Antibiotics in wastewater have become a growing problem in urban and peri-urban areas in developing countries as a result of increased use and misuse of antibiotics. A simple dynamic model, that describes the most important removal processes of antibiotic from the wastewater stabilization pond system (WSP) “Mafisa” in Morogoro, Tanzania, was developed using STELLA® software package. The model was based on liquid chromatography tandem mass spectrometry (LCMS/MS) analysis of trimethoprim, in water collected in the WSP. Concentrations of trimethoprim measured in the dry season and the rainy season were used in development of the model. To determine the model's applicability to simulate the removal of trimethoprim, a calibration was performed using concentrations from the dry season and a validation was performed using concentrations from the rainy season. To test the model's capacity to simulate the removal of other antibiotics than trimethoprim, a second validation was performed for three other antibiotics; metronidazole, sulfamethoxazole and ciprofloxacin. A two-tailed t-test with a confidence interval of 95% showed no significant difference (P = 0.7819) between the values given by the model (CSIM) and the values measured by LCMS/MS (COBS) of the first validation, and the standard deviation (SD) between the differences was 1%. The second validation gave a mean SD = 18% (found by a two-tailed t-test with a confidence interval of 95%) of the differences between CSIM and COBS. The model was developed under the assumption that settling, biodegradation, hydrolysis and photolysis were the only removal processes other than outlet. The major removal processes for trimethoprim and sulfamethoxazole were through settling and the outlet. Ciprofloxacin was removed by settling in the first pond. Metronidazole was mainly removed through the outlet, but settling and hydrolysis/photolysis also played a role. A sensitivity analysis (±10%) showed that the soil adsorption coefficient, the amount of suspended matter and the ratio of flow rate and volume were the most sensitive parameters. To strengthen the model, the exact removal processes should be further analysed. To apply the model on other WSP, a calibration of the settling rate constant and the amount of suspended matter should be performed.