By a slight abuse of notation, we denote the local cohomology of these sheaves by
${H}_{max/min}^{*}({\mathcal{U}}_{\alpha},g,a)$
, respectively. If
${\mathcal{U}}_{\alpha}$
is disjoint from all singular strata, then
$${H}^{j}({\mathcal{U}}_{\alpha},{\mathcal{\mathcal{L}}}_{max/min,a})=\{\begin{array}{ccc}\mathbb{R}& & j=0\\ 0& & j>0.\end{array}$$
Neither the weight function
${x}^{a}$
nor the metric
$g$
play a role. On the other hand, first note that
$$\left({\mathcal{\mathcal{L}}}_{max/min,a}\right(\mathcal{V}\times {C}_{1}\left(F\right)),{d}_{max/min,a})=\left({L}^{2}{\Omega}^{*}\right(\mathcal{V}),{d}_{max})\widehat{\otimes}\left({L}^{2}{\Omega}^{*}\right({C}_{1}\left(F\right),g,a),{d}_{max/min,a}).$$
Furthermore,
$dim{H}^{j}(\mathcal{V})<\infty $
so the range of
${d}_{max}$
on
${L}^{2}{\Omega}^{*}(\mathcal{V})$
is closed. Hence by Proposition 1 and quasiisometry invariance,
$$\begin{array}{ccc}{H}_{max/min}^{j}(\mathcal{V}\times {C}_{1}(F),g,a)& =& {}^{j}{\oplus}_{\ell =0}{H}^{\ell}(\mathcal{V})\otimes {H}_{max/min}^{j\ell}\left({C}_{1}\right(F),d{x}^{2}+{x}^{2}\kappa ,a)\end{array}$$  
$$\begin{array}{ccc}& =& {H}_{max/min}^{j}\left({C}_{1}\right(F),d{x}^{2}+{x}^{2}\kappa ,a).\end{array}$$  
We have now reduced the computation of weighted de Rham cohomology on
$X$
to that of the truncated cone. For the next step we use the following regularity result
Lemma 6.
$${}^{\infty}{\bigcap}_{\ell =1}\mathcal{D}({\Delta}_{abs/rel}^{\ell},a)\subset {\mathcal{A}}^{*}{\Omega}^{*}\left({C}_{1}\right(F\left)\right).$$

Proof.
The proof relies on the mapping properties of a parametrix for
${\Delta}_{abs/rel,a}$
; this parametrix is an element
$G\in {x}^{2}{\Psi}_{b}^{2,\mathcal{\mathcal{E}}}(M,{\Omega}^{*})$
, the calculus of
$b$
pseudodifferential operators on
$M$
, such that
$G\Delta =IR$
for some residual element
$R$
which satisfies
$R:{L}^{2}{\Omega}^{*}\to {\mathcal{A}}^{*}{\Omega}^{*}$
. (Both
$G$
and
$R$
depend on
$a$
and the choice of extension.) Suppose that
$u\in \mathcal{D}\left({\Delta}_{max,a}\right)$
, so in particular
$u,\Delta u\in {x}^{a}{L}^{2}{\Omega}^{*}$
. Then
$u=Gf+Ru$
, hence
$u\in {x}^{a+2}{H}_{b}^{2}{\Omega}^{*}+{\mathcal{A}}^{*}$
. Induction on
$\ell $
gives the result. We refer to [18] and [16] for more details.
Now we prove three lemmas taken from [
4]
. In the following,
$g=d{x}^{2}+{x}^{2}\kappa $
on
${C}_{1}\left(F\right)$
.
Proposition 5.
Let
$r:{C}_{1}\left(F\right)\backslash \left\{0\right\}\to F$
be the canonical retraction map. Then
$${r}^{*}:{L}^{2}{\Omega}^{k}\left(F\right)\cap kerd\u27f6\mathcal{D}\left({d}_{max,a}\right)\cap {x}^{a}{L}^{2}{\Omega}^{k}\left({C}_{1}\right(F\left)\right)$$
is welldefined and bounded if and only if
$k<\frac{f+1}{2}a$
.

Proof.
Let
$\alpha \in {L}^{2}{\Omega}^{k}\left(F\right)$
; then
$${\int}_{{C}_{1}\left(F\right)}\alpha {}_{g}^{2}{x}^{2a}d{V}_{g}=\left\alpha \right{}_{{L}^{2}\left(F\right)}^{2}{\int}_{0}^{1}{x}^{f2k2a}dx,$$
and this is finite if and only if
$k<\frac{f+1}{2}a$
. Since
$d{r}^{*}\alpha ={r}^{*}d\alpha =0$
, the image of
${r}^{*}$
lies in
$\mathcal{D}\left({d}_{max,a}\right)$
.
Proposition 6.
There exists a
$K>0$
such that for any
$\omega =\alpha +dx\wedge \beta \in {x}^{a}{L}^{2}{\Omega}^{k}\left(C\right(F\left)\right)$
there exists some
$c\in (1/2,1)$
for which
$$\left\right\alpha \left(c\right){}_{{L}^{2}\left(F\right)}^{2}\le K\left\omega \right{}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}.$$

Proof.
If not, then for any
$N>0$
, there exists some (nontrivial)
$\omega \in {x}^{a}{L}^{2}{\Omega}^{k}\left({C}_{1}\right(F\left)\right)$
such that
$N\left\right\omega {}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}<\left\alpha \right(x\left)\right{}_{{L}^{2}\left(F\right)}^{2}$
for all
$x\in (1/2,1)$
. However, if this is the case, then for this
$\omega $
,
$$\left\right\omega {}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}\ge \left\alpha \right{}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}\ge {\int}_{1/2}^{1}\left\right\alpha \left(x\right){}_{{L}^{2}\left(F\right)}^{2}{x}^{f2k2a}dx$$
$$>N\left\right\omega {}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}{\int}_{1/2}^{1}{x}^{f2k2a}dx.$$
This is a contradiction when
$N$
is large.
Proposition 7.
If
$k<\frac{f+3}{2}a$
, then for any
$c\in (1/2,1)$
, the map
$${x}^{a}{L}^{2}{\Omega}^{k}\left({C}_{1}\right(F\left)\right)\ni \omega =\alpha +dx\wedge \beta \u27f6{K}_{c}\left(\omega \right)={\int}_{c}^{x}\beta \left(s\right)ds\in {x}^{a}{L}^{2}{\Omega}^{k1}\left({C}_{1}\right(F\left)\right)$$
is bounded.

