Spectral response and the retrieval of canopy chlorophyll content under interspecific competition in wetlands - case study of wetlands in the Yangtze River Estuary

Chlorophyll content can effectively reflect the health (growth condition) of vegetation. At present, chlorophyll retrieval is mostly based on pure species and traditional spectral indices but seldom considers the influence of canopy structural differences caused by interspecific competition of mixed vegetation on the accuracy of chlorophyll retrieval. Therefore, in this research, the native species Phragmites australis and invasive species Spartina alterniflora in the wetland of the Yangtze River Estuary were studied, and the mixed growth of the two species in the field was simulated accurately by setting different competition ratios. The differences in canopy spectral characteristics caused by different competition ratios and the effect of interspecific competition on the retrieval accuracy of canopy chlorophyll content (CCC) were analyzed. It is found that different competition ratios will lead to different spectral characteristics of the vegetation canopy, but these differences poorly reflect the spectral dimension. To solve this problem, the Hilbert-Huang transform (HHT) theory is introduced to decompose the spectrum and obtain frequency domain information, which is combined with spectral information to research the spectral response characteristics and retrieval of CCC under different interspecific competition ratios. Finally, the necessity of considering interspecific competition under mixed growth is analyzed by using the PROSAIL-D model from remote sensing. The results show that (1) the retrieval accuracy of CCC based on pure spectral indices is lower than that based on the mixed indices under different competition ratios, especially under the competition ratio of P:S = 1:1, and the R-2 of the pure spectral index model is 0.714, but the R-2 of the model increases to 0.803 after the HHT indices are added, with the accuracy of the retrieval model with the mixed index being the highest, with R-2 and RMSE values of 0.874 and 17.378 mu g.cm-2, respectively. (2) There are differences in the distribution of spectral index values under different competition ratios based on the remote sensing mechanism results of simulation data from the PROSAIL-D model, especially under mixed growth of Phragmites australis and Spartina alterniflora. (3) The HHT indices ESP1 and ESP2 are more sensitive to changes in the competition ratio under different competition conditions, showing a greater range of values. All the above results indicate that the frequency domain indices can reflect the change in canopy structure of mixed vegetation more accurately, and it is necessary to consider the level of interspecific competition in the retrieval of CCC from mixed vegetation.