Temporal dynamics of spatial heterogeneity over cropland quantified by time-series NDVI, near infrared and red reflectance of Landsat 8 OLI imagery

Yanling Ding,Kai Zhao*,Xingming Zheng, Tao Jiang.International Journal of Applied Earth Observation and Geoinformation/30:139-145

Abstract:Spatial heterogeneity is an important characteristic of the land surface. Because multi-spectral bands areused to describe the land surface, an approach has to be established to characterize the surface spatialheterogeneity from multi-spectral remote-sensing observations. This work aims at quantifying the spa-tial heterogeneity of cropland using varigorams for multi-temporal NDVI, near infrared (NIR) and redreflectance. A concept of mean length variability is proposed to compare the difference in spatial het-erogeneity detected by variables with different magnitudes. The important temporal changes in spatialheterogeneity observed by NDVI, NIR and red bands over cropland are a result of changes in the fractionof vegetation cover. The results indicate the following: (1) the NIR and red variables detect a similarspatial heterogeneity of the cropland with similar values of the mean length variability before the sow-ing of crops; (2) the NDVI, NIR and red values capture different degrees of spatial heterogeneity whenvegetation cover is low; (3) over medium vegetation cover, the NDVI and NIR values capture similarspatial heterogeneity, which is low compared to the red band due to the homogeneity of soil; and (4)the spatial heterogeneity quantified by the NIR value is more heterogeneous than those of the NDVI andred values when vegetation cover is high. The red reflectance is sensitive to soil properties while the NIRreflectance responds to vegetation. The spatial heterogeneity of red reflectance decreases and that of theNIR reflectance increases with the growth of vegetation. The NDVI value shows the greatest heterogene-ity in the early stage of crop growth. With an increase in the image pixel size, the spatial heterogeneityquantified by the mean length variability of the NDVI, NIR and red variables tends to be the same.

keywords:Spatial heterogeneity,Mean length variability,NDVI,Near infrared and red reflectance,Fractional vegetation cover