Deriving green crop area index and canopy chlorophyll density of winter wheat from spectral reflectance data

Broge, Niels; Mortensen, J.

doi:10.1016/s0034-4257(01)00332-7

Cited by 279 publications

(149 citation statements)

References 46 publications

Supporting

Mentioning

139

Contrasting

Unclassified

Order By: Relevance

“…Lee et al (2004) came to the conclusion that regression models using AVIRIS channels performed better to predict LAI than those based on broadband data. However, the advantage of hyperspectral over multispectral data does not always seem to be the case: for instance, Broge and Mortensen (2002) came to the conclusion that hyperspectral VI derived from field spectral measurements were not better at estimating green crop area index (a variable related to LAI) than traditional broadband VI. Then again, these authors found that the prediction of canopy chlorophyll density was improved using narrow bands across the red edge.…”

Section: Broadband Versus Hyperspectral VImentioning

confidence: 99%

“…A systematic investigation on the performance of various broadband multispectral and narrow band hyperspectral vegetation indices has been done on agricultural crops (Boegh et al, 2002;Broge & Mortensen, 2002) and very recently, also in forests. For instance, Gong et al (2003) estimated forest LAI using vegetation indices derived from Hyperion hyperspectral data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Remote sensing of forest biophysical variables using HyMap imaging spectrometer data

Schlerf¹,

Atzberger²,

Hill³

2005

Remote Sensing of Environment

278

171

View full text Add to dashboard Cite

This study systematically evaluated linear predictive models between vegetation indices (VI) derived from radiometrically corrected airborne imaging spectrometer (HyMap) data and field measurements of biophysical forest stand variables (n=40). Ratio-based and soil-linerelated broadband VI were calculated after HyMap reflectance had been spectrally resampled to Landsat TM channels. Hyperspectral VI involved all possible types of two-band combinations of ratio VI (RVI) and perpendicular VI (PVI) and the red edge inflection point (REIP) computed from two techniques, inverted Gaussian Model and Lagrange Interpolation. Cross-validation procedure was used to assess the prediction power of the regression models. Analyses were performed on the entire data set or on subsets stratified according to stand age. A PVI based on wavebands at 1088 nm and 1148 nm was linearly related to leaf area index (LAI) (R 2 =0.67, RMSE=0.69 m 2 m À2 (21% of the mean); after removal of one forest stand subjected to clearing measures: R 2 =0.77, RMSE=0.54 m 2 m À2 (17% of the mean). A PVI based on wavebands at 885 nm and 948 nm was linearly related to the crown volume (VOL) (R 2 =0.79, RMSE=0.52). VOL was derived from measured biophysical variables through factor analysis (varimax rotation). The study demonstrates that for hyperspectral image data, linear regression models can be applied to quantify LAI and VOL with good accuracy. For broadband multispectral data, the accuracy was generally lower. It can be stated that the hyperspectral data set contains more information relevant to the estimation of the forest stand variables LAI and VOL than multispectral data. When the pooled data set was analysed, soil-line-related VI performed better than ratio-based VI. When age classes were analysed separately, hyperspectral VI performed considerably better than broadband VI. Best hyperspectral VI in relation with LAI were typically based on wavebands related to prominent water absorption features. Such VI are related to the total amount of canopy water; as the leaf water content is considered to be relatively constant in the study area, variations of LAI are retrieved. D

show abstract

Section: Broadband Versus Hyperspectral VImentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Remote sensing of forest biophysical variables using HyMap imaging spectrometer data

Schlerf¹,

Atzberger²,

Hill³

2005

Remote Sensing of Environment

278

171

View full text Add to dashboard Cite

show abstract

“…NDVI is calculated utilising the great difference in canopy reflectance between the red and the near-infrared bands. Because of this chain of related properties, NDVI has been used as a valuable tool in estimating crop variables like LAI, crop biomass, grain yield and photosynthetic activity (Penuelas et al 1995;Broge and Mortensen 2002;Hansen and Schjoerring 2003;Groeneveld et al 2007;Xue et al 2007). Based on the above-mentioned data, this study explored the possibilities of estimating the RUE and HI from canopy reflectance by correlation analysis.…”

Section: Discussionmentioning

confidence: 99%

Radiation-use efficiency and the harvest index of winter wheat at different nitrogen levels and their relationships to canopy spectral reflectance

Luo

2011

Crop Pasture Sci.

