Evaluation and Refinement of the Nitrogen Reflectance Index (NRI) for Site-Specific Fertilizer Management

Environmental and economic concerns have collectively created a demand for moreefficient agricultural fertilizer application. Remote sensing of plant parameters has recentlyshown the potential to improve fertilizer application efficiency by showing corn (Zea mays L.)producers when and where to apply fertilizers based on crop needs. Previous studies haveindicated that a normalized near-infrared over green (nir/green) waveband reflectance ratio,called the Nitrogen Reflectance Index (NRI), could be used to assess in-season corn N status(CNS) in small plots. This study evaluated the NRI using small plot data and then testedpossible index improvements in a larger area. Data from four small plot site years indicated thatthe NRI explained > 70% of the variability occurring with in-season CNS for canopies with LeafArea Index (LAI) 2. These results led to the development of an early-season feasibilityindicator (ESFI) a non-invasive method of assessing LAI. This created a simplistic guidelinefor appropriate use of the NRI based on minimum ground-cover. The ESFI, calculated bysubtracting the green and red reflectance, utilized a minimum threshold level of 0.01 to removemeasurements where LAI < 2. Another proposed improvement to the NRI illustrated that thecumulative density function of the nir/green ratio could be used to develop N variability mapsand improve fertilizer management. Further testing within a 22.5 ha area was used to determineplatform compatibility and performance comparisons of the NRI versus other indices. Afterfiltering with the ESFI, results showed strong compatibility between aerial imagery and ground-basedradiometer measurements. Comparing the NRI to the NDVI, Green NDVI, SAVI, andMSAVI showed that all of these indices accurately assessed N variability based on total Nuptake measurements collected.