OctoInformBeta

OpenAI preprint, September 2026 · §4. Geodesic passages and the index form

Lemma 4.5 [A gauge on a turning excursion]: The normal index splits into coordinate energy and endpoint contributions

A turning gauge controls return boundary data.

Hover over a blue or green term in the text, or a node in the graph, to find it in the other. Select it to read its declaration both in Lean and in words.

The Setting

Consider a geodesic excursion into a neck, with return duration LL and conserved transverse momentum pp. A variation has an axial component and a transverse component, with separate endpoint data.

The split tangent space is ℝ×ℝ2\mathbb R\times\mathbb R^2; write U=(Ut,U⟂)U=(U^t,U^\perp) for a split tangent vector.✦ Transverse vectors belong to ℝ2\mathbb R^2, and transverse matrices belong to ℝ2×2\mathbb R^{2\times2}. For transverse matrices H,QH,Q, transverse vectors u,z∈ℝ2u,z\in\mathbb R^2, and split vectors U,V∈ℝ×ℝ2U,V\in\mathbb R\times\mathbb R^2, the matrix bilinear pairing ℬH\mathcal B_H and the split neck metric gQg_Q are the forms in (1): ℬH(u,z)=u⋅Hz,gQ(U,V)=UtVt+ℬQ(U⟂,V⟂).\mathcal B_H(u,z)=u\cdot Hz, \qquad g_Q(U,V)=U^tV^t+\mathcal B_Q(U^\perp,V^\perp).(1)

The one-sided cutoff aa and its integrated cutoff AA are the real functions in (2): a(u)={exp⁡(−12/3−u),u<2/3,0,u≥2/3,A(u)=∫0ua(y)dy.a(u)= \begin{cases} \exp\!\left(-\dfrac{1}{2/3-u}\right),&u<2/3,\\ 0,&u\geq2/3, \end{cases} \qquad A(u)=\int_0^u a(y)\,dy.(2) Let ϑ\vartheta denote the standard smooth transition from zero to one. For a real neck parameter δ\delta and neck height tt, the shear transition χ\chi, cometric shear σ\sigma, and diagonal coefficients d1,d2d_1,d_2 are the real functions in (3): χ(t)=ϑ(6(t−5/12)),σ(δ,t)=δ2χ(t),d1(δ,t)=1+δA(t),d2(δ,t)=1+δA(1−t).\begin{aligned} \chi(t)&=\vartheta\!\left(6(t-5/12)\right), &\sigma(\delta,t)&=\delta^2\chi(t),\\ d_1(\delta,t)&=1+\delta A(t), &d_2(\delta,t)&=1+\delta A(1-t). \end{aligned}(3) The cometric matrix P(δ,t)P(\delta,t) is the real matrix in (4), where di=di(δ,t)d_i=d_i(\delta,t) and σ=σ(δ,t)\sigma=\sigma(\delta,t): P(δ,t)=(d1+d2σ2d2σd2σd2).P(\delta,t)= \begin{pmatrix} d_1+d_2\sigma^2&d_2\sigma\\ d_2\sigma&d_2 \end{pmatrix}.(4)✦

The endpoint frame matrix C2(δ)C_2(\delta) and terminal cometric Pend(δ,τ)P_{\mathrm{end}}(\delta,\tau), at inward endpoint height τ\tau, are given by (5): C2(δ)=(011−δ2),Pend(δ,τ)=C2(δ)P(δ,1−τ)C2(δ)𝖳.C_2(\delta)=\begin{pmatrix}0&1\\1&-\delta^2\end{pmatrix}, \qquad P_{\mathrm{end}}(\delta,\tau) =C_2(\delta)P(\delta,1-\tau)C_2(\delta)^{\mathsf T}.(5) For an endpoint choice ε∈{0,1}\varepsilon\in\{0,1\}, the coordinate cometric Pε,δP_{\varepsilon,\delta} and transverse metric matrix Qε,δQ_{\varepsilon,\delta} are the matrix families in (6): Pε,δ(τ)={P(δ,τ),ε=0,Pend(δ,τ),ε=1,Qε,δ(τ)=Pε,δ(τ)−1.P_{\varepsilon,\delta}(\tau)= \begin{cases} P(\delta,\tau),&\varepsilon=0,\\ P_{\mathrm{end}}(\delta,\tau),&\varepsilon=1, \end{cases} \qquad Q_{\varepsilon,\delta}(\tau)=P_{\varepsilon,\delta}(\tau)^{-1}.(6)

