Abstract: Saltwater intrusion and drought represent major hydrological threats to water supply security in coastal regions, particularly under the escalating impacts of climate change and increasing anthropogenic pressures on water resources. This study presents a comprehensive analysis of the co-occurrence risk of drought and saltwater intrusion in the Modaomen waterway, a critical freshwater channel and primary water source for the Pearl River Delta region. The analysis utilized monthly hydrological data spanning the period from 2005 to 2023. Drought severity was characterized using long-term streamflow records from the Makou hydrological station on the main stem of the Xijiang River, while monthly maximum salinity levels were obtained from five strategically located monitoring sites along the waterway: Denglongshan sluice, Lianshiwan sluice, Nanzhen water plant, Quanlu water plant, and Renyi water plant. A two-dimensional Archimedean Copula approach was employed to model the joint probability distribution of drought intensity and salinity extremes, providing a robust framework for assessing compound risk. Kendall’s rank correlation coefficient was applied to evaluate the dependence structure between the two variables. Parameters for three widely used Archimedean Copula functions—Gumbel, Clayton, and Frank—were estimated using the maximum likelihood estimation method. Model selection was performed based on rigorous statistical criteria, including RMSE and AIC, to identify the optimal Copula function for each monitoring site. The results demonstrate a statistically significant positive correlation between drought intensity at Makou station and monthly maximum salinity levels across all monitoring sites, with Kendall’s correlation coefficients consistently exceeding 0.500 0. The strongest correlation was observed at Lianshiwan sluice (0.715 5), while the weakest was identified at Renyi water plant (0.503 5). Spatial heterogeneity was evident in the selection of optimal Copula functions: the Frank Copula provided the best fit for Denglongshan sluice and Quanlu water plant, the Gumbel Copula was optimal for Nanzhen and Renyi water plants, and the Clayton Copula was most suitable for Lianshiwan sluice. The joint probability density functions for all sites exhibited distinct bimodal characteristics, indicating two predominant regimes of compound risk occurrence under both moderate and extreme hydrological conditions. Tail dependence analysis provided further insights into extreme-value behavior. The Gumbel Copula models, selected for Nanzhen and Renyi, exhibited strong upper-tail dependence, with coefficients of 0.755 7 and 0.656 9 respectively, underscoring a significantly elevated risk of concurrent extreme drought and high-salinity events. In contrast, the Clayton Copula model at Lianshiwan sluice showed pronounced lower-tail dependence (0.915 3), indicating a strong correlation during mild to moderate event conditions. The Frank Copula models displayed symmetric dependence with no significant tail preference, suggesting a more uniform risk distribution across different intensity ranges. A comparative analysis with traditional univariate extreme value theory, under the assumption of independence, revealed that the Copula approach captures substantially higher joint exceedance probabilities. For instance, at the 95th percentile threshold, the joint exceedance probability at Renyi water plant derived from the Gumbel Copula was 0.033 4—approximately 13 times greater than the value calculated under the independence assumption (0.002 5). Similarly, at the 99th percentile threshold, the Copula-based probability (0.006 6) was 66 times higher than the theoretical independent value (0.000 1), unequivocally demonstrating that conventional univariate methods severely underestimate compound hydroclimatic risks. The observed spatial variability in optimal Copula types and tail dependence structures highlights the profound influence of local physiographic and hydrological factors, including tidal dynamics, riverbed morphology, inflow from tributaries, and human interventions. These findings emphasize the critical necessity of adopting bivariate and multivariate modeling approaches for robust risk assessment of compound drought and saltwater intrusion events. The study provides valuable insights into the coupled mechanisms governing these complex hydrological hazards and offers a scientific basis for enhancing water resource management and adaptation strategies. The methodology and results presented herein can serve as a valuable reference for water managers, policymakers, and researchers working in coastal cities and estuarine regions worldwide facing similar challenges, ultimately supporting the development of more resilient and sustainable water supply systems under changing environmental conditions and increasing climate variability.