The spatial mapping and prediction of groundwater quality (GWQ) is important for sustainable groundwater management, but several research gaps remain unexplored, including the inaccuracy of spatial interpolation, limited consideration of the geological environment and human activity effects, limitation to specific pollutants, and unsystematic indicator selection. This study utilized the entropy-weighted water quality index (EWQI), the LightGBM model, the pressure-state-response (PSR) framework and SHapley Additive exPlanations (SHAP) analysis to address the above research gaps. The normalized importance (NI) shows that NO₃⁻ (0.208), Mg²⁺ (0.143), SO₄²⁻ (0.110), Cr⁶⁺ (0.109) and Na⁺ (0.095) should be prioritized as parameters for remediation, and the skewness EWQI distribution indicates that although most sampled locations have acceptable GWQ, a few areas suffer from severely poor GWQ. The PSR framework identifies 13 indicators from geological environments and human activities for the SMP of GWQ. Despite high AUROCs (0.9074, 0.8981, 0.8885, 0.9043) across four random training and testing sets, it was surprising that significant spatial uncertainty was observed, with Pearson correlation coefficients (PCCs) from 0.5365 to 0.8066. We addressed this issue by using the spatial-grid average probabilities of four maps. Additionally, population and nighttime light are key indicators, while net recharge, land use and cover (LULC), and the degree of urbanization have the lowest importance. SHAP analysis highlights both positive and negative impacts of human activities on GWQ, identifying point-source pollution as the main cause of the poor GWQ in the study area. Due to the limited research on this field, future studies should focus on six key aspects: multi-method GWQ assessment, quantitative relationships between indicators and GWQ, comparisons of various spatial mapping and prediction models, the application of the PSR framework for indicator selection, the development of methods to reduce spatial uncertainty, and the use of explainable machine learning techniques in groundwater management.

Based on the information from an actual petroleum-contaminated site, a one-dimensional soil column was used to examine the vertical transportation diversities of different petroleum components under the influence of water table fluctuations, and the results revealed the following: (1) There were two obvious pollution accumulation zones under the condition of water table fluctuations: (i) The pollution infiltration zone dominated by the residual phase was formed at the leakage position, and (ii) the floating zone dominated by the free phase was formed near the water table. Combined with the viscosity of the organic components, the concentrations of the residual phase were octanoic acid > hexadecane > cyclohexane > toluene. Compared to coarse sand, clay can adsorb more components. (2) Different fluctuation frequencies had a great influence on the migration process of components. The free phase can transfer into the residual phase during the low groundwater table fluctuation. In the case of high-frequency groundwater fluctuations, there were more free phase components that can be carried by the water flow. However, due to the continuous flushing by the water, pollutants were finally spread to the whole underground system. (3) A cost-effectiveness remediation strategy is based on the difference in pollutant transportation. Therefore, the conclusions in this paper are fully applied in the actual contaminated sites. Specifically, the air-sparing (AS) and soil vapor extraction (SVE) devices were installed in the vadose zone to remove volatile substances (such as toluene). The permeable reactive barriers (PRBs) were set in the groundwater fluctuation zone to repair the residual pollutants (alkanes, cyclane, and asphaltenes hydrocarbons) that are continuously converted from the residual phase to the dissolved phase and free phase due to water level fluctuations. Hence, the results of this study provided a specific, targeted, and comprehensive strategy for petroleum pollution treatment.

Underground engineering construction in the karst regions of Southwestern China has become a focal point of China’s advancing regional urban development. However, construction activities interfere with the karst groundwater environment, which is characterized by irregular pore distributions and complex, variable flow patterns. This study establishes a numerical model of the karst water system traversed by Line 2 of the Guiyang Rail Transit in China. Incorporating hydrogeological conditions and tunnel engineering parameters, the model simulates the effects of tunnel construction on the karst groundwater system. The flow-field distribution of the karst groundwater system is altered at various stages of tunnel construction. During tunnel excavation, a drainage zone centered around the subway forms in the groundwater system, altering the groundwater flow field and causing fluctuations in the groundwater level. During the lining phase, the tunnel area gradually transforms into a waterproof zone. Although the groundwater level gradually recovers under rainfall recharge, the waterproofing effect of the tunnel drives the formation of a new groundwater flow field within the groundwater system, changing both the groundwater level and the original flow field. This work offers support for the coordinated development of underground engineering and environmental protection in karst areas, facilitating sustainable urbanization.

Urban underground space (UUS) is a significant natural resource to support many aspects of city development, but it is not sustainably developed and utilized during the urbanization process. This study considered 11 conditional and two sensitive factors and combined analytic hierarchy process (AHP) and variable weight theory (VWT) for the suitability evaluation of UUS development and utilization (SEUUSD&U) by taking the Jining city planning zone (JNPZ) as a case study. The results show that mining subsidence and groundwater-related factors are critical factors, which align with the real conditions. A significant increase in the weight of shallow groundwater can be observed after applying the VWT, rising from 0.1586 to 0.2544. This may result from significant extreme values, which WVT accurately identified and therefore increased the weights. From shallow to deep UUS, both the most suitable and least suitable areas increase, rising from 32.91% to 68.20% and from 0.57% to 3.01%, respectively. Based on two sensitive factors (key urban development and ecological protection), the study area was divided into four management zones. These sensitive factors often exhibit a “barrel effect”, showing the power to either definitively affirm or veto the outcomes. More importantly, this study proposes a generalized SEUUSD&U framework comprising six key steps, with particular emphasis on three aspects: “local conditions”, “barrel effect integration”, and “adaptive management strategies aligned with the United Nations sustainable development goals (SDGs)”. We strongly recommend that this framework be highly promoted in future research and strongly encourage future studies to place greater emphasis on the ultimate goal of achieving the SDGs by 2030 during updates to models, variable weight functions, factors, and frameworks.

Urbanization always leads to increasing challenges to the groundwater resources in karst regions due to intensive land use, infrastructure development, and the rapid transmission potential of pollutants. This study proposed an improved groundwater vulnerability assessment (GVA) framework by modifying the widely used COP (Concentration of flow, Overlying layers, and Precipitation) model, through the integration of three additional indicators: urban underground rail transportation (UURT), land use and cover (LULC), and karst development (K). Guiyang, a typical urbanized karst city in southwest China, was selected as the case study. The improved COP model, namely the COPK model, showed stronger spatial differentiation and a higher Pearson correlation coefficient ($r$) with nitrate concentrations ($r$ = 0.4388) compared to the original COP model (R = 0.3689), which validates the effectiveness of the newly introduced indicators. However, both R values remained below 0.5, even after model modification, suggesting that intensive human activities play a role in influencing nitrate distribution. The pollution load index (PI) was developed based on seven types of pollution sources, and it was integrated with the COPK vulnerability index using a risk matrix approach, producing a groundwater risk map classified into five levels. Global Moran’s I analysis (0.9171 for COP model and 0.8739 for COPK model) confirmed strong and significant spatial clustering patterns for the two models. The inclusion of UURT and LULC improved the model’s sensitivity to urban-related pressures and enhanced its capacity to detect local risk zones. It is a scalable tool for groundwater risk assessment in urbanized karst areas and offers practical insights for land use planning and sustainable groundwater management.