The Chagan Lake Catchment is located in the midwest of Songnen Plain, which is a typical high fluoride groundwater area. High fluoride water has an important impact on the economic development and ecosystem stability of Chagan Lake. In this study, the spatial distribution characteristics and influencing factors of fluorine in Chagan Lake Catchment are discussed by using hydrochemistry and mathematical statistical analysis. The groundwater in the study area was characterized as Na⁺-rich and Ca²⁺-poor, with a high pH value and high HCO₃^– content. The average concentration of F^– was 3.02 mg/L, which was the highest in Qian’an County. The dissolution of fluorine-containing minerals and the desorption of F^– in soil provided the source of F^– in groundwater, while calcite and dolomite precipitation, cation exchange, and evaporation concentration provided favorable conditions for F^– dissolving, migration, and enrichment in water. In addition, the concentration of F^– in surface water was 4.56 mg/L, and the highest concentration was found in Hongzi Pool and Hua’ao Pool. The elevated concentrations of F^– in both surface water and groundwater in the study were affected by human factors, such as rice planting and water conservancy project construction.

Soil samples from T (0~20 cm) and S (20~40 cm) layers of four saline–alkali rice fields (R5, R15, R20, and R35) with different reclamation years were selected to study the distribution of soil aggregates and the contents of readily oxidizable organic carbon (ROC), dissolved organic carbon (DOC), microbial biomass carbon (MBC), potentially mineralizable carbon (PMC), and soil organic carbon (SOC). The effects of large macroaggregate (>2 mm, LMA), small macroaggregate (0.25 to 2 mm, SMA), and microaggregate ( LMA > MA. An effective way to increase carbon sink and improve the ecological environment in western Jilin Province is to change the soil environment by planting rice in saline–alkali land.

Urban building energy models (UBEMs) are key tools for quantifying citywide building energy demand and on-site renewable supply, yet most studies focus on a single city and rely on locally available data under heterogeneous modelling standards, leaving no consistent basis for cross-climate comparison. This study proposes an integrated UBEM framework built on globally accessible open data. A nested pipeline couples microclimate correction with building energy demand and rooftop photovoltaic (PV) potential simulation at the block scale, and is applied to Melbourne, New York, Hong Kong, Helsinki and Dubai to generate a comparable dataset of 1,358 blocks. Across the five cities, the block-level Spearman correlation between energy use intensity (EUI) and PV generation per unit roof area is consistently strong and positive, while the Jaccard overlap between the top 10% hotspots of EUI and PV ranges from 0.23 to 0.65, indicating that local supply and demand can be spatially mismatched at the block level. Three tree-based surrogate models are benchmarked, with XGBoost reaching an R2 of 0.897 for EUI and 0.871 for PV. A global sensitivity analysis based on the Hilbert-Schmidt independence criterion (HSIC) shows that urban morphology dominates both targets in all five cities, while cluster-level rankings reveal within-city heterogeneity. Shapley additive explanations (SHAP) provide directional and per-block attribution; their partial disagreement with HSIC on the EUI ranking highlights a caveat of surrogate-based interpretation. The framework offers a consistent parameter system and an interpretable surrogate for cross-climate UBEM analysis, supporting early screening of demand-supply matching in urban energy planning.