An assessment of angular variations of red and NIR reflectances in multi-date IRS-1D wide field sensor data

Mahtaba, Anjum ; Sridhara, V. N. ; Navalgund, R. R. (2009) An assessment of angular variations of red and NIR reflectances in multi-date IRS-1D wide field sensor data International Journal of Remote Sensing, 30 (17). pp. 4599-4619. ISSN 0143-1161

Full text not available from this repository.

Official URL: http://www.tandfonline.com/loi/gfer20?open=30#vol_...

Related URL: http://dx.doi.org/10.1080/01431160802621226

Abstract

A study has been carried out to assess angular variations in red and near infrared (NIR) reflectance of different features of the Earth's surface in a common overlap area of accumulated four-date Indian Remote Sensing Satellite (IRS-1D) Wide Field Sensor (WiFS) data from the first fortnight of October 2003. An improved dark object subtraction (DOS) method has been used to perform image based atmospheric corrections. Red and NIR reflectance variations of four structurally different classes-dense vegetation (shrub), sparse crop (pearl millet/maize), wasteland and forest with Sun-target-sensor geometry were analysed. A linearly constrained least squares technique was used to estimate red and NIR model coefficients of the linear Ross Thick-Li Sparse (RTLS) semi- empirical Bidirectional Reflectance Distribution Function (BRDF) model and compared with Moderate Resolution Imaging Spectrometer (MODIS) BRDF product coefficients. The relative reflectance difference between two dates as well as anisotropic factors for red and NIR for all classes and dates were also computed. Red and NIR reflectance of the four land cover classes at different locations with different observation geometry were estimated using both WiFS derived and MODIS BRDF product RTLS model coefficients and compared with WiFS observed reflectance. It was found that the mean relative difference in red and NIR reflectances between consecutive dates varied between 4-11% and 6-8%, respectively,while the computed mean anisotropy factors varied between 3-10% in the red and 8-11% in the NIR. A small anisotropy in the Normalized Difference Vegetation Index (NDVI) as a function of the scattering angle was observed for the four land cover classes. This may imply that angular effects in WiFS are relatively small and do not exceed 10-11 % for the land cover classes considered here.

Item Type:Article
Source:Copyright of this article belongs to Taylor and Francis Group.
ID Code:89344
Deposited On:26 Apr 2012 13:16
Last Modified:26 Apr 2012 13:16

Repository Staff Only: item control page