Search

This study analyzed the sensitivities of three vegetation biochemical parameters [chlorophyll content (Cab), leaf water content (Cw), and leaf area index (LAI)] to the changes of canopy reflectance, with the effects of each parameter on the wavelength regions of canopy reflectance considered, and selected three vegetation indices as the optimization comparison targets of cost function. Then, the Cab, Cw, and LAI were estimated, based on the particle swarm optimization algorithm and PROSPECT + SAIL model. The results showed that retrieval efficiency with vegetation indices as the optimization comparison targets of cost function was better than that with all spectral reflectance. The correlation coefficients (R2) between the measured and estimated values of Cab, Cw, and LAI were 90.8%, 95.7%, and 99.7%, and the root mean square errors of Cab, Cw, and LAI were 4.73 microg x cm(-2), 0.001 g x cm(-2), and 0.08, respectively. It was suggested that to adopt vegetation indices as the optimization comparison targets of cost function could effectively improve the efficiency and precision of the retrieval of biochemical parameters based on PROSPECT + SAIL model.

This paper studied the clonal architecture of two divergent Leymus chinensis types (grey-green type and yellow-green type) in Songnen grassland, and compared their internode length, spacer length, interbranching length, interbranching angle, and ramet population density and height under the same habitat. The results showed that there was no significant difference in these clonal characteristics except spacer length and ramet population density between the two types of L. chinensis, and yellow-green type, with less spacer length and more ramet density than grey-green type, should be more adaptable to the resourceful habitat. Moreover, the V-indices of the clonal architecture of two divergent L. chinensis types were all close to 1, and the difference was not significant. Therefore, both of the two types belonged to typical guerilla clonal plant.