The minimum ecological land is the key ecological land in a region, which can ensure that the regional ecosystems provide and maintain the integrity of important ecosystem services. Under the background of rapid urbanization, ecological land has been seriously occupied and destroyed, resulting in the decline of ecosystem services and even affecting the ecological security of the whole region. Therefore, quantifying the minimum ecological land is of great significance for maintaining the health and sustainable development of regional ecosystem. Taking Dianchi Lake Basin as an example, this paper firstly analyzed the connotation of the minimum ecological land. Secondly, from the perspective of ecosystem services, a research framework of the minimum ecological land quantification was proposed based on minimum cumulative resistance (MCR) model and scenario simulation method. Finally, landscape pattern index method was used to compare the quantitative results of the minimum ecological land of these two methods. The results are as follows: (1) Based on MCR model, the number of grids of the minimum ecological land in the study area is 1275560, with an area of 1148.00km², accounting for 50.68% of the ecological land area and 39.32% of the total area of the basin. (2) Based on scenario simulation method, four scenarios accounting for 30%, 40%, 50% and 60% of the total ecological land area in the study area are compared, and the minimum ecological land area is identified as 1129.55km², covering for 50% of the ecological land area and 38.96% of the total area of the basin. (3) Further comparative analysis reveals that the distribution characteristics of land use types of the minimum ecological land identified by scenario simulation method show higher stability and rationality than by MCR model. At the watershed scale, the land use type with the highest proportion of the minimum ecological land determined by MCR model and scenario simulation method are both forestland, which are 78.17% and 47.32%, respectively. In terms of landscape pattern characteristics, the largest patch index, contagion index and aggregation index of the minimum ecological land identified by scenario simulation method are all higher than those determined by MCR model, with values of 10.62, 39.57 and 93.39, respectively. This indicates that, compared with MCR model, the minimum ecological land determined by scenario simulation method has better dominant patch connectivity, higher landscape agglomeration degree, and smaller landscape fragmentation degree. Accordingly, it is necessary to comprehensively play the advantages of these two methods in the process of the minimum ecological land quantification.