The chlorophyll content can indicate the general health of vegetation, and can be estimated from hyperspectral data. The aim of this study is to estimate the chlorophyll content of mangroves at different stages of restoration in a coastal wetland in Quanzhou, China, using proximal hyperspectral remote sensing techniques. We determine the hyperspectral reflectance of leaves from two mangrove species, Kandelia candel and Aegiceras corniculatum, from short-term and long-term restoration areas with a portable spectroradiometer. We also measure the leaf chlorophyll content (SPAD value). We use partial-least-squares stepwise regression to determine the relationships between the spectral reflectance and the chlorophyll content of the leaves, and establish two models, a full-wave-band spectrum model and a red-edge position regression model, to estimate the chlorophyll content of the mangroves. The coefficients of determination for the red-edge position model and the full-wave-band model exceed 0.72 and 0.82, respectively. The inverted chlorophyll contents are estimated more accurately for the long-term restoration mangroves than for the short-term restoration mangroves. Our results indicate that hyperspectral data can be used to estimate the chlorophyll content of mangroves at different stages of restoration, and could possibly be adapted to estimate biochemical constituents in leaves.2 of 15 expected to reach 70,000 hm 2 by 2020 [4]. The chlorophyll content is an indicator of vegetation stress and can be used as a basis for estimating other biochemical parameters [5] and should be measured when monitoring, restoring and managing coastal wetland ecosystems. As such, it would therefore be useful if there were methods that resource managers could use to monitor the chlorophyll content of mangroves from hyperspectral data [6][7][8][9].Mangroves adapt to high-salt habitats using a range of physiological, biochemical, and morphological mechanisms. It is generally accepted that mangroves fall into two types, namely salt-secreting species with salt glands and salt-excluding species without salt glands [10,11]. Aegiceras plants have salt glands that secrete Na and Cl through their leaves to maintain the salt balance while Kandelia plants are known as salt-rejecting mangrove species that isolate the matrix from salt water mainly through high negative pressure in the xylem [12,13]. The mechanisms used by mangroves are unique within the plant kingdom, and the two different survival modes of mangroves have led to some differences between the leaves of the two mangrove species. The growth and physiological metabolism of mangroves are distinctive during different stages of restoration [14,15]. To date, hyperspectral data have not been used to compare the growth of these two mangrove species at different stages of restoration; instead, remote sensing has been used frequently to classify mangrove species and estimate biomass [16][17][18][19][20]. For example, Jia et al. mapped the Maipu mangrove species using hyperspectral data from the...

Studying the stoichiometric characteristics of plant C, N, and P is an effective way of understanding plant survival and adaptation strategies. In this study, 60 fixed plots and 120 random plots were set up in a reed-swamp wetland, and the canopy spectral data were collected in order to analyze the stoichiometric characteristics of C, N, and P across all four seasons. Three machine models (random forest, RF; support vector machine, SVM; and back propagation neural network, BPNN) were used to study the stoichiometric characteristics of these elements via hyperspectral inversion. The results showed significant differences in these characteristics across seasons. The RF model had the highest prediction accuracy concerning the stoichiometric properties of C, N, and P. The R2 of the four-season models was greater than 0.88, 0.95, 0.97, and 0.92, respectively. According to the root mean square error (RMSE) results, the model error of total C (TC) inversion is the smallest, and that of C/N inversion is the largest. The SVM yielded poor predictive results for the stoichiometric properties of C, N, and P. The R2 of the four-season models was greater than 0.82, 0.81, 0.81, and 0.70, respectively. According to RMSE results, the model error of TC inversion is the smallest, and that of C/P inversion is the largest. The BPNN yielded high stoichiometric prediction accuracy. The R2 of the four-season models was greater than 0.87, 0.96, 0.84, and 0.90, respectively. According to RMSE results, the model error of TC inversion is the smallest, and that of C/P inversion is the largest. The accuracy and stability of the results were verified by comprehensive analysis. The RF model showed the greatest prediction stability, followed by the BPNN and then the SVM models. The results indicate that the accuracy and stability of the RF model were the highest. Hyperspectral data can be used to accurately invert the stoichiometric characteristics of C, N, and P in wetland plants. It provides a scientific basis for the long-term dynamic monitoring of plant stoichiometry through hyperspectral data in the future.

No abstract