Assessing Canopy Spectral Reflectance of Hybrid Bermudagrass under Various Combinations of Nitrogen and Water Treatments

Abstract

Remote sensing technology was tested as a non-destructive tool to assess the nitrogen (N) and water status of hybrid bermudagrass 'Tifway' [Cynodon dactylon (L.) Pers. × C. transvaalensis Burtt-Davy]. Ten combinations of water (based on leaching fraction: LF) and N treatments including cyclic and steady treatments were imposed on 20 plots (6.1 × 6.1 m) from October 2001 to October 2002 and January 2003 to June 2003. Each plot contained a non-weighing lysimeter which was used to assess the weekly water balance. Leaf xylem water potential, canopy temperature, tissue moisture content (TM), chlorophyll meter index, visual color ratings, and tissue nitrogen concentration (TN) were assessed twice per month. Canopy spectral reflectance measurements were also acquired twice per month using a portable spectroradiometer mounted on a movable cart. Analysis of variance (ANOVA) showed that spectral reflectance at individual wavelengths, vegetation indices, and soil-plant parameters separated based on N, LF and/or N × LF interaction (P < 0.05). Prediction equations for individual wavelengths and vegetation indices were generated from plant and soil parameters that included TN (Nitrogen Content index (NC): R2 = 0.73), TM (R555 : R2 = 0.66), or TM × TN (Normalized Difference Vegetation Index (NDVI) : R2 = 0.71). Multiple regression analysis also produced significant predictive models for four plant variables based on a selected number of wavelengths (between 3 and 9): canopy color (R2 = 0.48), TN (R2 = 0.56), TM × TN (R2 = 0.43), and chlorophyll meter index (R2 = 0.78). Normalized difference vegetation index (NDVI) showed a strong threshold relationship with canopy color rating (R2 = 0.64, P < 0.001, n = 194) and a consistent positive curvilinear relationship with the chlorophyll meter index (R2 = 0.75, P < 0.001, n = 200). The Water Band Index (WBI) significantly correlated with TM (R2 = 0.76, P < 0.001) and N content index (NC) correlated with TN (R2 = 0.45, P < 0.001) for single day measurements, but not for the larger data sets. Results from this experiment demonstrate that canopy reflectance measured with a ground-based spectroradiometer has potential for improving turfgrass N and irrigation management under field conditions, which could lead to significant N and water savings on large turfgrass areas such as golf courses and sports fields.