Estimation of Pure NDVI Seasonal Profiles of Crops by Temporal Unmixing of MODIS Dataset

DOI JASI Web Site

Bibliographic Information

Other Title
  • ミクセル分解法を用いたMODISデータセットからの作物別NDVI季節変化の推定
  • ミクセル ブンカイホウ オ モチイタ MODIS データセット カラ ノ サクモツ ベツ NDVI キセツ ヘンカ ノ スイテイ

Search this article

Description

Abundant temporal information of MODIS data is useful for capturing phenological pattern of crops and for estimating their productivity with high reliability. However, the spatial resolution of MODIS data is so coarse that several types of land cover may simultaneously occur within a pixel, thus it is made difficult to extract pure spectral response of crops from MODIS data only. In this study, we tried to estimate pure NDVI seasonal profiles of irrigated crops in south central Turkey by the application of the unmixing model in combination with precise land cover maps derived from Landsat TM images. Estimated profiles showed a good agreement with the NDVI seasonal profiles from pure pixels with an average relative error of 0.05 for crops. The results also agreed well with the crop calendar of the area. A fair correlation (R2=0.74) was found between the estimated NDVI and crop coefficient (Kc) of the irrigation planning in the area at full cover, indicating that the estimated NDVI can be employed to estimate evapotranspiration to some extent.

Journal

Keywords

Details 詳細情報について

Report a problem

Back to top