The northern Qaidam Basin has abundant coal and coalbed methane (CBM) resources, and quantitative evaluation of adsorption pore characteristics has great significance for optimum selection of CBM-favorable areas. Based on vitrinite reflectance, coal maceral, proximate analysis, low-temperature N 2 adsorption, and methane isothermal adsorption experiments, the heterogeneities of adsorption pores (pore diameter <100 nm) were quantitatively characterized, and relationships between fractal dimensions and physical parameters of low-ranked coal reservoirs were revealed. The results show that the micropore volume percentage ranges between 33.70 and 86.44% with an average of 64.94%. Based on N 2 adsorption/desorption curves and pore diameter distribution characteristics, the adsorption pore structures of low-ranked coals were divided into 3 types. According to the FHH model, fractal dimension D 1 (relative pressure between 0 and 0.5) and D 2 (relative pressure between 0.5 and 1) were calculated. Fractal dimension D 1 , representing adsorption pore surface area, ranges from 2.001 to 2.345, with lower values. Fractal dimension D 2 (adsorption pore structure) is from 2.641 to 2.917, with relatively high values, which has a decreasing tendency from west to east in the study area. There are positive relationships between fractal dimension D 1 and Langmuir volume and specific surface area, whereas negative correlations are found between fractal dimension D 2 and Langmuir pressure, ash yield, moisture content, volatile content, and average pore diameter. Combined with the related data for middle-and high-rank coals, the characteristics of pore surface and methane adsorption capacity can be analyzed based on the variation of vitrinite reflectance. Furthermore, the complexity of pore structure can also be predicted according to the averaged pore size and micropore content to some degree.