For fixed momentum p∈ℝ2p\in\mathbb R^2, the reduced force FpF_p and force derivative Fp′F'_p are the real functions in (7); primes on matrix families denote differentiation in inward height: Fp(τ)=ℬPε,δ′(τ)(p,p),Fp′(τ)=ℬPε,δ″(τ)(p,p).F_p(\tau)=\mathcal B_{P'_{\varepsilon,\delta}(\tau)}(p,p), \qquad F'_p(\tau)=\mathcal B_{P''_{\varepsilon,\delta}(\tau)}(p,p).(7) Along an inward height curve τ(s)\tau(s) parametrized by geodesic time ss, write Pˆ(s)=Pε,δ(τ(s))\widehat P(s)=P_{\varepsilon,\delta}(\tau(s)) and Qˆ(s)=Qε,δ(τ(s))\widehat Q(s)=Q_{\varepsilon,\delta}(\tau(s)). For a height velocity v(s)v(s), the velocity at fixed momentum is the split vector field TT of (8): T(s)=(v(s),Pˆ(s)p).T(s)=\bigl(v(s),\widehat P(s)p\bigr).(8)✦

The Tangential Gauge

Let V,TV,T be split vector fields and QQ a transverse matrix family, and put q(s)=gQ(s)(V(s),T(s))q(s)=g_{Q(s)}(V(s),T(s)). The affine tangential interpolation α\alpha is the real function in (9), with fixed real parameters L,v0L,v_0; a quotient with zero denominator is assigned zero in this formula: α(s)=q(0)−Vt(0)v0+sL[q(L)+Vt(L)v0−(q(0)−Vt(0)v0)].\alpha(s)=q(0)-\frac{V^t(0)}{v_0} +\frac{s}{L}\left[ q(L)+\frac{V^t(L)}{v_0} -\left(q(0)-\frac{V^t(0)}{v_0}\right)\right].(9)✦ The tangential correction ff is the scalar function in (10): f(s)=α(s)−q(s).f(s)=\alpha(s)-q(s).(10)✦ The adjusted field WW is the split vector field defined by (11): W(s)=V(s)+f(s)T(s).W(s)=V(s)+f(s)T(s).(11)✦