Proof.
First,
$$\begin{array}{c}\left\right{K}_{c}\omega {}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}={\int}_{0}^{1}{\int}_{F}{\left{\int}_{c}^{x}\beta \left(s\right)ds\right}_{\kappa}^{2}{x}^{f2k+22a}dxd{V}_{\kappa}.\end{array}$$ 
(12)

Next, for any real number
$b$
,
$${\left{\int}_{c}^{x}\beta \left(s\right)ds\right}_{\kappa}^{2}\le {\left({\int}_{c}^{x}\left\beta \right(s){}_{\kappa}ds\right)}^{2}={\left({\int}_{c}^{x}{s}^{b}\left{s}^{b}\beta \right(s){}_{\kappa}ds\right)}^{2}$$
$$\le {\int}_{c}^{x}{s}^{2b}ds{\int}_{c}^{x}\left\beta \right(s){}_{\kappa}^{2}{s}^{2b}ds=\{\begin{array}{cc}\frac{{x}^{12b}{c}^{12b}}{12b}{\int}_{c}^{x}\left\beta \right(s){}_{\kappa}^{2}{s}^{2b}ds& b\ne 1/2\\ (log(x)log(c\left)\right){\int}_{c}^{x}\left\beta \right(s){}_{\kappa}^{2}{s}^{2b}ds& b=1/2.\end{array}$$
Setting
$2b=f2k+22a$
and using this in ( 12 ) gives
$$\left\right{K}_{c}\omega {}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}$$
$$\le \{\begin{array}{cc}{\int}_{0}^{1}\frac{xc(x/c{)}^{f2k+22a}}{1(f2k+22a)}{\int}_{c}^{x}\left\beta \right(s){}_{\kappa}^{2}{s}^{f2k+22a}dsd{V}_{\kappa}dx& k\ne \frac{f+1}{2}a\\ {\int}_{0}^{1}x(ln(x)ln(c\left)\right){\int}_{c}^{x}\left\beta \right(s){}_{\kappa}^{2}{s}^{f2k+22a}dsd{V}_{\kappa}dr& k=\frac{f+1}{2}a\end{array}$$
$$\le \{\begin{array}{cc}\left({\int}_{0}^{1}\frac{xc(x/c{)}^{f2k+22a}}{1(f2k+22a)}dx\right)\left\right\beta {}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}& k\ne \frac{f+1}{2}a\\ \left({\int}_{0}^{1}x(ln(x)ln(c\left)\right)dx\right)\left\right\beta {}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}& k=\frac{f+1}{2}a.\end{array}$$
Since
$c$
is bounded away from
$0$
, both of these coefficients on the right are uniformly bounded when
$k\le \frac{f+3}{2}a$
.
3.5 Poincaré lemma
We now compute the weighted cohomologies of the truncated cone.
Lemma 7.
$${H}_{max}^{k}\left({C}_{1}\right(F),g,a)=\{\begin{array}{ccc}{H}^{k}\left(F\right)& & k<(f+1)/2a\\ 0& & k\ge (f+1)/2a,\end{array}$$
and
$${H}_{min}^{k}\left({C}_{1}\right(F),g,a)=\{\begin{array}{ccc}{H}^{k}\left(F\right)& & k\le (f1)/2a\\ 0& & k>(f1)/2a.\end{array}$$

