Abstract: The paper use the lunar DEM with 20m resolution as the data source, 16 typical lunar areas in different latitudes were selected. Then, we build a parallel computing framework to conduct numerical simulations of PSD and SR on each experimental sample area in 2020. Then, utilizing the slope spectrum model on digital terrain analysis, we use the rectangular window expansion method to define the PSD spectrum and SR spectrum. On this basis, the slope-mean PSD spectrum in a year, slope-mean SR spectrum in a year and aspect-mean SR spectrum in a year were derived. Meanwhile, the stable area values of different spectrums were extracted. The results show that: It can be seen that the spatial distribution of PSD and SR is significantly affected by terrain ; On the whole, with the increase of latitude, the solar altitude angle decreases, the terrain shielding effect becomes more obvious, resulting in the decrease of PSD spectrum and SR spectrum; The stable area value of the slope-mean possible sunshine duration spectrum on sample areas is positively correlated with the variation coefficient of slope. The result can be employed as a import support for the construction of the lunar base in the future.