Perturbations of local maxima and comparison principles for boundary-degenerate linear differential equations

Abstract

We develop strong and weak maximum principles for boundary-degenerate elliptic and parabolic linear second-order partial differential operators, Au := -tr(aD2u)-<b, Du> + cu, with partial Dirichlet boundary conditions. The coefficient, a(x), is assumed to vanish along a non-empty open subset, ∂0O, called the degenerate boundary portion, of the boundary, ∂O, of the domain O⊂Rd, while a(x) is non-zero at any point of the non-degenerate boundary portion, ∂1O := ∂O∂0O. If an A-subharmonic function, u in C2(O) or W2,dloc(O), is C1 up to ∂0O and has a strict local maximum at a point in ∂0O, we show that u can be perturbed, by the addition of a suitable function w∈ C2(O) C1(Rd), to a strictly A-subharmonic function v=u+w having a local maximum in the interior of O. Consequently, we obtain strong and weak maximum principles for A-subharmonic functions in C2(O) and W2,dloc(O) which are C1 up to ∂0O. Only the non-degenerate boundary portion, ∂1O, is required for boundary comparisons. Our results extend those in Daskalopoulos and Hamilton (1998), Epstein and Mazzeo [arXiv:1110.0032], and the author [arXiv:1204.6613, 1306.5197], where tr(aD2u) is in addition assumed to be continuous up to and vanish along ∂0O in order to yield comparable maximum principles for A-subharmonic functions in C2(O), while the results developed here for A-subharmonic functions in W2,dloc(O) are entirely new.

0

Turn this paper into a lesson

ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.

Discussion (0)

Sign in to join the discussion.

Loading comments…