According to the daily precipitation data of 106 meteorological stations in East China and NCEP reanalysis data from 1961 to 2016, the atmospheric circulation patterns that cause the summer precipitation anomaly in Shandong Peninsula and their correlations with the underlying surface factors (SST and soil moisture) are analyzed. The results are as follows. 1) When anomalous southwesterly winds occur over the Bay of Bengal, anomalous anticyclonic circulations occur over the south of Japanese Islands, and anomalous cyclonic circulations occur over the southwest of Lake Baikal, the transfer of water vapor to Shandong Peninsula is intensified, and coordinated with anomalous atmospheric ascending movement, the summer precipitation in Shandong Peninsula is more. On the contrary, with the accurance of anomalous northwesterly winds over the Bay of Bengal, anomalous cyclonic circulations over the south of Japanese Islands, and anomalous anticyclonic circulations over the southwest of Lake Baikal, Shandong Peninsula gets less precipitation in summer. 2) A significant positive correlation between geopotential height of the whole troposphere and underlying SST in the critical regions of the Bay of Bengal and central Pacific can persist from spring to summer, and when the geopotential height anomaly of the whole troposphere in the two regions is in the positive phase, the strong westerly wind over the Bay of Bengal and the anticyclone circulation over the south of Japanese Islands occur in summer, respectively. 3) The anomalous sensible heat and latent heat flux caused by the soil moisture anomaly could explain the geopotential height anomaly in the critical regions of Lake Baikal and local convection anomaly in Shandong Peninsula. Higher soil moisture in the southwest of Lake Baikal corresponds to negative anomaly of geopotential height in the upper troposphere and higher soil moisture in Shandong Peninsula corresponds to anomalous ascending movement in the upper troposphere. 4) Using the underlying surface factors (SST and soil moisture) in spring in critical regions, a multivariate statistical regress-forecasting model for summer precipitation is established in Shandong Peninsula, whose anomaly coincidence rate is about 75% by leave-one-out cross validation. These results can provide important reference for summer precipitation prediction in Shandong Peninsula.