Proof.
By Lemma 6 , we may work exclusively with conormal forms. First let
$k<\frac{f+1}{2}a$
.
We wish to define a map
$$R:{H}_{max}^{k}\left({C}_{1}\right(F),g,a)\u27f6{H}^{k}\left(F\right),$$
and show that it is an isomorphism. Let
$\left[\omega \right]\in {H}_{max}^{k}\left({C}_{1}\right(F),g,a)$
, and choose a conormal representative
$\omega =\alpha +dx\wedge \beta \in \mathcal{D}(dmax,g,a)$
. For any
$c\in (1/2,1)$
, set
$R\left(\right[\omega \left]\right)=\left[\alpha \right(c\left)\right]$
.
To check that
$R$
is defined independently of all choices, first note that
$d\omega ={d}_{F}\alpha +dx\wedge ({\alpha}^{\prime}{d}_{F}\beta )=0$
, so
${d}_{F}\alpha \left(c\right)=0$
for any
$c$
. Next, if
$\stackrel{~}{\omega}=\stackrel{~}{\alpha}+dx\wedge \stackrel{~}{\beta}$
is another conormal representative of
$\left[\omega \right]$
, then there exists
$\eta =\mu +dx\wedge \nu \in {x}^{a}{L}^{2}{\Omega}^{k1}\left({C}_{1}\right(F\left)\right)\cap {\mathcal{A}}^{*}$
with
$\stackrel{~}{\omega}=\omega +d\eta $
. This implies in particular that
$\stackrel{~}{\alpha}\left(c\right)=\alpha \left(c\right)+{d}_{F}\mu \left(c\right)$
, so
$\left[\alpha \right(c\left)\right]=\left[\stackrel{~}{\alpha}\right(c\left)\right]$
.
Similarly,
${\int}_{c}^{{c}^{\prime}}\beta \left(s\right)ds\in {L}^{2}{\Omega}^{k1}\left(F\right)$
and hence
${\alpha}^{\prime}={d}_{F}\beta $
implies that
$$d{\int}_{c}^{{c}^{\prime}}\beta \left(s\right)ds={\int}_{c}^{{c}^{\prime}}{d}_{F}\beta \left(s\right)ds={\int}_{c}^{{c}^{\prime}}{\alpha}^{\prime}\left(s\right)ds=\alpha \left({c}^{\prime}\right)\alpha \left(c\right);$$
thus
$\left[\alpha \right(c\left)\right]=\left[\alpha \right({c}^{\prime}\left)\right]$
.
$R$
is certainly linear; it is bounded by Proposition 6 and surjective by Proposition 5 , so we must only show that it is injective. Suppose
$\omega =\alpha +dx\wedge \beta \in {\mathcal{A}}^{*}$
and
$R\left(\right[\alpha \left]\right)=\left[0\right]$
.
Then
$\alpha \left(c\right)={d}_{F}\eta $
for some
$\eta \in {\mathcal{C}}^{\infty}{\Omega}^{k1}\left(F\right)$
. By Proposition 5 ,
${r}^{*}\eta \in \mathcal{D}(dmax,g,a)$
, while Proposition 7 gives that
${\int}_{c}^{x}\beta \left(s\right)ds\in {x}^{a}{L}^{2}{\Omega}^{k1}\left({C}_{1}\right(F\left)\right)\cap {\mathcal{A}}^{*}$
. Thus
$$d\left(\eta +{\int}_{c}^{x}\beta \left(s\right)ds\right)={d}_{F}\eta +dx\wedge \beta \left(x\right)+{\int}_{c}^{x}{d}_{F}\beta \left(s\right)ds$$
$$=\alpha \left(c\right)+dx\wedge \beta \left(x\right)+{\int}_{c}^{x}{\alpha}^{\prime}\left(s\right)ds=\omega .$$
Since
$\omega \in {x}^{a}{L}^{2}{\Omega}^{k}\left({C}_{1}\right(F\left)\right)$
, this implies that
$\eta +{\int}_{c}^{x}\beta \left(s\right)ds\in \mathcal{D}(dmax,g,a)$
, hence
$\omega $
is exact in the maximal complex, i.e.
$\left[\omega \right]=\left[0\right]$
, as desired.
We next show that when
$k\ge \frac{f+1}{2}a$
, any
$\left[\omega \right]\in {H}_{max}^{k}\left({C}_{1}\right(F),g,a)$
is trivial. Thus, for any representative
$\omega =\alpha +dx\wedge \beta \in \mathcal{D}(dmax,g,a)\cap {\mathcal{A}}^{*}$
, we must find a
$(k1)$
form
$\eta \in \mathcal{D}(dmax,g,a)\cap {\mathcal{A}}^{*}$
with
$\omega ={d}_{max,a}\eta $
. Assume
$\omega {}_{\kappa}=O({x}^{p})$
for some
$p$
. The condition
${\int}_{0}^{1}\omega {}_{\kappa}^{2}{x}^{f2k2a}dxd{V}_{\kappa}<\infty $
gives that
$p>a+k\frac{f+3}{2}$
. Furthermore, by assumption,
$a+k\frac{f+3}{2}>1$
, so
${K}_{0}\left(\omega \right)={\int}_{0}^{x}\beta \left(s\right)ds$
is defined. Using
$\left\beta \right(s){}_{\kappa}=O({s}^{p+1})$
, we deduce that the integral
$$\left\right{K}_{0}\omega {}_{{x}^{a}{L}^{2}\left({C}_{1}\right(F\left)\right)}^{2}={\int}_{0}^{1}{\int}_{F}{\left{\int}_{0}^{x}\beta \left(s\right)ds\right}_{\kappa}^{2}{x}^{f2k+22a}dxd{V}_{\kappa}$$
is finite. Now,
$\alpha \in {\mathcal{A}}^{*}\cap {x}^{a}{L}^{2}{\Omega}^{k}\left({C}_{1}\right(F\left)\right)$
, so
$\alpha {}_{\kappa}=O({x}^{q})$
where
$2q+f2k2a>1$
, i.e.
$q>0$
, so
$$d\left({K}_{0}\right(\omega \left)\right)=dx\wedge \beta \left(x\right)+{\int}_{0}^{x}{\alpha}^{\prime}\left(s\right)ds=\omega \left(x\right),$$
as desired. We have now shown that
${K}_{0}\left(\omega \right)\in \mathcal{D}(dmax,g,a)$
and hence
$\left[\omega \right]=0$
in
${H}_{max}^{k}\left({C}_{1}\right(F),g,a)$
. This completes the computation of
${H}_{max}^{*}\left({C}_{1}\right(F),g,a)$
.
The computation of
${H}_{min}^{*}\left({C}_{1}\right(F),g,a)$
proceeds identically when
$k\le (f1)/2a$
or
$k\ge (f+3)/2a$
, but the remaining cases are treated slightly differently.
Claim 1.
If
$k\in \left(\right(f1)/2a,(f+1)/2)a)$
and
$\omega =\alpha +dx\wedge \beta \in {x}^{a}{L}^{2}{\Omega}^{k}\left({C}_{1}\right(F\left)\right)\cap \mathcal{D}\left({d}_{min,a}\right)\cap {\mathcal{A}}^{*}$
is closed, then
$\omega =d\eta $
for some
$\eta \in \mathcal{D}\left({d}_{min,a}\right)\cap {\mathcal{A}}^{*}$
. As a consequence,
${H}_{min}^{k}\left({C}_{1}\right(F),g,a)=0$
.

