Hyperspectral Analysis and Regression Modeling of SPAD Measurements in Leaves of Three Mangrove Species

Abstract

Mangroves have important roles in regulating climate change, and in reducing the impact of wind and waves. Analysis of the chlorophyll content of mangroves is important for monitoring their health, and their conservation and management. Thus, this study aimed to apply four regression models, eXtreme Gradient Boosting (XGBoost), Random Forest (RF), Partial Least Squares (PLS) and Adaptive Boosting (AdaBoost), to study the inversion of Soil Plant Analysis Development (SPAD) values obtained from near-ground hyperspectral data of three dominant species, Bruguiera sexangula (Lour.) Poir. (B. sexangula), Ceriops tagal (Perr.) C. B. Rob. (C. tagal) and Rhizophora apiculata Blume (R. apiculata) in Qinglan Port Mangrove Nature Reserve. The accuracy of the model was evaluated using R2, RMSE, and MAE. The mean SPAD values of R. apiculata (SPADavg = 66.57), with a smaller dispersion (coefficient of variation of 6.59%), were higher than those of C. tagal (SPADavg = 61.56) and B. sexangula (SPADavg = 58.60). The first-order differential transformation of the spectral data improved the accuracy of the prediction model; R2 was mostly distributed in the interval of 0.4 to 0.8. The accuracy of the XGBoost model was less affected by species differences with the best stability, with RMSE at approximately 3.5 and MAE at approximately 2.85. This study provides a technical reference for large-scale detection and management of mangroves.

Can SPAD Values and CIE L*a*b* Scales Predict Chlorophyll and Carotenoid Concentrations in Leaves and Diagnose the Growth Potential of Trees? An Empirical Study of Four Tree Species

“…For example, for mangrove plants, a report showed that the SPAD reading ranges for Avicennia marina, Rhizophora stylosa, Bruguiera gymnorhiza, and Aegiceras corniculatum were 41. 50-56.8, 70.40-80.10, 52.80-69.90, and 42.30-73.60, respectively [42], while another study recorded that the average values for Bruguiera sexangula, Ceriops tagal, and Rhizophora apiculata were 58.60, 61.56, and 66.57, respectively [23]. Intrinsic plant factors (such as developmental stages) and external factors (such as light intensity) can both affect SPAD values.…”

“…One of them is a CIE L*a*b* system, such as an NR20XE Precision Colorimeter (Shenzhen ThreeNH Technology, Shenzhen, China) [18], a Hand-held Color Analyzer (CM-2500D, Konica Minolta, Osaka, Japan) [19], or a Minolta Chroma Meter (Model CR200/300, Minolta, Ramsey, NJ, USA) [20]. Another is a portable Chl meter (SPAD-502Plus, Minolta Camera Co., Osaka, Japan) [21][22][23]. The SPAD-502 has been widely used for assessing or diagnosing a plant's nitrogen status [24][25][26], health status [27], pest and disease stress [28], leaf senescence [29], and coloration effects [30].…”