A reliable estimate of the gross primary productivity (GPP) is crucial for understanding the global carbon balance and accurately assessing the ability of terrestrial ecosystems to support the sustainable development of human society. However, there are inconsistencies in variations and trends in current GPP products. To improve the estimation accuracy of GPP, a deep learning method has been adopted to merge 23 CMIP6 data to generate a monthly GPP merged product with high precision and a spatial resolution of 0.25°, covering a time range of 1850–2100 under four climate scenarios. Multi-model ensemble mean and the merged GPP (CMIP6_DL GPP) have been compared, taking GLASS GPP as the benchmark. Compared with the multi-model ensemble mean, the coefficient of determination between CMIP6_DL GPP and GLASS GPP was increased from 0.66 to 0.86, with the RMSD being reduced from 1.77 gCm⁻²d⁻¹ to 0.77 gCm⁻²d⁻¹, which significantly reduced the random error. Merged GPP can better capture long-term trends, especially in regions with dense vegetation along the southeast coast. Under the climate change scenarios, the regional average annual GPP shows an upward trend over China, and the variation trend intensifies with the increase in radiation forcing levels. The results contribute to a scientific understanding of the potential impact of climate change on GPP in China.

Accurate and timely monitoring of soil salinity is essential for the sustainable management and remediation of coastal salinization. This study utilized a UAV-based remote sensing platform to collect multispectral imagery and concurrent in situ soil salinity samples from an experimental zone within the Yellow River Delta National Nature Reserve in July 2024. We constructed multiple spectral indices and employed advanced feature selection methods—namely VIP, MultiSURF, and PSO-SFLA—to identify the most informative index combination. We established a soil salinity retrieval model utilizing a stacking ensemble framework. This architecture integrated TabPFN, SVM, and Ridge regression as the base learners, while employing XGBoost as the meta-learner to synthesize the final predictions. Model interpretability was assessed using SHAP (SHapley Additive explanations) values, while predictive performance was evaluated using the coefficient of determination (R²), Standardized Root Mean Square Error (S_RMSE), and the Ratio of Performance to Deviation (RPD). Results indicate that the stacking model, when coupled with PSO-SFLA for feature selection, outperformed all other model configurations. It achieved the highest prediction accuracy on the test set, with an R² of 0.754, S_RMSE of 0.310, and RPD of 1.941. The resulting soil salinity distribution map exhibited a high degree of spatial agreement with the ground-truth survey data. This study demonstrates that leveraging a stacking algorithm with UAV multispectral data provides an accurate and reliable method for monitoring soil salinity in coastal wetlands, offering valuable technical support for effective soil salinization management.

To enhance the meteorological forecasting and early warning service capability for urban waterlogging risks in Jinan, this study aims to investigate the relationship between rainfall and urban waterlogging. Based on minute-scale precipitation observations from 38 automatic weather stations and records from 70 waterlogging monitoring sites in the urban area of Jinan from 2011 to 2024, this study systematically analyzes the spatiotemporal characteristics of precipitation and waterlogging events and quantifies their response relationship. The main findings are summarized as follows. Heavy precipitation and waterlogging events are strongly temporally coincident, primarily occurring during the main flood season from June to August. Regarding diurnal variation, short-duration heavy rainfall and waterlogging events are concentrated between 14:00 and 20:00. The water depth of most waterlogging events ranges from 0.11 m to 1.04 m, with a median of 0.26 m, and the distribution of waterlogging exhibits a pronounced right-skewed pattern. A moderate positive spatial autocorrelation was observed in waterlogging depth, suggesting that severe urban waterlogging events are more likely to occur in the northern region of Jinan. The precipitation preceding waterlogging events is predominantly short-duration heavy rainfall. A strong temporal relationship exists between peak precipitation and maximum waterlogging depth. In nearly 90% of the waterlogging events, peak precipitation occurs within 2 h before the maximum waterlogging depth, with an average lead time of approximately 55 min. The relationship between antecedent cumulative precipitation and peak waterlogging depth is strongest at the 120 min timescale. About 90% of maximum rainfall over 10 min, 1 h, and 2 h did not exceed the 1-year return period threshold, indicating that the precipitation causing waterlogging events in Jinan is generally non-extreme.