Proof.
As above,
${\alpha}^{\prime}={d}_{F}\beta $
, so if
$c,{c}^{\prime}\in (0,1)$
,
$${\Pi}_{0}\left(\alpha \right({c}^{\prime})\alpha (c\left)\right)={\Pi}_{0}{\int}_{c}^{{c}^{\prime}}{\alpha}^{\prime}\left(s\right)ds={\Pi}_{0}{d}_{F}{\int}_{c}^{{c}^{\prime}}\beta \left(x\right)dx=0.$$
Since
$\omega \in \mathcal{D}\left({d}_{min,a}\right)$
,
${\Pi}_{0}\left(\alpha \right(x\left)\right)=o\left(1\right)$
, so
${\Pi}_{0}\left(\alpha \right(x\left)\right)=0$
for all
$x$
. This gives that
$$d{\int}_{c}^{x}\beta \left(s\right)ds=dx\wedge \beta +\alpha \left(x\right)\alpha \left(c\right)=\omega \alpha \left(c\right)=\omega {\Pi}_{\perp}\alpha \left(c\right)=\alpha d\left({r}^{*}\eta \right)$$
for some
$\eta \in {L}^{2}{\Omega}^{k1}\left(F\right)\cap \mathcal{D}\left(d\right)$
. Thus
$\alpha =d(\eta +{\int}_{c}^{x}\omega (s\left)ds\right)$
.
As for its domain, it suffices by Lemma
3 to show that
$\eta +{\int}_{c}^{x}\beta \left(s\right)ds\in \mathcal{D}\left({d}_{max,a}\right)$
since
$k1<(f1)/2a$
. But on the one hand,
$d(\eta +{\int}_{c}^{x}\beta (s\left)ds\right)=\omega \in {x}^{a}{L}^{2}{\Omega}^{k}\left({C}_{1}\right(F\left)\right)$
; furthermore,
$(\eta +{\int}_{c}^{x}\omega (s\left)ds\right)\in {L}^{2}{\Omega}^{k1}\left(C\right(F),g,a)$
since
${r}^{*}\eta \in {x}^{a}{L}^{2}{\Omega}^{k1}\left({C}_{1}\right(F\left)\right)$
, again because
$k1<(f+1)/2a$
, and
$\beta =o\left({x}^{(2k2+2af1)/2}\right)=o\left({x}^{p}\right)$
for some
$p>1$
, so
${\int}_{c}^{x}\beta \left(s\right)ds\in {x}^{a}{L}^{2}{\Omega}^{k1}\left({C}_{1}\right(F\left)\right)$
. This proves the claim.
Claim 2.
The map
${H}_{min}^{k}\left({C}_{1}\right(F),g,a)\to {H}^{k}\left(F\right)$
is injective when
$k\in \left(\right(f+1)/2a,(f+3)/2a)$
; hence, for
$k$
in this range,
${H}_{min}^{k}\left({C}_{1}\right(F),g,a)\sim ={H}_{max}^{i}\left(C\right(F),g,a)\sim =0$
.

Proof.
If
$\omega =\alpha +dx\wedge \beta \in {\mathcal{A}}^{*}$
represents a class in
${H}_{min}^{k}\left({C}_{1}\right(F),g,a)$
, then we already know that
$\omega =d\eta $
for some
$\eta \in \mathcal{D}\left({d}_{max,a}\right)\cap {\mathcal{A}}^{*}$
. So we must show that we can arrange for
$\eta $
to lie in
$\mathcal{D}\left({d}_{min,a}\right)$
as well. Since
$k1\in \left(\right(f1)/2+a,(f+1)/2a)$
,
$\beta \left(x\right)=o\left({x}^{p}\right)$
for some
$\nu >1$
, and so
$${\Pi}_{0}{\int}_{c}^{0}\beta \left(s\right)ds$$
is defined. We can choose
$\zeta $
which solves
$\alpha \left(c\right)={d}_{F}\zeta $
by specifying that
${\Pi}_{0}\left(\zeta \right)={\Pi}_{0}{\int}_{c}^{0}\beta \left(s\right)ds$
. Now let
$\eta =\zeta +{\int}_{0}^{x}\beta \left(s\right)ds$
. As before,
$\eta \in {\mathcal{A}}^{*}$
and
$d\eta =\omega $
. Furthermore,
${\Pi}_{0}\left(\eta \right(0\left)\right)=0$
and
${\partial}_{x}\left({\Pi}_{0}\right(\eta \left)\right)={\Pi}_{0}\left(\beta \right(x\left)\right)=o\left({x}^{\left(2\right(k1)+2af1)/2}\right)=o\left({x}^{p}\right)$
for
$p>1$
, so
${\Pi}_{0}\left(\eta \right)=o\left(1\right)$
. This means that
$\eta \in \mathcal{D}\left({d}_{min,a}\right)$
.
This completes the calculation of the local cohomology for the sheaves
${\mathcal{\mathcal{L}}}_{max/min,a}$
on
$X$
. By Proposition 4 , we now obtain one of our main results:
Theorem 4.
If
$(M,g)$
is a manifold with an incomplete edge metric and
$X$
is the associated stratified space, then
$${H}_{max}^{*}(M,g,a)=\{\begin{array}{ccc}I{H}_{\overline{\mathfrak{m}}+\ll a1\gg}^{*}(X,B)& & f\text{odd}\\ I{H}_{\overline{\mathfrak{m}}+\ll a1/2\gg}^{*}(X,B)& & f\text{even}\end{array}$$
and
$${H}_{min}^{*}(M,g,a)=\{\begin{array}{ccc}I{H}_{\underline{\mathfrak{m}}+<a>}^{*}(X,B)& & f\text{odd}\\ I{H}_{\underline{\mathfrak{m}}+<a1/2>}^{*}(X,B)& & f\text{even}\end{array};$$
here
$\ll t\gg $
denotes the least integer strictly greater than
$t$
and
$<t>$
denotes the least integer greater than or equal to
$t$
.
There are two important special cases which we single out:
Corollary 5.
The maximal and minimal de Rham cohomologies at weight zero correspond to upper and lower middle perversity intersection cohomology.
$$\begin{array}{c}\begin{array}{ccc}{H}_{max}^{*}(M,g,0)& =& I{H}_{\overline{\mathfrak{m}}}^{*}\left(X\right)\\ {H}_{min}^{*}(M,g,0)& =& I{H}_{\underline{\mathfrak{m}}}^{*}\left(X\right)\end{array}\end{array}$$ 
(13)

Moreover, when
$f$
is even, the maximal and minimal de Rham cohomologies at weights
$\pm 1/2$
coincide, and again correspond to upper and lower middle perversity intersection cohomology.
$$\begin{array}{c}\begin{array}{ccc}{H}_{max/min}^{*}(M,g,1/2)& =& I{H}_{\overline{\mathfrak{m}}}^{*}\left(X\right)\\ {H}_{max/min}^{*}(M,g,1/2)& =& I{H}_{\underline{\mathfrak{m}}}^{*}\left(X\right).\end{array}\end{array}$$ 
(14)

