This study elucidates Y chromosome distribution patterns in the three general provincial populations of historical Tibet, Amdo (n = 88), Dotoe (n = 109) and U-Tsang (n = 153) against the backdrop of 37 Asian reference populations. The central aim of this study is to investigate the genetic affinities of the three historical Tibetan populations among themselves and to neighboring populations. Y-SNP and Y-STR profiles were assessed in these historical populations. Correspondence analyses (CA) were generated with Y-SNP haplogroup data. Y-STR haplotypes were determined and employed to generate multidimensional scaling (MDS) plots based on Rst distances. Frequency contour maps of informative Y haplogroups were constructed to visualize the distributions of specific chromosome types. Network analyses based on Y-STR profiles of individuals under specific Y haplogroups were generated to examine the genetic heterogeneity among populations. Average gene diversity values and other parameters of population genetics interest were estimated to characterize the populations. The Y chromosomal results generated in this study indicate that using two sets of markers (Y-SNP, and Y-STR) the three Tibetan populations are genetically distinct. In addition, U-Tsang displays the highest gene diversity, followed by Amdo and Dotoe. The results of this transcontinental biogeographical investigation also indicate various degrees of paternal genetic affinities among these three Tibetan populations depending on the type of loci (Y-SNP or Y-STR) analyzed. The CA generated with Y-SNP haplogroup data demonstrates that Amdo and U-Tsang are closer to each other than to any neighboring non-Tibetan group. In contrast, the MDS plot based on Y-STR haplotypes displays Rst distances that are much shorter between U-Tsang and its geographic nearby populations of Ladakh, Punjab, Kathmandu and Newar than between it and Amdo. Moreover, although Dotoe is isolated from all other groups using both types of marker systems, it lies nearer to the other Tibetan collections in the Y-SNP CA than in the Y-STR MDS plot. High resolution and shallow evolutionary time frames engendered by Y-STR based analyses may reflect a more recent demographic history than that delineated by the more conserved Y-SNP markers.