. The experiment was a split-plot design with three replications using a factorial arrangement of treatments. Three irrigation (whole-plot) treatments included dry land, irrigation based on 0.5 and full evapotranspiration (ET). Sub-plot treatments included five N rates (0, 45, 90, 134, 269 kg N ha-1). Multi-spectral imagery consisted of four wavebands; blue (485 nm +/- 35 nm), green (550 nm +/- 35 nm), red (660 nm +/- 30 nm), and near-infrared (NIR) (830 nm +/- 70nm). Imagery was collected on various dates throughout the growing season; biomass sampling was performed within two days of image collection. Grain yield, in-season biomass and N concentration increased with increasing N rate regardless of sampling date or year. Yield was only affected by irrigation during 1997 due to a significant difference in rainfall between 1997 and 1998, 247 versus 457 mm, respectively. Regression correlation coefficients for the 1998 values were generally higher compared to 1997 values across years for imagery collected at similar growth stages, possibly due to differences in sensor sensitivity or the increased plant response to applied N. Regression correlation coefficients increased as the growing season progressed. The green waveband and normalized difference greenness vegetation index (GNDVI) had the greatest ability to estimate grain yield in the presence of varying N and/or drought stresses. This study demonstrates the ability of multi-spectral imagery analysis to estimate grain production in the presence of N and/or drought stresses.