MA40254 Differential and geometric analysis : Exercises 8
Hand in answers by 1:15pm on Wednesday 29 November for the Seminar of Thursday 30 November Homepage: http://moodle.bath.ac.uk/course/view.php?id=57709 ex08
  • 0 (Warmup). Let M be an n-dimensional submanifold of Rs. Prove (from the definition of d on submanifolds) that if αΩk(M) is exact, i.e., α=dβ, then α is closed, i.e., dα=0.

    Solution
    Suppose α=dβ. We have seen in lectures that d2=0 so dα=d2β=0, but let us recall how to prove this. Note that it suffices to prove dα=0 locally on M, i.e., on sufficiently small open neighbourhoods UM of each pM. There are two ways to proceed.

    First by definition, we may assume that on UM, β=ιγ, where γΩk(U) and ι:UMU is the inclusion. Thus dα=d2(ιγ)=d(ιdγ)=ι(d2γ)=0 on UM. Alternatively, we may assume that there is a parametrisation φ:U~UM. Then φdα=φd2β=dφdβ=d2φβ=0, so dα=0 on UM because φ:Ωk(UM)Ωk(U~) is a bijection with inverse (φ1).

  • 1. Let MRs be an n-dimensional submanifold and αΩk(M) for k>0.

    • (i) Suppose that M is diffeomorphic to Rn and α is closed; prove that α is exact.

      Hint
      Let φ:RnM be a diffeomorphism, and consider the pullback φαΩk(Rn).

    • (ii) Suppose k=n; show that α is closed.

      Hint
      What is the dimension of Altn+1(TxM)?

    • (iii) If k=n, does α have to be exact?

      Hint
      Look for examples on M=S1. A previous exercise may help.

  • 2. Give an example of an n-dimensional submanifold MRs and an αΩk(M) such that α is not the pullback by the inclusion of any βΩk(Rs).

    Hint
    The simplest examples have n=s=1 and k=0.

  • 3. Let S2={zR3:z=1}, and U=S2{(0,0,1)}, and consider the parametrisation φ:R2U defined by

    (x1,x2)11+x2(2x1,2x2,x21).

    Let α=idz3 where i:S2R3 is the inclusion and z1,z2,z3 denote the coordinate functions on R3. Compute φα.

    Hint
    φidz3=(iφ)dz3=.

MA40254 Differential and geometric analysis : Solutions 8

  • 1.

    • (i) Suppose α is closed. Let φ:RnM be a diffeomorphism. Then φαΩk(Rn) is closed, so exact by the Poincaré lemma, i.e, there is some γΩk1(Rn) such that dγ=φα. If we let β=(φ1)γΩk1(M) then dβ=(φ1)dγ=(φ1)(φα)=(φφ1)α=α.

    • (ii) For each pM, dαp is an element of Altn+1(TpM). Since n+1>dimTpM, we have Altn+1(TpM)={0}, so dα is zero everywhere.

    • (iii) No. Consider M=S1R2, and let αΩ1(S1) be the pullback by i:S1R2{0} of the non-exact 1-form

      γ=x2dx1+x1dx2x12+x22Ω1(R2{0})

      from Exercises 6 Q 4(iii). The same argument as there shows that α is not exact either.

  • 2. We can for instance take M=(0,)R, and k=0. Then an element of Ωk(M) is just a smooth function (0,)R and pullback reduces to restriction. If we take it to be x1x then that is not the restriction of any smooth function RR.

  • 3. We have φidz3=(iφ)dz3=d((iφ)z3) and (iφ)z3=(x12+x221)/(1+x12+x22). The exterior derivative of this is

    4x1dx1+4x2dx2(1+x12+x22)2.