From §2, these weighted de Rham cohomology spaces are identified with the nullspaces of the associated (absolute and relative) Laplacians, and we conclude that the nullspaces of
${\Delta}_{abs/rel,a}$
are (finite dimensional and) identified with particular intersection cohomology spaces. Note that we are not asserting anything about the nullspaces of the `ordinary' Laplacians
${\Delta}_{abs/rel,0}$
on the weighted spaces
${x}^{a}{L}^{2}{\Omega}^{*}$
; indeed, these nullspaces are either infinite dimensional when
$a\ll 0$
or vanish identically when
$a\gg 0$
, cf. [18] .
4 Elliptic edge operators and minimal Hodge cohomology
To proceed further in the study of these weighted de Rham complexes, we must use elliptic methods. More specifically, we still wish to study the question of when there is a unique closed extension for
$d$
on
${x}^{a}{L}^{2}{\Omega}^{*}\left(M\right)$
for incomplete edge metrics, and we also wish to compute the minimal Hodge cohomology. We shall study both of these questions using the formally symmetric operator
${D}_{a}=d+{\delta}_{a}$
on
${x}^{a}{L}^{2}{\Omega}^{*}\left(M\right)$
. The proper context for this analysis is the calculus of pseudodifferential edge operators, and in the next subsection we review the generalities of this theory. After that we show how it applies to the specific problems at hand.
4.1 Edge operators
We now review the general theory of elliptic edge operators. This is the correct context to study
${D}_{a}$
for an incomplete edge metric (and also the corresponding operator for a complete edge metric). This theory is developed fully in [
18]
, and we refer there for more details Fix a local coordinate system
$({y}_{1},\dots ,{y}_{b})$
on
$B$
and
$({z}_{1},\dots ,{z}_{f})$
on
$F$
, so that
$w=(x,y,z)$
is a local coordinate system in some neighbourhood of a singular stratum in
$X$
.
By including the hypersurface
$\{x=0\}$
, we are blowing up
$B$
in
$X$
; the resulting manifold with boundary is denoted
$\overline{M}$
and its interior is denoted
$M$
.
A differential operator
$L$
on
$M$
is called an edge operator of order
$m$
if it can be expressed in the form
$$\begin{array}{c}L={\sum}_{j+\left\alpha \right+\left\beta \right\le m}{a}_{j,\alpha}(x,y,z)\left(x{\partial}_{x}{)}^{j}\right(x{\partial}_{y}{)}^{\alpha}{\partial}_{z}^{\beta},\end{array}$$ 
(15)

where the (scalar or matrixvalued) coefficients are bounded. We shall assume that these coefficients are smooth in these variables, down to
$x=0$
. For example, if
$g$
is a complete edge metric, then the scalar or Hodge Laplacian is an operator of this type; similarly, if
$g$
is an incomplete edge metric, then its Laplacian is of the form
${x}^{2}L$
, where
$L$
is an edge operator of order
$2$
.
4.1.1 Ellipticity and model operators
There is a welldefined symbol in this setting:
$$\sigma \left(L\right)(x,y,z;\xi ,\eta ,\zeta ):={\sum}_{j+\left\alpha \right+\left\beta \right=m}{a}_{j,\alpha}(x,y,z){\xi}^{j}{\eta}^{\alpha}{\zeta}^{\beta},$$
and we say that
$L$
is elliptic in the edge calculus provided
$\sigma \left(L\right)(x,y,z;\xi ,\eta ,\zeta )$
is invertible when
$(\xi ,\eta ,\zeta )\ne 0$
.
Ellipticity alone does not guarantee that
$L$
is Fredholm between appropriate function spaces; one must also require that certain model operators for
$L$
also be invertible. There are two such operators:

∙
The normal operator of
$L$
is defined by
$$N\left(L\right):={\sum}_{j+\left\alpha \right+\left\beta \right\le m}{a}_{j,\alpha ,\beta}(0,y,z)\left(s{\partial}_{s}{)}^{j}\right(s{\partial}_{u}{)}^{\alpha}{\partial}_{z}^{\beta}(s,u)\in {\mathbb{R}}^{+}\times {\mathbb{R}}^{b};$$
here
$y\in B$
enters only parametrically and the operator acts on functions on
${\mathbb{R}}^{+}\times {\mathbb{R}}^{b}\times F$
. This operator can be regarded as
$L$
with its coefficients frozen (in an appropriate sense) at
$x=0$
, acting on functions (or sections of an appropriate bundle) on the space
${\mathbb{R}}_{s}^{+}\times {\mathbb{R}}_{u}^{b}\times {F}_{z}$
.

