This study advances environmental monitoring by predicting the spatial and temporal distribution of Harmful Algal Blooms (HABs) in the Republic of Korea through a hybrid approach that combines geostatistical and deep learning methods. Using 3D universal kriging, the study interpolates missing HAB concentration values, transforming geospatial point data into spatially continuous grid images that serve as the foundation for predictive modeling. These interpolated images are then used as input for a ConvLSTM (Convolutional Long Short-Term Memory) network, which integrates convolutional layers to capture spatial patterns and LSTM units to model temporal dependencies. By leveraging this spatiotemporal modeling framework, the ConvLSTM network effectively predicts future HAB concentrations with improved accuracy. This innovative methodology highlights the utility of combining 3D universal kriging for spatial interpolation with image-based ConvLSTM prediction, offering valuable insights into HAB dynamics and supporting sustainable strategies for environmental management and public health.

Fisheries and Aquatic Sciences

Traditional linear models in urban informatics struggle to capture the complex, non-linear spatial non-stationarity inherent in metropolitan housing markets. To overcome these constraints, this study introduces a data-driven computational framework integrating a Spatio-Temporal Graph Convolutional Network (STGCN) with gradient-based Explainable Artificial Intelligence (XAI) and Geographically Weighted Regression (GWR). This framework is empirically tested using 217,598 apartment transactions in Busan, the Republic of Korea, augmented with high-resolution micro-demographic grids and Digital Elevation Model (DEM) topographical data. Utilizing unsupervised K-Means clustering, the region is spatially stratified into a dense Urban Core and a dispersed Suburban Periphery. The STGCN demonstrates overwhelming predictive superiority ($R^2=0.802$) over the traditional Spatial Error Model ($R^2=0.437$). Crucially, gradient-based XAI and localized GWR coefficients successfully unspool the deep learning “black box,” visualizing hyper-localized economic realities that global linear models obscure. The analysis expose stark regional market segmentation driven by environmental topography, mathematically quantifying non-linear dynamics such as coastal high-floor premiums, severe mountainous altitude penalties, and latent urban reconstruction premiums. Ultimately, this research bridges the gap between predictive computational power and spatial economic interpretability, offering a robust informatics framework for equitable urban planning.

Timely assessment of macroeconomic conditions is essential because official gross domestic product (GDP) statistics are released with substantial delays and are often revised. This study examines whether high-frequency highway traffic volumes, disaggregated by vehicle type, improve short-term GDP nowcasting in the Republic of Korea. Using nationwide expressway traffic data from 328 toll plazas over the period from September 2008 to September 2025, we integrate traffic series with conventional macroeconomic indicators into a mixed-frequency dynamic factor model and evaluate pseudo-real-time nowcasting performance against official quarterly GDP releases. Time-series diagnostics indicate that traffic volumes contain short-horizon predictive information for GDP and satisfy stationarity requirements after appropriate transformation. In the full evaluation sample, the macro-only benchmark records an RMSE of 1.0258 and an MAE of 0.8716. Adding aggregated traffic changes these metrics only marginally (RMSE = 1.0269, MAE = 0.8696), whereas the model augmented with the heaviest freight class (Vehicle Type 6) performs best, lowering RMSE to 1.0179 and MAE to 0.8652. During the COVID-19 period, forecast accuracy deteriorates across specifications: aggregated traffic increases RMSE and MAE to 1.3456 and 1.2096 relative to the macro-only benchmark (RMSE = 1.3082, MAE = 1.2020), while Vehicle Type 6 lowers MAE to 1.1683 but still records a higher RMSE of 1.3198. These findings show that aggregate mobility measures add limited value, whereas freight-oriented vehicle-type disaggregation provides the most informative highway traffic signal for real-time GDP nowcasting.

Chlorophyll-a monitoring characterizes algal biomass but does not directly measure cyanotoxins or human exposure. We examined ecological spatial associations between chlorophyll-a and population-weighted estimates of Alzheimer’s disease (AD) and Parkinson’s disease (PD) patient counts in the Nakdong River basin, South Korea. Data for 2010–2022 combined chlorophyll-a from a nationwide 939-site monitoring network, older-adult populations, and administrative patient counts across 33,049 1 km grids. Geographically weighted regression described local associations. Higher chlorophyll-a occurred along river reaches and downstream areas; positive local coefficients were strongest in selected inland districts, including Sangju-si and Jinju-si, and weaker in metropolitan areas. Administrative sensitivity analyses showed that association direction and magnitude depended on population adjustment and model specification. Supplementary bandwidth analyses also identified scale dependence and persistent residual spatial autocorrelation. Because the grid outcomes were constructed from administrative counts, interpolation accuracy was unquantified, and source-specific clinical definitions were unverified; these findings indicate exploratory ecological covariation rather than a toxicological effect or causal risk. The mapped contrasts may guide targeted water-quality and cyanotoxin monitoring and epidemiological validation.