Sürdürülebilir Kalkınma Amaçları
Yayınlar
- MakaleYalnız künye2026Very fast Kolmogorov-Arnold network for landslide susceptibility prediction
Günen, Mehmet Akif; Tona, Aziz Uğur
Landslides represent critical geological events causing loss of life and property. Prediction of landslide susceptibility is essential for risk reduction and decision-making. XGBoost, LightGBM, NGBoost, CatBoost, MLP, and FastKAN were used to predict landslide susceptibility in Gümüşhane, Türkiye, a region with complex topographic and geological structures. 20 factors were used: 9 topographical, 3 hydrological, 4 climatic, 2 geological, and 2 environmental and infrastructure-related factors. Model performance was assessed using Accuracy, Precision, Recall, and F-Score, with SHAP analysis showing feature contributions. A multicollinearity analysis was used to determine the final feature set. Topographical variables (Aspect, Slope) showed lower impacts, while geological and environmental factors (Lithology, Distance to Fault) demonstrated moderate influence across models. For FastKAN, DEM, precipitation, and distance to fault were key factors, while for boosting methods, GHI, Tmax, and DEM were most significant. FastKAN achieved the highest performance under the internal pixel-wise evaluation scheme, with an accuracy of 96.54%, followed by XGBoost at 95.56%. McNemar’s statistical test showed no significant difference between NGBoost and MLP. Post-inventory landslide events observed in 2025 and 2026 were spatially consistent with high and very high susceptibility zones, providing a qualitative consistency check rather than rigorous independent quantitative validation.
- MakaleAçık Erişim2026Assessment of Atmospheric Acidifying Pollutant Trends and Their Potential Impact on Aquatic Carbon Stability in a Semi-Arid Basin: The Case of Konya
Tona, Aziz Uğur; Demir, Vahdettin
The behavior of the carbon cycle within the Land-Ocean Aquatic Continuum (LOAC) is shaped not only by aquatic processes but also by chemical interactions occurring at the atmosphere–water interface. In particular, the transport of acid rain precursors such as SO2 and NOx to surface waters via deposition can alter the water’s pH balance, thereby affecting Dissolved Inorganic Carbon (DIC) fractions and CO2 emission potential. In this study, air quality measurements from three monitoring stations (Bosna, Karatay, and Meram) in Konya province of Türkiye, along with precipitation and temperature data from a representative meteorological station for the period 2021–2023, were analyzed using the Mann–Kendall Trend Test. Additionally, seasonal pH values of groundwater were examined, and their trends were compared with those of the other variables. The findings reveal striking differences on a station basis. At the Bosna station, while NO (Z = 10.80), NO2 (Z = 9.47), and NOx (Z = 10.04) showed strong increasing trends, O3 decreased significantly (Z = −15.14). At the Karatay station, significant increasing trends were detected for CO (Z = 10.01), PM10 (Z = 8.59), SO2 (Z = 5.55), and NOx (Z = 2.44), whereas O3 exhibited a negative trend (Z = −6.54). At the Meram station, a significant decrease was observed in CO (Z = −11.63), while NO2 showed an increasing trend (Z = 3.03). Analysis of meteorological series indicated no significant trend in precipitation (Z = −0.04), but a distinct increase in temperature (Z = 2.90, p < 0.01). These findings suggest that the increasing NOx load in the Konya atmosphere accelerates O3 consumption and, combined with rising temperatures, creates a potential for change in the carbon chemistry of aquatic systems. The results demonstrate that atmospheric pollutant trends constitute an indirect but significant pressure factor on the aquatic carbon cycle in semi-arid regions and highlight the necessity of integrating atmospheric processes into carbon budget analyses within the scope of LOAC.