∙
The indicial operator of
$L$
is defined by
$$I\left(L\right):={\sum}_{j+\left\beta \right\le m}{a}_{j,0,\beta}(0,y,z)(s{\partial}_{s}{)}^{j}{\partial}_{z}^{\beta}.$$
For example, the normal and indicial operators associated to the scalar Laplacian for the complete edge metric
${x}^{2}(d{x}^{2}+h)+\kappa $
are
$$N\left({\Delta}_{g}\right)={s}^{2}{\partial}_{s}^{2}+(1b)s{\partial}_{s}+{s}^{2}{\Delta}_{u}+{\Delta}_{\kappa},I\left({\Delta}_{g}\right)={s}^{2}{\partial}_{s}^{2}+(1b)s{\partial}_{s}+{\Delta}_{\kappa}.$$
The indicial operator captures some fundamental invariants associated to
$L$
:
Definition 5.
The number
$\gamma \in \mathbb{C}$
is said to be an indicial root of
$L$
at
${y}_{0}\in B$
if there exists a function
$v\left(z\right)$
on
$F$
such that (in local coordinates where
${y}_{0}$
corresponds to
$y=0$
)
$$I\left(L{)}_{{y}_{0}}\right({s}^{\gamma}v\left(z\right))=\left({\sum}_{j+\left\beta \right\le m}{a}_{j,0,\beta}(0,0,z)(s{\partial}_{s}{)}^{j}\right){s}^{\gamma}v(z)=\mathcal{O}({s}^{\gamma +1}).$$
Indicial roots may often be calculated in terms of eigenvalues for an induced elliptic operator on the fibre
${F}_{{y}_{0}}$
, and might depend on
${y}_{0}\in B$
.
The operator
$L$
acts naturally on weighted Sobolev spaces. Let
$M$
be a manifold with complete edge metric
$G$
. For
$\ell \in \mathbb{N}$
and
$\delta \in \mathbb{R}$
, define
$${x}^{\delta}{H}_{e}^{\ell}\left(M\right)=\{u={x}^{\delta}v:(x{\partial}_{x}{)}^{j}(x{\partial}_{y}{)}^{\alpha}{\partial}_{z}^{\beta}v\in {L}^{2}(M,d{V}_{G})\forall j+\alpha +\beta \le \ell \}.$$
(By interpolation and duality, these spaces can be defined for any
$\ell \in \mathbb{R}$
.) Clearly, if
$L$
is any edge operator of order
$m$
, then
$$\begin{array}{c}L:{x}^{\delta}{H}_{e}^{\ell +m}\left(M\right)\u27f6{x}^{\delta}{H}_{e}^{\ell}\left(M\right)\end{array}$$ 
(16)

for any
$\delta ,\ell $
. Further hypotheses, beyond the ellipticity of
$L$
are required to ensure that this mapping is wellbehaved.
The first instance of this is that the indicial roots of
$L$
yield weights
$\delta $
for which ( 16 ) does not have closed range; these are precisely the weights
$\delta $
for which an indicial root
$\gamma $
`just fails' to lie in
${x}^{\delta}{L}^{2}$
near
$x=0$
, i.e. where
${x}^{\gamma}\in {x}^{\delta \epsilon}{L}^{2}$
for any
$\epsilon >0$
but
${x}^{\gamma}/\in {x}^{\delta}{L}^{2}$
.
We denote this critical weight
$\delta $
associated to a given indicial root
$\gamma $
as
$\delta \left(\gamma \right)$
. With respect to the measure
$dxdydz$
,
$\delta \left(\gamma \right):=\text{Re}\gamma +1/2$
. However, the measure appearing in our application below is
${x}^{f2a}dxdydz$
, and
${x}^{\gamma}\in {x}^{\delta}{L}^{2}\left({x}^{f2a}dxdydz\right)$
near
$x=0$
if and only if
$\delta <\gamma +(f+1)/2a$
, so that we shall define
$\delta \left(\gamma \right)=\text{Re}\gamma +(f+1)/2a$
.
Even when
$\delta $
is not equal to one of these critical values, the behaviour of the normal operator at weight
$\delta $
plays another very important role.
Proposition 8.
Let
$L$
be an elliptic differential edge operator of order
$m$
.
Fix
$\delta $
such that
$\delta \ne \delta \left(\gamma \right(b\left)\right)$
for any indicial root
$\gamma \left(b\right)$
,
$b\in B$
. Suppose also that
$N\left(L\right):{s}^{\delta}{H}_{e}^{m}\to {s}^{\delta}{L}^{2}$
is surjective (for all
$b\in B$
). Then ( 16 ) is essentially surjective, in the sense that its range is closed and of finite codimension. On the other hand, if
$N\left(L\right)$
is injective on
${s}^{\delta}{L}^{2}$
, then any element of the nullspace of
$L$
is necessarily conormal.
There are many more refined statements one can make about the mapping properties of
$L$
. For later applications, we state only one very special result. We shall restrict to a special setting, which is what arises in our applications below. The hypothesis that the the normal operator
$N\left(L\right)$
is surjective is equivalent to the injectivity of the normal operator for the adjoint
${L}^{*}$
. This adjoint depends on the choice of measure, and we shall assume (as in our applications) that the adjoint of
$L$
on
${x}^{\delta}{L}^{2}$
corresponds to the same operator
$L$
on a `dually weighted' space
${x}^{{\delta}^{*}}{L}^{2}$
, for some
${\delta}^{*}>\delta $
. The fact that
$L$
has closed range implies the existence of a generalized inverse
$G:{x}^{\delta}{L}^{2}\to {x}^{\delta}{H}_{e}^{m}$
which satisfies
$LG=IP$
, where
$P$
is the orthogonal projector onto the cokernel. By duality, elements of this cokernel are identified with elements of the nullspace of
$L$
on
${x}^{{\delta}^{*}}{L}^{2}$
, and by the result above, these are conormal.
Proposition 9.
Let
$L$
satisfy the special assumptions of the preceding paragraph. Suppose furthermore that the interval
$(\delta ,{\delta}^{*})$
contains a finite set of indicial roots
${\gamma}_{j}$
,
$j=1,\dots ,N$
, all of which are constant in
$b\in B$
. Let
$f\in {x}^{{\delta}^{*}}{L}^{2}\cap {\mathcal{A}}^{*}$
. Then
$u=Gf$
satisfies
$Lu=f\phi $
where
$\phi \in {x}^{{\delta}^{*}}{L}^{2}\cap {\mathcal{A}}^{*}$
,
$L\phi =0$
, and
$u={\sum}_{j=1}^{N}{u}_{j}(y,z){x}^{{\gamma}_{j}}+v$
; where each
${u}_{j}(y,z)\in {\mathcal{C}}^{\infty}$
solves the indicial equation
$I\left(L\right)\left({s}^{{\gamma}_{j}}{u}_{j}\right(y,z\left)\right)=0$
, and the error term
$v\in {\mathcal{A}}^{*}\cap {x}^{{\delta}^{*}}{L}^{2}$
.
The proofs rely on the construction of a pseudodifferential parametrix
$G$
for
$L$
, depending on
$\delta $
. This is an element in the calculus of pseudodifferential edge operators
${\Psi}_{e}^{*}\left(M\right)$
. We do not define this calculus here, but remark only that these operators are described by specifying the precise asymptotic behaviour of their Schwartz kernels, near the diagonal and also near the boundaries and corner of
$M\times M$
. We refer as before to [18] . (We should note also that the results stated here are slightly more general than what is written explicitly in that source because we are allowing the possibility of variable indicial roots outside the critical interval; however, these can be derived easily from the same techniques.)
4.2 Edge analysis of
${D}_{a}$
We now proceed to apply the methods of the last subsection to the analysis of
${D}_{a}$
.
The first tasks are the calculation of the indicial roots of
${D}_{a}$
and the analysis of the normal operator
$N\left(x{D}_{a}\right)$
.
4.2.1 Indicial roots of
${D}_{a}$
For simplicity, first consider the calculation of the indicial roots for
${D}_{a}$
for the metric
$g=d{x}^{2}+{x}^{2}\kappa $
on the cone
${C}_{1}\left(F\right)$
. To do this, we decompose this operator as much as possible. Thus, first regard
${D}_{a}$
as a
$2\times 2$
matrix acting on pairs
$(\alpha ,\beta )\leftrightarrow \alpha +dx\wedge \beta $
, where
$\alpha \left(x\right),\beta \left(x\right)\in {\Omega}^{*}\left(F\right)$
for each
$x$
; normalize by writing the
$k$
form part of
$\alpha $
as
${x}^{k}{\alpha}_{k}$
, and similarly for
$\beta $
. A short calculation then shows that, acting on pairs
$({\alpha}_{k},{\beta}_{k})$
,
$${I\left(x{D}_{a}\right)}_{{\Omega}^{k}\oplus {\Omega}^{k}}=\left(\begin{array}{cc}\frac{1}{x}{D}_{F}& {\partial}_{x}\frac{fk2a}{x}\\ {\partial}_{x}+\frac{k}{x}& \frac{1}{x}{D}_{F}\end{array}\right);$$
the full indicial operator is the direct sum over
$k$
of these matrices. Similarly, the indicial family is the direct sum of matrices
$$\begin{array}{c}I\left(x{D}_{a}{)}_{k}\right(\gamma )=\left(\begin{array}{cc}{D}_{F}& \gamma (fk+2a)\\ \gamma +k& {D}_{F}\end{array}\right)\end{array}$$ 
(17)

