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 high-resolution images, as they tend to require a large amount of learning data from diverse environments, owing to high intra-class heterogeneity.

The integration of remote sensing and state-of-the-art deep learning models has enabled the generation of highly accurate semantic segmentation maps to serve the agricultural sector, for which continuous land monitoring is required. However, despite their wide presence in the research field, only a few such products are used in on-site decision-making processes. This is due to their incompatibility with existing datasets that are at the core of current operating processes. In this study, paddy rice mapping in South Korea was examined to determine whether it produces qualified products that can complement on-site surveys and simultaneously be compatible with existing domestic datasets. Cases of early predictions for timely rice supply control were examined using a recurrent U-Net architecture with diverse applications: chronological batch training (CBT), time-inversed padding (TIP), and super-resolution (SR). In addition, the paddy area was confirmed using diverse datasets by standardizing its spatial extent in the definition of each data manual and calibrating the levee error, which was considered a major source of incompatibility. The robustness of the recurrent U-Net in early predictions dramatically increased upon CBT and TIP, recording an F1 score of over 0.75 on July 10, when the on-site survey was performed; meanwhile, the best performance score was 0.81 at the end of the growing period. SR enhanced the spatial details of rice mapping near the levee area, which had an estimated width of 60 cm; however, the area was more similar to that in existing datasets when it was calibrated with the predicted probability of the levee ratio rather than SR. The calibration was scalable from the patch to city level, with the paddy area at both levels recording high R² for the farm map and statistics (0.99 for both the farm map and statistics at the city level, and 0.93 and 0.95, respectively, at the patch level). This study shows that remote-sensing-based paddy rice mapping can produce not only accurate but also timely and compatible predictions by integrating deep learning applications. The results show that the predictions are compatible with domestic datasets as much as they are with each other; therefore, remote-sensing approaches are expected to be more actively and practically integrated into agricultural decision-making processes.

Although the importance of developing reliable and systematic GHG inventory has increased, the GIS/RS-based national scale LULUCF (Land Use, Land-Use Change and Forestry) sector analysis is insufficient in the context of the Paris Agreement. In this study, the change in CO₂ storage of forest land due to land use change is estimated using two GIS/RS methodologies, Sampling and Wall‐to‐Wall methods, from 2000 to 2010. Particularly, various imagery with sampling data and land cover maps are used for Sampling and Wall-to-Wall methods, respectively. This land use matrix of these methodologies and the national cadastral statistics are classified by six land‐use categories (Forest land, Cropland, Grassland, Wetlands, Settlements, and Other land). The difference of area between the result of Sampling methods and the cadastral statistics decreases as the sample plot distance decreases. However, the difference is not significant under a 2 km sample plot. In the 2000s, the Wall-to-Wall method showed similar results to sampling under a 2 km distance except for the Settlement category. With the Wall-to-Wall method, CO₂ storage is higher than that of the Sampling method. Accordingly, the Wall-to-Wall method would be more advantageous than the Sampling method in the presence of sufficient spatial data for GHG inventory assessment. These results can contribute to establish an annual report system of national greenhouse gas inventory in the LULUCF sector.

Air pollution is one of the most significant environmental hazards. The elderly, young, and poor are more vulnerable to air pollution. The risk of air pollution was assessed based on the risk framework published by the Intergovernmental Panel on Climate Change (IPCC) in terms of three aspects: hazard, exposure, and vulnerability. This study determined the concentrations of hazardous pollutants using satellite images from 2015 at 1 km² spatial resolution. In addition, the study identified vulnerable groups who are exposed to hazardous air pollutants. The study highlighted the degree of vulnerability based on environmental sensitivity and institutional abilities, such as mitigation and social adaption policies, using statistical data. Based on the results, Seoul City and Gyeonggi Province have low air pollution risk owing to good institutional abilities, while the western coastal area has the highest air pollution risk. Three adaption pathway scenarios were assessed in terms of the effect of increases in the budget for social adaptation policies on the level of risk. The study found that the risk can be reduced when the social adaptation budget of 2015 base level is increased by 20% in Gyeonggi Province and by 30% in the western coastal area. In conclusion, this risk assessment can support policy-making to target more vulnerable groups based on scientific evidence and to ensure environmental justice at the national level.