Groundwater salinization is a major challenge in coastal aquifers of arid and semi-arid settings, where limited supply of water and extreme environmental conditions are key constraints on groundwater usage. Hydrochemical data from groundwater wells in the Al-Qatif coastal aquifer have been assessed by a copula-driven scheme to measure the joint salinity extremes and their spatial risk. Electrical Conductivity (EC) ranged from about 3000 to 40,000 mu S cm- 1 and salinity was greater than 35,000 mg L- 1 in the acutely affected areas. Lognormal marginal distributions showed closer fits to salinity indicators than Gamma models, with differences in Akaike Information Criterion (AIC) exceeding ten units. Upper-tail dependence analysis indicated that there was nearly perfect co-occurrence among EC, chloride, sodium, sulfate, and calcium, with lambda(u) values getting closer to unity at high quantiles while magnesium had weaker tail coupling at 0.77. Maximum probabilities of joint exceedance probability surfaces approached 0.27 at moderate limits and decayed quickly at extreme levels, in keeping with heavy-tailed dependence described by Student's t copulas. The relationships were incorporated into an Extreme Salinity Risk Index (ESRI), generating spatial risk maps capturing localized hotspots where the ESRI value was greater than 0.8 and uncertainty was less than 0.15. Results indicate that extreme salinity is driven by coordinated upper-tail behavior across major ions, not by one-shot parameter exceedance. Contrary to conventional correlation or facies-based methods that show average conditions, the proposed framework provides quantities of compound extremes and their spatial probability, hence creating a structure for monitoring and management under worst-case salinity scenarios.