This can be reduced further using the eigendecomposition for
${\Delta}_{F}$
. In particular, we see that the operator in ( 17 ) is noninvertible if and only if for some eigenvalue
${\lambda}^{2}$
for
${\Delta}_{F}$
,
$$\left(\begin{array}{cc}\lambda & \gamma (fk2a)\\ \gamma +k& \lambda \end{array}\right)$$
is singular, or equivalently
$${\gamma}^{2}+(f2a)\gamma +k(fk2a){\lambda}^{2}=0.$$
Hence the indicial roots come in pairs:
$$\begin{array}{c}{\gamma}_{\lambda ,k}^{\pm}=a\frac{f}{2}\pm \frac{1}{2}{\left[(f2a2k{)}^{2}+4{\lambda}^{2}\right]}^{1/2}.\end{array}$$ 
(18)

The extension of these calculations to incomplete edge metrics requires only some mild alterations. Write
$g=d{x}^{2}+\stackrel{~}{g}\left(x\right)$
, where
$\stackrel{~}{g}\left(x\right)=h+{x}^{2}\kappa $
is a degenerating family of metrics on
$Y=\partial M$
. When
$\alpha $
is a
$(p,q)$
form on
$Y$
, its pointwise norm satisfies
$$\alpha {}_{\stackrel{~}{g}\left(x\right)}={x}^{q}\alpha {}_{\stackrel{~}{g}\left(1\right)}.$$
Denote by
${D}_{Y}^{x}$
the operator
$D$
on
$Y$
for the metric
${G}_{x}$
. By Proposition 3, with respect to the metric
$\stackrel{~}{g}\left(1\right)$
on
$Y$
,
$${d}_{Y}={d}_{F}+{\stackrel{~}{d}}_{B}\mathbb{I}\mathbb{I}+R,{\delta}_{Y}={\delta}_{F}+({\stackrel{~}{d}}_{B}{)}^{*}\mathbb{I}{\mathbb{I}}^{*}+{R}^{*};$$
hence,
${D}_{Y}={D}_{F}+{\overline{D}}_{B}+\overline{R}$
where
$${\overline{D}}_{B}={\stackrel{~}{d}}_{B}+({\stackrel{~}{d}}_{B}{)}^{*}\mathbb{I}\mathbb{I}\mathbb{I}{\mathbb{I}}^{*},\text{and}\overline{R}=R+{R}^{*}.$$
A quick review of the definitions shows that
$${D}_{Y}^{x}=\frac{1}{x}{D}_{F}+{\overline{D}}_{B}+x\overline{R},$$
where all the components on the right are the corresponding operators at
$x=1$
.
Now let
$Z\to B$
be the bundle with fibre
${C}_{1}\left(F\right)$
obtained from
$Y\to B$
, with metric
$g=d{x}^{2}+\stackrel{~}{g}\left(x\right)$
. Decompose any form
$\omega =\alpha +dx\wedge \beta $
on
$Z$
as
$\alpha =\sum {x}^{k}{\alpha}_{k}$
, where
${\alpha}_{k}$
is of type
$(*,k)$
on
$Y$
, and similarly for
$\beta $
. Thus the pointwise norms satisfy
$\omega {}_{g}^{2}=\sum \left({\alpha}_{k}{}_{\stackrel{~}{g}\left(1\right)}^{2}+{\beta}_{k}{}_{\stackrel{~}{g}\left(1\right)}^{2}\right)$
. In terms of these decompositions and normalizations, the restriction of the operator
${D}_{a}$
to pairs of
$(*,k)$
forms on
$Z$
is given by
$$\begin{array}{c}{D}_{a}=\left(\begin{array}{cc}\frac{1}{x}{D}_{F}+{\overline{D}}_{B}+x\overline{R}& {\partial}_{x}\frac{fk2a}{x}\\ {\partial}_{x}+\frac{k}{x}& \left(\frac{1}{x}{D}_{F}+{\overline{D}}_{B}+x\overline{R}\right)\end{array}\right).\end{array}$$ 
(19)

