Groundwater contaminant transport is an intricate process that involves the movement, distribution, and alteration of contaminants through subterranean formations. The procedure integrates environmental and hydro-chemical variables to develop models that have the potential to predict behavioral characteristics of diverse contaminants in the subsurface of water bodies. The goal of the study is to unravel the complex interactions and pathways of contaminants within the sub-surface water domain. As a result, we employed different regression models: Linear Regression (LR), Random Forest Regression (RFR), Extreme Gradient Boosting (XGB), Light Gradient Boosting Machine (LGBM), Support Vector Regression (SVR), Multi-layer Perceptron (MLP), and Boosted Decision Tree Regression (BDTR). The dataset is from earthdata: search.earthdata.nasa.gov/downloads/3912644044 arranged as 27 variables in 27 columns and 3081 rows. The result shows that RFR had an outstanding performance, with the lowest Root Mean Squared Error (RMSE) of 0.46, a Mean Absolute Error (MAE) of 0.20, an Explained Variance Score (EVS), and an R-squared (R2) value of 0.63 respectively. Next are LGBM and XGB both of which show excellent performance of 0.48 and 0.50 as RMSE with varying values for MAE, EVS, MSE, and R2 metrics. However, LR had a lower performance value of RMSE (0.57), 0.32 for both MAE and MSE, and 0.44 for EVS and R2 concurrently. BDTR are also valuable in groundwater contamination transport predictions with 0.50 as RMSE, 0.23 for MAE, and 0.56 for both EVS and R2. Therefore, this study not only presents groundwater contamination transport prediction modeling techniques but also provides scholars, hydrogeologists, and hydro-policymakers helpful recommendations on the models that present capabilities of selected models for predictions of groundwater contaminants.