View full text Add to dashboard Cite

Abstract. Radiation-use efficiency (RUE, g/MJ) and the harvest index (HI, unitless) are two helpful characteristics in interpreting crop response to environmental and climatic changes. They are also increasingly important for accurate crop yield simulation, but they are affected by various environmental factors. In this study, the RUE and HI of winter wheat and their relationships to canopy spectral reflectance were investigated based on the massive field measurements of five nitrogen (N) treatments. Crop production can be separated into light interception and RUE. The results indicated that during a long period of slow growth from emergence to regreening, the effect of N on crop production mainly showed up in an increased light interception by the canopy. During the period of rapid growth from regreening to maturity, it was present in both light interception and RUE. The temporal variations of RUE APAR (aboveground biomass produced per unit of photosynthetically active radiation absorbed by the canopy) during the period from regreening to maturity had different patterns corresponding to the N deficiency, N adequacy and N-excess conditions. Moreover, significant relationships were found between the RUE APAR and the accumulative normalised difference vegetation index (NDVI) in the integrated season (R 2 = 0.68), between the HI and the accumulative NDVI after anthesis (R 2 = 0.89), and between the RUE grain (ratio of grain yield to the total amount of photosynthetically active radiation absorbed by the canopy) and the accumulative NDVI of the whole season (R 2 = 0.89) and that after anthesis (R 2 = 0.94). It suggested that canopy spectral reflectance has the potential to reveal the spatial information of the RUE APAR , HI and RUE grain . It is hoped that this information will be useful in improving the accuracy of crop yield simulation in large areas.

show abstract

“…As the REP of the grassland in the Lagrangian method goes towards shorter wavelength and lower reflectance than the Linear method, it can be concluded that the Lagrangian approach offers more accurate results than the Linear method in estimating grassland, because the derivative approaches (Lagrangian technique) minimizes the soil background reflectance effects [27,[50][51][52][53][54][55] . In classifying different ages of coniferous species, both methods offer the same results in extracting REP for different ages of coniferous species.…”

Section: Performance Of the Lagrangian Interpolation Techniquementioning

confidence: 99%

Hyperspectral Remote Sensing of Vegetation Using Red Edge Position Techniques

Shafri¹,

Salleh²,

Ghiyamat³

2006

American J. of Applied Sciences

View full text Add to dashboard Cite

Two red edge position (REP) techniques, Linear and Lagrangian, were applied on hyperspectral data acquired from the HyMap sensor for a forested area in Thetford Forest, UK. Red edge positions of different vegetation covers were extracted with the two approaches from the hyperspectral data. Based on the estimated REPs, the Linear and Lagrangian interpolation methods were compared with ground reference image to analyse different vegetation types and ages. Experimental results of both interpolation techniques indicate that the wavelength and reflectance of REP for younger plants (higher chlorophyll content) shift towards longer wavelength and of higher reflectance in comparison with older plants (lower chlorophyll content).

show abstract

Deriving green crop area index and canopy chlorophyll density of winter wheat from spectral reflectance data

Cited by 279 publications

References 46 publications

Remote sensing of forest biophysical variables using HyMap imaging spectrometer data

Remote sensing of forest biophysical variables using HyMap imaging spectrometer data

Radiation-use efficiency and the harvest index of winter wheat at different nitrogen levels and their relationships to canopy spectral reflectance

Hyperspectral Remote Sensing of Vegetation Using Red Edge Position Techniques

Contact Info

Product

Resources

About