KJCKorea
Journal Central

The Climate Change Impact Assessment system is a key instrument for achieving carbon neutrality; however, its current application is limited to some plans and projects subject to Environmental Impact Assessment. This study aims to provide quantitative evidence for expanding the scope of projects subject to Climate Change Impact Assessment. To this end, 203 Environmental Impact Assessment reports were collected for seven project types that are not currently subject to Climate Change Impact Assessment. Greenhouse gas emissions were analyzed using three evaluation criteria: total GHG emissions during construction, annual GHG emissions during operation, and the previous-study criterion based on stationary combustion and electricity use during operation. Carbon storage and sequestration losses were analyzed separately as supplementary carbon sink indicators. The results indicated that operational emissions from development projects in specific areas and harbor construction projects exceeded the current criteria for Climate Change Impact Assessment, indicating the need to prioritize these project types for consideration in future expansion of the assessment scope. In railroad construction projects, greenhouse gas emissions during the construction phase were identified as a major environmental impact, whereas in tourism complex development projects and sports facility establishment projects, carbon sink losses resulting from forest loss were found to be major environmental impacts. This study empirically examined the characteristics of greenhouse gas emissions and carbon sink losses by project type, thereby revealing blind spots in the current Climate Change Impact Assessment system. Based on these findings, it presents policy implications for expanding the scope of Climate Change Impact Assessment by comprehensively considering GHG emissions during the construction and operation phases and carbon sink conservation.

The high-rise and high-density development of apartment complexes creates complex microclimates. However, conventional landscape planting has often failed to accurately reflect these environmental characteristics, heavily relying on designers' intuition or simple compass directions. Therefore, this study aimed to quantitatively categorize spatial shading characteristics formed by buildings and to investigate the impact of differences in the light environment (light intensity and R:FR ratio) on the actual growth (diameter at root collar) of landscape trees. For this purpose, 36 standard plots across six apartment complexes were investigated. Using hemispherical photography, the sky obstruction ratio and the number of buildings in eight directions were calculated. Subsequently, the spaces were classified into Open, Partially shaded, and Completely shaded types through K-Means clustering analysis. Following this, the light environment and the annual root collar diameter growth of the planted trees were statistically compared across the spatial types. The results confirmed that unconditionally open spaces are not the optimal environments for tree growth. The overall tree growth was the highest in the partially shaded spaces (p<0.001). Furthermore, the light environment analysis revealed that although the partially shaded spaces had a lower total light intensity than the open spaces, they maintained the essential light quality (R:FR ratio) crucial for plant growth at the highest and most stable levels during the day. This study is significant as it quantitatively classifies spaces within apartment complexes and empirically proves that appropriate shading can actually foster a positive light environment for tree growth. These findings can serve as fundamental data for future landscape tree selection and the establishment of planting guidelines that accurately reflect spatial characteristics.

This study forecasts water quality at Paldang Lake and interprets the basis of its predictions to identify priority nonpoint source (NPS) areas in the watershed. Because rainfall-driven NPS pollution originates from widely dispersed sources, evidence is needed to allocate limited management resources. To this end, 63 monitoring stations across the Namhan River, Bukhan River, and Gyeongan Stream were represented as graph nodes, and Graph WaveNet was used to forecast BOD, COD, T-N, and T-P at the Paldang Lake station from 1 to 10 days ahead. SHapley Additive exPlanations (SHAP) was then applied to quantify variable- and station-wise contributions, which were compared according to whether rainfall occurred. For 1-day-ahead forecasts, the explanatory power (R²) was at least 0.89 for all variables (0.94 for T-N), equal to or higher than the LSTM, TCN, and Transformer baselines, and T-N maintained an R² of 0.37 even at the 10-day horizon. Conditional SHAP analysis showed that nutrient contributions from the remote Bokha Stream system in Icheon increased markedly under rainfall; for T-P, rainfall contributions exceeded dry-condition contributions in all 250 training runs, indicating a rainfall-sensitive NPS source. The same tendency appeared in the Gyeongan Stream watershed. These results can serve as a basis for prioritizing watershed-scale NPS reduction projects.