The Coordinate Normal Index

The neck Christoffel operator ΓP,H\Gamma_{P,H} and its axial derivative Γt;P,H,J\Gamma_{t;P,H,J}, associated to transverse matrices P,H,JP,H,J, are the bilinear operators on split vectors given in (12): ΓP,H(U,V)=(−12ℬH(U⟂,V⟂),12(UtPHV⟂+VtPHU⟂)),Γt;P,H,J(U,V)=(−12ℬJ(U⟂,V⟂),12(Ut(PJ−PHPH)V⟂+Vt(PJ−PHPH)U⟂)).\begin{aligned} \Gamma_{P,H}(U,V) &=\left(-\frac12\mathcal B_H(U^\perp,V^\perp), \frac12\bigl(U^tPHV^\perp+V^tPHU^\perp\bigr)\right),\\ \Gamma_{t;P,H,J}(U,V) &=\left(-\frac12\mathcal B_J(U^\perp,V^\perp), \frac12\bigl(U^t(PJ-PHPH)V^\perp +V^t(PJ-PHPH)U^\perp\bigr)\right). \end{aligned}(12) With these matrix arguments suppressed on Γ\Gamma and Γt\Gamma_t, the coordinate curvature tensor RP,H,JR_{P,H,J} is the trilinear operator on split vectors U,V,ZU,V,Z in (13): RP,H,J(U,V)Z=UtΓt(V,Z)−VtΓt(U,Z)+Γ(U,Γ(V,Z))−Γ(V,Γ(U,Z)).R_{P,H,J}(U,V)Z =U^t\Gamma_t(V,Z)-V^t\Gamma_t(U,Z) +\Gamma(U,\Gamma(V,Z))-\Gamma(V,\Gamma(U,Z)).(13) The normal index density 𝒩P,Q,H,J\mathcal N_{P,Q,H,J} is the scalar function of split vectors T,V,DT,V,D defined in (14): 𝒩P,Q,H,J(T,V,D)=gQ(K,K)−gQ(V,RP,H,J(V,T)T)−gQ(K,T)2,K=D+ΓP,H(T,V).\begin{aligned} \mathcal N_{P,Q,H,J}(T,V,D) &=g_Q(K,K)-g_Q\bigl(V,R_{P,H,J}(V,T)T\bigr)-g_Q(K,T)^2,\\ K&=D+\Gamma_{P,H}(T,V). \end{aligned}(14)✦

The returning normal index density nY,D⟂(s)n^\perp_{Y,D}(s) of split vector fields Y,DY,D is the normal index density evaluated at the coordinate cometric, its inverse transverse metric matrix and the first two height derivatives of that metric matrix at τ(s)\tau(s), with the velocity T(s)T(s) of (8) and the vectors Y(s),D(s)Y(s),D(s). For a differentiable split vector field YY along the returning trajectory, the returning normal index density nY⟂n_Y^\perp is the scalar function obtained by setting D=Y′D=Y′ in nY,D⟂n^\perp_{Y,D}, and the normal index I⟂[Y]I^\perp[Y] is the integral of that density in (15). nY⟂(s)=𝒩Pˆ(s),Qˆ(s),Qε,δ′(τ(s)),Qε,δ″(τ(s))(T(s),Y(s),Y′(s)),I⟂[Y]=∫0LnY⟂(s)ds.\begin{aligned} n_Y^\perp(s) &=\mathcal N_{\widehat P(s),\widehat Q(s), Q'_{\varepsilon,\delta}(\tau(s)),Q''_{\varepsilon,\delta}(\tau(s))} \bigl(T(s),Y(s),Y'(s)\bigr),\\ I^\perp[Y]&=\int_0^L n_Y^\perp(s)\,ds. \end{aligned}(15) For a transverse matrix family QQ, vector families k,z′k,z', and scalar functions h,w,w′h,w,w', the completed square energy density 𝔢Q,k,h\mathfrak e_{Q,k,h} is the real function in (16): 𝔢Q,k,h(z′,w,w′;s)=ℬQ(s)(z′(s)−w(s)k(s),z′(s)−w(s)k(s))+w′(s)2+h(s)w(s)2.\mathfrak e_{Q,k,h}(z',w,w';s) =\mathcal B_{Q(s)}\bigl(z'(s)-w(s)k(s),z'(s)-w(s)k(s)\bigr) +w'(s)^2+h(s)w(s)^2.(16)✦ The cometric coordinate energy E(z,w)E(z,w) of a differentiable split field (w,z)(w,z) is the integral in (17), with coefficients evaluated along the returning trajectory: E(z,w)=∫0L𝔢Qˆ,k,h(z′,w,w′;s)ds,k(s)=Pε,δ′(τ(s))p,h(s)=−12Fp′(τ(s)).\begin{aligned} E(z,w)&=\int_0^L\mathfrak e_{\widehat Q,k,h}(z',w,w';s)\,ds,\\ k(s)&=P'_{\varepsilon,\delta}(\tau(s))p, \qquad h(s)=-\frac12F'_p(\tau(s)). \end{aligned}(17) Write z1z_1 for the first coordinate of a transverse vector or vector field.

