Abstract:The normalized difference vegetation index(NDVI) is the most widely used vegetation index,which is not only applied into many operational applications,but also is an important parameter for some models.The objective of the study is to evaluate the effect of the red band and near-infrared(NIR) bands on NDVI,which is the function of red and NIR spectral bands,and to explore the relationship between band position and bandwidth of NDVI with the aim of monitoring the biophysical and biochemical parameters of rice.Canopy hyperspectral reflectance data of rice at seven stages were collected by portable spectroradiometer with a spectral range from 350 nm to 2500 nm.Extreme values of rice reflectance spectra in the visional and NIR regions at different stages were identified in order to determine the suitable band center positions for red and NIR bands of NDVI.Through the analysis of hyperspectral reflectance,the spectral bands at 674 nm and 860 nm were selected as red and NIR bands center positions,and for the computation NDVI.Then,the NDVI values were calculated under variable band position and bandwidth from hypersectral data of rice at different stages.Three scenarios were adopted to study the effects of band position and bandwidth on NDVI as follows.The NDVI values were calculated(1) from a constant red band,centered at 674nm with a 10-nm bandwidth,and variable NIR band positions and bandwidths for rice,(2) from a constant NIR band,centered at 860 nm with 10-nm bandwidth,and variable red band positions and bandwidths,(3) from a constant NIR band,centered at 860 nm with a 10-nm bandwidth,and variable red band bandwidths with 674nm,645nm as band center position,respectively.Moreover,the interactions between bandwidth and the central position of red band under the a precision requirement of 99% were analyzed too.The result indicated that band position and bandwidth of NIR channel have no significant influence on the NDVI of rice canopy at different development stages while those of red channel affect NDVI significantly,especially when band position approaches the red minimum(near 670nm).That means NDVI is more sensitive to band position and bandwidth of red band than those of NIR band.The sensitivity varied with rice development stages.Relative to the middle and later development stages of rice,the NDVI values derived from reflectance data at early development stages are more easily affected by band position.Although NDVI of rice was sensitive to band position,the difference between NDVI values was less affected by band position,therefore,the data sources should be consistent when NDVI was used to study rice.In addition,the NDVI of rice at different developments was essentially affected by proximity of the red and NIR bands to red edge region(690—740 nm).Under the 1% error of NDVI,the bandwidth of red band becomes narrower when the central position of the band moves towards longer wavelengths at bloom stages,and it reaches its minimum at round 690 nm.Beyond that position,the bandwidth becomes a little wider.However,for early and later development stages,the variation of red bandwidth with central position was more complicated due to the fluctuation of red bandwidth around 648 nm.The research on the effects of band position and bandwidth on NDVI will provide useful information for remote sensing of rice.

Effects of Band Position and Bandwidth on NDVI Measurements of Rice at Different Growth Stages

Rice Yield Prediction with Panicle Spectral Indices

Relationship Between Narrow Band Normalized Deference Vegetation Index and Rice Agronomic Variables

[Estimation of Rice LAI by Using NDVI at Different Spectral Bandwidths].

Hyperspectral Vegetation Indices and Their Relationships with Rice Agronomics Variables

Potential of RapidEye and WorldView-2 Satellite Data for Improving Rice Nitrogen Status Monitoring at Different Growth Stages

Comparability of Red/Near-Infrared Reflectance and NDVI Based on the Spectral Response Function Between MODIS and 30 Other Satellite Sensors Using Rice Canopy Spectra.

New Vegetation Index and Its Application in Estimating Leaf Area Index of Rice

Wavebands Selection for Rice Information Extraction Based on Spectral Bands Inter-Correlation

Using an Active-Optical Sensor to Develop an Optimal NDVI Dynamic Model for High-Yield Rice Production (yangtze, China).

Estimating Leaf Nitrogen Concentration with Three-Band Vegetation Indices in Rice and Wheat

A Comparative Analysis of Broadband and Narrowband Derived Vegetation Indices in Predicting LAI and CCD of a Cotton Canopy

Assessing the Impact of Spatial Resolution on the Estimation of Leaf Nitrogen Concentration Over the Full Season of Paddy Rice Using Near-Surface Imaging Spectroscopy Data

Feasibility of field evaluation of rice nitrogen status from reflectance spectra of canopy

Estimation of rice yield per unit using MODIS spectral index

Analysis of Common Canopy Reflectance Spectra for Indicating Leaf Nitrogen Concentrations in Wheat and Rice

Monitoring Leaf Photosynthesis with Canopy Spectral Reflectance in Rice

Research on Sensitive Spectrum of Rice Canopy for Nitrogen under Nitrogen and Phosphorus Interaction

Rice Mapping and Growth Monitoring Based on Time Series GF-6 Images and Red-Edge Bands

RELATIONSHIP BETWEEN SPECTRAL VEGETATION INDICES AND LAI IN RICE

A New Spectral Index for Estimating Protein Nitrogen Concentrations in Top Leaves of Rice