p-adic "laplacian'

[hung]

hello, do anyone know of a good analogue of the laplacian for the p-adic field? if so, what is its spectrum, say on an interval, a + (P^n)Zp?
thank you,
thl