We investigated the use of full-range (400–2,500 nm) hyperspectral data obtained by sampling foliar reflectances to discriminate 46 plant species in a tropical wetland in Jamaica. A total of 47 spectral variables, including derivative spectra, spectral vegetation indices, spectral position variables, normalized spectra and spectral absorption features, were used for classifying the 46 species. The Mann–Whitney U-test, paired one-way ANOVA, principal component analysis (PCA), random forest (RF) and a wrapper approach with a support vector machine were used as feature selection methods. Linear discriminant analysis (LDA), an artificial neural network (ANN) and a generalized linear model fitted with elastic net penalties (GLMnet) were then used for species separation. For comparison, the RF classifier (denoted as RFa) was also used to separate the species by using all reflectance spectra and spectral indices, respectively, without applying any feature selection. The RFa classifier was able to achieve 91.8% and 84.8% accuracy with importance-ranked spectral indices and reflectance spectra, respectively. The GLMnet classifier produced the lowest overall accuracies for feature-selected reflectance spectra data (52–77%) when compared with the LDA and ANN methods. However, when feature-selected spectral indices were used, the GLMnet produced overall accuracies ranging from 79 to 88%, which were the highest among the three classifiers that used feature-selected data. A total of 12 species recorded a 100% producer accuracy, but with spectral indices, and an additional 8 species had perfect producer accuracies, regardless of the input features. The results of this study suggest that the GLMnet classifier can be used, particularly on feature-selected spectral indices, to discern vegetation in wetlands. However, it might be more efficient to use RFa without feature-selected variables, especially for spectral indices.

Most developing countries have failed to adopt hydroacoustics to aid with the management of their aquatic natural resources. We tested the ability of single-beam sonar (SBES) to discern and map substrates in three rivers from the largest wetland in Jamaica, the Black River Lower Morass (BRLM). We used five supervised classification methods (including C5.0; random forest, RF; and naïve Bayes, NB) and four interpolation algorithms (indicator kriging (iks), fixed path simulation (fpth), random path simulations (rpth) and multinomial categorical simulation (mcs) based on transitional rates and incorporated into Markov Chain). Irrespective of the classifier used, mcs consistently produced higher overall classification accuracies (OAC) and kappa statistics; however, rpth interpolation produced the lowest balanced error rate (BER) recorded. For all three rivers, OAC, kappa and BER statistics were 49.7–87.1, 32.8–81.0 and 15.3–45.1% respectively. All interpolation algorithms produced maps with higher OAC and kappa indices from data classified using the tree-based classifiers (C5.0 and RF) in the absence of gravel-free substrates. At a lower spatial resolution, comparable maps were obtained by interpolating discrete sample points acquired by grab samples. Given that most of rivers in island states are small, sinuous, shallow and sometimes non-navigable by boat, the use of SBES as the most cost-effective and efficient way of mapping river substrates is questionable, but the interpolation of grab samples might suffice.