With the rapid development of industrialization and urbanization, air pollution is becoming increasingly serious. Accurate prediction of PM_2.5 concentration is of great significance to environmental protection and public health. Our study takes Nanning urban area, which has unique geographical, climatic and pollution source characteristics, as the object. Based on the dual-time resolution raster data of the China High-resolution and High-quality PM_2.5 Dataset (CHAP) from 2012 to 2023, the PM_2.5 concentration prediction study is carried out using SARIMA, Prophet and LightGBM models. The study systematically compares the performance of each model from the spatial and temporal dimensions using indicators such as mean square error (MSE), mean absolute error (MAE) and coefficient of determination (R²). The results show that the LightGBM model has a strong ability to mine complex nonlinear relationships, but its stability is poor. The Prophet model has obvious advantages in dealing with seasonality and trend of time series, but it lacks adaptability to complex changes. The SARIMA model is based on time series prediction theory and performs well in some scenarios, but has limitations in dealing with non-stationary data and spatial heterogeneity. Our research provides a multi-dimensional model performance reference for subsequent PM_2.5 concentration predictions, helps researchers select models reasonably according to different scenarios and needs, provides new ideas for analyzing concentration change patterns, and promotes the development of related research in the field of environmental science.

Against the background of the rapid transformation of traditional economies and societies and continuous global climate change, how to ensure the long-term stability of the coastal ecological environment has become a key issue to be studied. In this paper, we take the 20 km buffer zone extending inland from the South African coastal zone as the study area. By constructing a vegetation vulnerability evaluation system, the current and future scenarios are compared in depth based on the base period (2010–2020), the near term (2030–2059), and the long term (2070–2099) with the help of GIS spatial analysis, the Moran index, and other methods. The results show that there are obvious spatial differences in vegetation vulnerability in the South African coastal zone. The extremely vulnerable areas of vegetation are mostly distributed on the west coast of South Africa, and some areas have obvious high–high aggregation patterns. The transfer of SSP1-2.6 scenarios in the near term is relatively stable, and the vegetation vulnerability level rebounds significantly in the long term; the vulnerability level of SSP2-4.5 scenarios has increased in both the near term and the long term, indicating that the risk of vegetation vulnerability has increased; while the SSP5-8.5 scenario has a significant deterioration trend in the long term, and the risk of vegetation vulnerability shifting to a high vulnerability level has increased significantly. Land use type has a significant impact on the response of vegetation vulnerability to SSP prediction. In the process of transformation from the base period to the long term, the proportion of vegetation vulnerability shifting to extremely vulnerable and severely vulnerable levels is notably high for both cultivated land and forest land—particularly under high-emission scenarios, driven by agricultural intensification for cultivated land and climate stress for forest land. This paper deeply explores the spatiotemporal evolution law and driving mechanism of vegetation vulnerability in the South African coastal zone under different shared socioeconomic pathway (SSP) scenarios, providing decision support for better development and protection of the South African coastal zone in the future.