At present, the monitoring network of China cannot provide sufficient data to estimate land-based pollutant loads that enter the sea, and estimation methods are imprecisely used. In this study, the selection of monitoring stations, monitoring frequency, and pollutant load estimation methods was studied in Qingdao City, a typical coastal city in China, taken as an example. Land-based pollutant loads from Qingdao were estimated, and load distribution, density, and composition were analyzed to identify the key pollution source regions (SRs) that need to be monitored and controlled. Results show that the administrative land area of Qingdao can be divided into 25 sea-sink source regions (SSRs). A total of 14 more rivers and 62 industrial enterprises should be monitored to determine the comprehensive pollutant loads of the city. Furthermore, the monitoring frequency of rivers should not be less than three times/year; a monitoring frequency of five or more times is preferable. The findings on pollutant load estimation with the use of different estimation methods substantially vary; estimation results with the use of ratio-based methods were 10 and 22 % higher than those with the use of monitoring-based methods in terms of chemical oxygen demand (COD) and total nitrogen (TN), respectively. None-point sources contributed the majority of the pollutant loads at about 70 % of the total COD and 60 % of the total TN.