Long-term land cover changes play a significant driver of ecosystem and function of natural biodiversity. Hence, their analysis can be used for evaluating and supporting government plans, especially conservation and management of natural habitats such as sago palm. In Papua Province of Indonesia, sago palm has been stated as one of the priority plants in the Medium-Term Development Plan (R.P.J.M.). However, limited studies have examined this palm in one of the Regencies of Papua Province, namely, Merauke Regency. In this study, we performed remotely sensed data imagery and supervised classification to produce land cover maps from 1990 to 2019. During the study period, twenty-one land cover classes were identified. The six classes of the natural forest consist of primary dryland forest, secondary dryland forest, primary mangrove forest, secondary mangrove forest, primary swamp forest, and secondary swamp forest; thus, fifteen classes of non-forested area. Concerning the sago palm habitat, our study evaluated two different categories (1) based on the land cover scheme from the Ministry of Environment and Forestry and (2) according to the peatland land cover ecosystem in Papua. Based on paired samples t-test, the result indicated statistically significant changes specifically at primary dryland (p-value = 0.015), grassland (p-value = 0.002) and swamp (p-value = 0.007). Twelve from 20 districts of Merauke Regency tend to lose the forecasted natural habitat of the sago palm. Therefore, this study suggests the further need to recognize and estimate the yield of sago palm area in these various ecosystems.

Sago palm tree, known as Metroxylon Sagu Rottb, is one of the priority commodities in Indonesia. Based on our previous research, the potential habitat of the plant has been decreasing. On the other hand, while the use of remote sensing is now widely developed, it is rarely applied for detection and classification purposes, specifically in Indonesia. Considering the potential use of the plant, local farmers identify the harvest time by using human inspection, i.e., by identifying the bloom of the flower. Therefore, this study aims to detect sago palms based on their physical morphology from Unmanned Aerial Vehicle (UAV) RGB imagery. Specifically, this paper endeavors to apply the transfer learning approach using three deep pre-trained networks in sago palm tree detection, namely, SqueezeNet, AlexNet, and ResNet-50. The dataset was collected from nine different groups of plants based on the dominant physical features, i.e., leaves, flowers, fruits, and trunks by using a UAV. Typical classes of plants are randomly selected, like coconut and oil palm trees. As a result, the experiment shows that the ResNet-50 model becomes a preferred base model for sago palm classifiers, with a precision of 75%, 78%, and 83% for sago flowers (SF), sago leaves (SL), and sago trunk (ST), respectively. Generally, all of the models perform well for coconut trees, but they still tend to perform less effectively for sago palm and oil palm detection, which is explained by the similarity of the physical appearance of these two palms. Therefore, based our findings, we recommend improving the optimized parameters, thereby providing more varied sago datasets with the same substituted layers designed in this study.