The Aral Sea was one of the largest lakes in the world, but almost 60,000 km 2 of the waterbody has dried up due to water withdrawal for irrigation. Afforestation on the desiccated seafloor could be important in preventing soil flation, dust storms, and negative impact on human health. In this study, we aimed to delineate potential vegetation establishment areas on the dried Aral Sea bed using remote-sensed data in support of the decision-making related to afforestation. Various indices such as normalized difference vegetation index (NDVI), topsoil grain size index (TGSI), soil salinity index (SSI), and normalized multiband drought index (NMDI) were calculated from the LANDSAT-8 OLI satellite imagery. As an indicator of vegetation existence, NDVI was classified into three groups and set as a base for classifying other indices by performing statistical analyses. Based on the decision tree method, indices were combined and the potential vegetation establishment area was detected. Higher NDVI was identified in the southeast than the west of the study area. The results of statistical analyses showed that TGSI had a positive correlation with NDVI, while SSI and NMDI had a negative correlation. Overall, the potential vegetation area comprised 7,295.21 km 2 (61.34%) of the 'unsuitable' area, 2,818.64 km 2 (23.7%) of the 'intermediate' area, 1,612.15 km 2 (13.56%) of the 'suitable' area, and 166.42 km 2 (1.4%) of the 'very suitable' area. The developed map enables to identify dried seafloor area suitable for vegetation establishment thus contributing to planning the land rehabilitation efforts and preventing further land degradation.
The applicability of deep learning to remote sensing is rapidly increasing in accordance with the improvement in spatiotemporal resolution of satellite images. However, unlike satellite images acquired in near-real-time over wide areas, there are limited amount of labeled data used for model training. In this article, three kinds of deep learning applications-data augmentation, semisupervised classification, and domain-adapted architecture-were tested in an effort to overcome the limitation of insufficient labeled data. Among the diverse tasks that can be used for classification, rice paddy detection in South Korea was performed for its ability to fully utilize the advantages of deep learning and high spatiotemporal image resolution. In the process of designing each application, the domain knowledge of remote sensing and rice phenology was integrated. Then, all possible combinations of the three applications were examined and evaluated with pixel-based comparisons in various environments and city-level comparisons using national statistics. The results of this article indicated that all combinations of the applications can contribute to increase classification performance, even though the uncertainty involved in imitating or utilizing unlabeled data remains. As the effectiveness of the proposed applications was experimentally confirmed, enhancement in the applicability of deep learning was expected in various remote sensing areas. In particular, the proposed applications would be significant when they are applied to a wide range of study areas and highresolution images, as they tend to require a large amount of learning data from diverse environments, owing to high intraclass heterogeneity.
Central Asian countries, which are included the Mid-Latitude Region (MLR), need to develop regional adaptive strategies for reducing Sand and Dust Storm (SDS)-induced negative damages based on adequate information and data. To overcome current limitation about data and assessment approaches in this region, the macroscale verified methodologies were required. Therefore, this study analyzed environmental conditions based on the SDS impacts and regional differences of SDS sources and receptors to support regional SDS adaptation plans. This study aims to identify environmental conditions based on the phased SDS impact and regional differences of SDS source and receptor to support regional adaptation plans in MLR. The Normalized Difference Vegetation Index (NDVI), Aridity Index (AI), and SDS frequency were calculated based on satellite images and observed meteorological data. The relationship among SDS frequency, vegetation, and dryness was determined by performing statistical analysis. In order to reflect phased SDS impact and regional differences, SDS frequency was classified into five classes, and representative study areas were selected by dividing source and receptor in Central Asia and East Asia. The spatial analysis was performed to characterize the effect of phased SDS impact and regional distribution differences pattern of NDVI and AI. The result revealed that vegetation condition was negatively correlated with the SDS frequency, while dryness and the SDS frequency were positively correlated. In particular, the range of dryness and vegetation was related to the SDS frequency class and regional difference based on spatial analysis. Overall, the Aral Sea and the Caspian Sea can be considered as an active source of SDS in Central Asia, and the regions were likely to expand into potential SDS risk areas compared to East Asia. This study presents the possibility of potential SDS risk area using continuously monitored vegetation and dryness index, and aids in decision-making which prioritizes vegetation restoration to prevent SDS damages with the macrolevel approach in the MLR perspective.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.