With the adjustment of industrial structure, many high pollution enterprises will gradually shut down. This process will produce a large number of high-risk pollution plots and cause a series of environmental problems. In this study, geophysical detection and direct push technology were applied to investigate soil and groundwater pollution in a typical brownfield in Zhenjiang, Jiangsu province, China. The results showed that Chlorotoluene and its ramification were the main pollutants for the brownfield. Pollution decreased with depth and was quite uneven in the brownfield. Production and wastewater treatment areas were heavily polluted, where volatile organic compounds contaminated and semi-volatile organic compounds were found 12 m and 15 m below the land surface, respectively. About 18880 m³ groundwater was contaminated. This study is significant to develop a geological survey work mode for the post-industrial era.

Due to large-scale geological deposition processes, slope structures are often stratified, which means that the spatial distribution of the parameters involved in slope reliability evaluation is statistically anisotropic. This paper studies the effect of the statistical anisotropy of undrained shear strength on the probability of slope failure (p_f) based on the Monte Carlo simulation. The results show that for the horizontally layered slope, the larger the horizontal correlation scale of undrained shear strength (λ_x) is, the larger p_f is, especially when λ_x is smaller than the slope length; for the vertically layered slope, the larger the vertical correlation scale (λ_y) is, the smaller p_f is, especially when λ_y is smaller than the slope height. Additionally, the mechanism of the above results is discussed by analyzing the displacement distribution at different correlation scales. The findings indicate that in the reliability evaluation of undrained slopes in stratified structures, either underestimation of λ_x or overestimation of λ_y leads to an unconservative estimate of p_f, resulting in an overestimation of the slope stability.

Coastal engineering construction suitability evaluation methods are too empirical and difficult to quantify. Considering these weaknesses, in order to determine the weight of each factor reasonably, and to analyze the suitability of coastal zone construction comprehensively, the theory of establishing a coastal zone construction suitability evaluation model based on a Rough Set (RS) and an Analytic Hierarchy Process (AHP) is proposed. In total, 20 typical coastal areas of Putian are selected, and the main impact factors are determined according to a port dock, pollution-prone industry, and an electric power plant. The contribution rate and weight of each factor for the construction of a coastal zone are analyzed by the combination evaluation model, and the final evaluation result is consistent with the actual investigation situation. Finally, 52 evaluation units of the Putian coastal zone are evaluated by Neural Networks (NNs). The weight of the impact factors is made more objective by using the training sample set of the combination evaluation model as the sample set of the neural network. The learning speed and accuracy of the network are improved, and the and the evaluation result is consistent with the actual investigation situation. In a word, it is effective to perform the suitability evaluation of the coastal zone construction using the RS-AHP-NN proposed model, and it can be applied in practical engineering.

Accurate multi-step runoff forecasting over China is important for flood control, water-resource management, and regional hydrological assessment. However, existing data-driven methods often struggle to jointly capture temporal variations at different time scales and the directional hydrological dependencies imposed by river networks, which limits forecasting accuracy and spatial structural consistency. To address this issue, this paper proposes a spatiotemporal forecasting framework that combines cross-scale temporal fusion with basin-topology-guided spatial modeling. Specifically, a multi-scale temporal module with cross-scale gating is introduced to adaptively integrate short-, medium-, and long-term runoff variations, while a basin-topology attention module incorporates upstream–downstream connectivity into spatial dependency learning. Experiments are conducted on a China-scale gridded runoff forecasting benchmark derived from the publicly available GloFAS Historical dataset through spatial filtering, valid-region masking, and forecasting-oriented sample construction. The proposed method achieves better overall performance than representative baselines in terms of MAE, RMSE, PSNR, and SSIM. In the overall comparison, it reaches MAE 0.0269 and RMSE 0.0603 in normalized log-scale runoff units, PSNR 24.39, and SSIM 0.9273, while maintaining a moderate parameter size and practical inference efficiency. The results demonstrate that the proposed framework reduces numerical errors and better preserves the spatial patterns of runoff fields. Ablation studies further confirm that both cross-scale gating and basin-topology attention contribute consistently to the overall improvement.