By reading the input from one track, and using two more tracks for the stacks, a tm can simulate the two stack pdm.
Conversely, given a tm, build a pdm that reads the input and pushes it onto the first stack. At the end-of-word marker, pop the first stack and push the letters onto the second stack, thereby reversing the word. Now the first character of the word is at the top of the second stack, and the first stack is empty, except for its bottom-of-stack symbol. From here, states and state transitions correspond. Move the pdm to q0, the start state of the tm. Based on the character at the top of the second stack and the current state, select the character that the tm would write, and push it on the first stack for a right move, or push it and the symbol from the first stack onto the second stack for a left move. The two stacks together form the tape of the tm.