The Claim

Theorem 1 (A gauge on a turning excursion). Let L,v0∈ℝL,v_0\in\mathbb R, let V,T:ℝ→ℝ×ℝ2V,T:\mathbb R\to\mathbb R\times\mathbb R^2 and Q:ℝ→ℝ2×2Q:\mathbb R\to\mathbb R^{2\times2}, and define WW by (11).

  1. Suppose v0≠0v_0\ne0 and Tt(0)=v0T^t(0)=v_0. The initial endpoint conclusion is (18): Wt(0)=0.W^t(0)=0.(18)✦

  2. At the terminal endpoint, assume L≠0L\ne0, v0≠0v_0\ne0, and Tt(L)=−v0T^t(L)=-v_0. The endpoint conclusion is (19): Wt(L)=0.W^t(L)=0.(19)✦

  3. At any time ss for which gQ(s)(T(s),T(s))=1g_{Q(s)}(T(s),T(s))=1, the tangential conclusion is (20): gQ(s)(W(s),T(s))=α(s).g_{Q(s)}(W(s),T(s))=\alpha(s).(20)✦

  4. If Q,V,TQ,V,T are smooth on a subset U⊂ℝU\subset\mathbb R, the regularity conclusion is (21): W∈C∞(U;ℝ×ℝ2).W\in C^\infty(U;\mathbb R\times\mathbb R^2).(21)✦

  5. Fix an endpoint choice ε\varepsilon, a real neck parameter δ\delta, momentum p∈ℝ2p\in\mathbb R^2, and L>0L>0. Suppose Pε,δ(τ(s))P_{\varepsilon,\delta}(\tau(s)) is nonsingular for every s∈[0,L]s\in[0,L]. Assume τ,v:ℝ→ℝ\tau,v:\mathbb R\to\mathbb R have derivatives τ′(s)=v(s)\tau'(s)=v(s) and v′(s)=−Fp(τ(s))/2v'(s)=-F_p(\tau(s))/2 at every point of [0,L][0,L], with τ(0)=τ(L)=0\tau(0)=\tau(L)=0 and v(L)=−v(0)v(L)=-v(0). Let z:ℝ→ℝ2z:\mathbb R\to\mathbb R^2 and w:ℝ→ℝw:\mathbb R\to\mathbb R have continuous derivatives on [0,L][0,L], with w(0)=w(L)=0w(0)=w(L)=0. Suppose the tangential pairing s↦gQˆ(s)((w(s),z(s)),T(s))s\mapsto g_{\widehat Q(s)}((w(s),z(s)),T(s)) has constant derivative m∈ℝm\in\mathbb R at every point of [0,L][0,L], where TT is given by (8). The returning normal index identity is (22): I⟂[(w,z)]=E(z,w)+δa(0)v(0)2(z1(0)2+z1(L)2)−(p⋅(z(L)−z(0)))2L.I^\perp[(w,z)] =E(z,w)+\frac{\delta a(0)v(0)}{2} \bigl(z_1(0)^2+z_1(L)^2\bigr) -\frac{\bigl(p\cdot(z(L)-z(0))\bigr)^2}{L}.(22)✦

  6. For nonnegative real numbers κ,v\kappa,v and transverse endpoint vectors z0,zL∈ℝ2z_0,z_L\in\mathbb R^2, the boundary estimate is (23): κv4((zL−z0)1)2≤κv2((z0)12+(zL)12).\frac{\kappa v}{4}\bigl((z_L-z_0)_1\bigr)^2 \leq\frac{\kappa v}{2}\bigl((z_0)_1^2+(z_L)_1^2\bigr).(23)✦

What Is Checked

The gauge assertions are stated over split vector fields and a transverse matrix family, the return identity over the endpoint cometric and differentiable returning trajectories with continuously differentiable variations, and the boundary estimate over arbitrary transverse endpoint vectors and nonnegative real coefficients.