Objective: To simulate changes in land use and food production in the Loess Plateau for past (2000-2015) and future (2015-2050) periods under four coupled climate-socio-economic scenarios. Methods: An integrated simulation and assessment method based on multi-source data was developed, whereby the direct and indirect impacts of urbanisation and cropland changes on forests and grasslands were assessed.Results: (1) An integrated simulation model of land use and land cover (LULC) based on system dynamics, lo-gistics equations, and cellular automata for simulating changes in spatiotemporal landscape patterns in multiple scenarios was developed. The simulation results indicated that urban land use will increase significantly, cropland and grassland will decrease; in addition, there was a clear trade-off between urban land and ecological land in some regions. (2) Using regression analysis to predict future changes in food production under multiple scenarios, an upward trend in yield per unit area and total grain output was observed. The urban-cropland changes from 2015 to 2050 will have significant indirect impacts on forest and grassland landscapes.Conclusion: The results of the present study provide insights into management and trade-off strategies for regulating the coupling between land use and key ecosystem services.