Different combinations of low impact development (LID) technologies can be applied for control of urban non-point source pollution. There are currently few evaluations of urban non-point source pollution and pollution load reduction based on a combination of porous asphalt and bio-retention. Taking Shenzhen International Low Carbon City as an example, road-deposited sediments were collected prior to and after rainfall events. Runoff was monitored under six typical rainfall events, from porous asphalt and the inlet/outlet of bio-retention. Through analysis of changes in the process of "build-up-wash-off-transport" of pollutant loads, the average build-up of road-deposited sediments in the study area was found to be (15.80±3.79) g·m-2; the mass percentage of road-deposited sediments (size>250 µm) was approximately 65.14%. The average wash-off percentage of six different intensity rainfall events was 17.15%, and road-deposited sediments (size<105 µm) were carried by 62.71%-74.94%. The average pollution loads of surface runoff pollutants SS, TN, and TP were 2.02, 0.025, and 0.0013 g·m-2, respectively. The removal rates of SS, TN, and TP through porous asphalt under infiltration and filtration were 70.26%, 46.29%, and 19.27%, respectively. The secondary purification removal rates of runoff water in bio-retention were 85.25%, 20.22%, and 70.27%, respectively. Pollutant loads into Dingshan River totaled 0.08, 0.011, and 0.0003 g·m-2, representing 4.05%, 43.47%, and 24.39% of runoff. The combination thus had a significant effect on runoff purification. Through quantitative research on the formation of non-point source pollution, this paper provides a scientific basis for estimating pollution loads of urban non-point source pollution and evaluating the performance of LID projects. It makes suggestions for the popularization and application of LID and sponge city design in China.