Abstract: In order to explore the food web structure of Lake Shankou, an alpine lake in northeast China, and then predict a more reasonable ecological management mode for this ecosystem, the effects of physicochemical factors on primary production (PP) of Lake Shankou were identified through multivariate stepwise regression analysis. Ecosystem data of Lake Shankou from 2014 were used for Ecopath modeling, and the Ecosim model was applied to predict the succession trend of plankton and major fish species in the next 20 years (since 2014) under different scenarios. The future management modes of Lake Shankou were also explored based on correlation analysis. The results showed that Lake Shankou was a phosphorus-limited lake, and the PP of this ecosystem was positively correlated with water temperature and total phosphorus. Complex predator-prey relationships were observed in Lake Shankou ecosystem. Most energy flows of this ecosystem were concentrated on trophic level II or above, and the keystone species was the "other fish" function group. The results of the Ecopath modeling were highly reliable according to the Pedigree index (0.537). The increase in production rate (Production/Biomass, PD/B) of phytoplankton significantly promoted the relative biomass increase of crucian carp (Carassius auratus), common carp (Cyprinus carpio), and silver carp (Hypophthalmichthys molitrix). However, this promotion effect was not significant when PD/B of phytoplankton decreased by more than 5% annually. The increasing search rate of Hypophthalmichthys molitrix would enhance its predatory efficiency to zooplankton (i.e. copepod, cladocera, and rotifer), and result in a slight increase in the relative biomass of phytoplankton via trophic cascade effects. Combined with scenario analysis and correlation analysis, the elevation of phytoplankton biomass would increase fishery production, and silver carp control for controlling algae was not effective in Lake Shankou. The controlling of exogenous nutrient input was more important in future ecosystem management.