Classification of different tree species in semiarid areas can be challenging as a result of the change in leaf structure and orientation due to soil moisture constraints. Tree species mapping is, however, a key parameter for forest management in semiarid environments. In this study, we examined the suitability of 5-band RapidEye satellite data for the classification of five tree species in mopane woodland of Botswana using machine leaning algorithms with limited training samples.We performed classification using random forest (RF) and support vector machines (SVM) based on EnMap box. The overall accuracies for classifying the five tree species was 88.75 and 85% for both SVM and RF, respectively. We also demonstrated that the new red-edge band in the RapidEye sensor has the potential for classifying tree species in semiarid environments when integrated with other standard bands. Similarly, we observed that where there are limited training samples, SVM is preferred over RF. Finally, we demonstrated that the two accuracy measures of quantity and allocation disagreement are simpler and more helpful for the vast majority of remote sensing classification process than the kappa coefficient. Overall, high species classification can be achieved using strategically located RapidEye bands integrated with advanced processing algorithms.

Land use and land cover (LULC) mapping is very important for evaluation and management of natural resources. Classification of remote sensing data in mountainous and other heterogeneous environments are particularly problematic because of their high spectral heterogeneity. The aim of this study is to conduct accuracy analyses of LULC classification of Landsat-8 OLI composite images and multisource ancillary data using four machine learning classifiers on the Google Earth Engine (GEE) platform in the Golden Gate Highland National Park (GGHNP) and its near surrounding. Our results indicate that the inclusion of short-wave infrared bands increased accuracy of image composites by more than 10% in landcover types with wider range of spectral characteristics. Best performance was achieved by MaxEnt with overall accuracy and kappa statistics of 0.846 and 0.775, respectively. Support Vector Machine (SVM), however, produces relatively higher accuracy for handling different characteristics of multisource data. In elevated areas, aspect was found to outperform other topographic indices in reducing misclassification errors associated with confusion of grasslands with bare lands and water. The LULC produced from image composites can provide some reliable information on vegetation, land and water characteristics and reduce confusion associated with complex biophysical environments on vegetation spectral signatures in sub-tropical mountainous grasslands.