This study applies a LOADEST model to hydrology and water quality data from 2008 to 2017 to examine the variation of pollutant loads in Shenzhen River estuary. These investigated pollutants include COD, NH3-N, TN and TP. The result from the LOADEST model, alone with situation of both precipitation and wastewater treatment plants, are used to analyze both natural and artificial impacts on the pollutants. The research concludes the following statements. 1) The net pollutant loads in Shenzhen River decreased significantly from 2008 to 2017, but still not enough to reach level V for river water quality standard (GB3838–2002). 2) The seasonal variations of net pollutant loads reduced from 2008 to 2017. 3) There is a strong correlation between precipitation and COD net load, which indicates that this pollutant is mainly originated from non-point sources. Precipitation and NH3-N, TN and TP net loads have weak correlations, which means that point sources are the main contributors of these pollutants. 4) There is a negative correlation between net pollutant loads and the amount of treated wastewater. This illustrates the enhanced capacity of waste water treatment plants, as well as the improved quality of discharged treated water will contribute to better water quality in Shenzhen River.