From this expression, we see that neither
${\overline{D}}_{B}$
nor
$\overline{R}$
appear in the indicial operator
$I\left(x{D}_{a}\right)$
. Hence the computation of the indicial roots is exactly the same as in the conic case; in other words, all indicial roots are of the form ( 18 ). Note, however, that the eigenvalues
${\lambda}^{2}$
may depend on
$b\in B$
, hence the same may be true of these indicial roots.
Notice that if
$\omega \in {\Omega}^{k}$
and
$\omega {}_{g}\sim {x}^{\gamma}$
, then
$\omega \in {x}^{a}{L}^{2}{\Omega}^{k}$
(near
$x=0$
) if and only if
$\gamma >af/2$
. The indicial roots which lie near to (and above) this `
${x}^{a}{L}^{2}$
cutoff ' are the ones which cause the difference between minimal and maximal domains. We explain this later, but for now record the
Corollary 6.
The indicial roots of the operator
${D}_{a}$
contained in the interval
$(a(f+1)/2,a(f1)/2)$
correspond to the eigenvalues
${\lambda}^{2}$
of
${\Delta}_{F}$
on
$k$
forms such that
$(f2a+2k{)}^{2}+4{\lambda}^{2}<1$
. In order for this condition to be nonvacuous, it is necessary that
$k\in \left(\right(f1)/2a,(f+1)/2a)$
. Note in particular that
${\gamma}_{0,k}^{\pm}=af/2\pm (f/2ak)=k,k+2af\in (a(f+1)/2,a(f1)/2)$
precisely when
$k\in \left(\right(f1)/2a,(f+1)/2a)$
and
${H}^{k}\left(F\right)\ne 0$
.
4.2.2 The normal operator of
${D}_{a}$
According to the discussion in the final paragraphs of §4.1, we must also study the mapping properties of the normal operator
$N\left(x{D}_{a}\right)$
. Before doing so, we address some `duality' issues. The main point is that
${D}_{a}$
is formally symmetric on
${x}^{a}{L}^{2}{\Omega}^{*}$
, hence the adjoint of the (closed range) operator
$$\begin{array}{c}{D}_{a}:{x}^{a}{L}^{2}{\Omega}^{*}\u27f6{x}^{a1}{L}^{2}{\Omega}^{*}\end{array}$$ 
(20)

is identified with
$$\begin{array}{c}{D}_{a}:{x}^{a+1}{L}^{2}{\Omega}^{*}\u27f6{x}^{a}{L}^{2}{\Omega}^{*}.\end{array}$$ 
(21)

Recall also that, according to the computations of the preceding subsection, there are at most two indicial roots
${\gamma}_{0}^{\pm}$
in the interval
$(a(f+1)/2,a(f1)/2)$
, and these are symmetric around the midpoint
$af/2$
. We shall assume that the metric
$g$
is such that
${\Delta}_{F}$
has no small nonzero eigenvalues, so that no other indicial roots intersect the closed interval
$[a(f+1)/2,a(f1)/2]$
.
By (
19 ),
$$\begin{array}{c}N\left(x{D}_{a}\right)=s{D}_{C\left(F\right),a}+s{D}_{{\mathbb{R}}^{b}},\end{array}$$ 
(22)

where the first operator on the right is the analogous weighted operator on the complete cone
$C\left(F\right)$
and the second is on Euclidean space. In fact, the identification of
${s}^{1}N\left(x{D}_{a}\right)$
with
${D}_{C\left(F\right),a}+{D}_{{\mathbb{R}}^{b}}$
may also be seen by naturality, since the operator on the left must equal, at
$b\in B$
, the Hodgede Rham operator on
${\mathbb{R}}_{s}^{+}\times {\mathbb{R}}_{u}^{b}\times F$
with respect to
${g}_{b}=d{s}^{2}+{s}^{2}du{}^{2}+{\kappa}_{b}$
.
Proposition 10.
$$N\left(x{D}_{a}\right):{s}^{a+1}{L}^{2}{\Omega}^{*}\u27f6{s}^{a+1}{L}^{2}{\Omega}^{*}$$
is injective, and hence ( 21 ) has a finite dimensional nullspace consisting of conormal forms

Proof.
Suppose that
$\omega \in {s}^{\alpha}{L}^{2}{\Omega}^{*}$
is in the nullspace of this operator. Take the Fourier transform in the
$u$
direction; denoting the dual variable by
$\eta $
, then whenever
$\eta \ne 0$
we can rescale, setting
$t=s\left\eta \right$
,
$\hat{\eta}=\eta /\left\eta \right$
. Then
$$\left({D}_{C\left(F\right),a}+i\text{cl}\left(\hat{\eta}\right)\right)\hat{\omega}=0,$$
where
$\text{cl}\left(\hat{\eta}\right)$
is Clifford multiplication
$\hat{\eta}\wedge \cdot +\iota \left(\hat{\eta}\right)\cdot $
. Apply
${D}_{C\left(F\right),a}+i\text{cl}\left(\hat{\eta}\right)$
to this equation to deduce that
$$\left({\Delta}_{C\left(F\right),a}+1\right)\hat{\omega}=0.$$
It is not hard to show, cf. [18] , that any solution of this equation either grows or decays exponentially as
$t\to \infty $
, and the
${L}^{2}$
hypothesis prohibits the former. Furthermore, solutions are polyhomogeneous as
$t\to 0$
, and hence decay at some indicial weight
${t}^{\gamma}$
with
$\gamma >a(f1)/2$
. Hence both
$N\left(d\right)\hat{\omega}$
and
$N\left({\delta}_{a}\right)\hat{\omega}$
decay like
${t}^{\gamma 1}$
, and in particular are still in
${t}^{a}{L}^{2}$
. This means we can integrate by parts to obtain
$$0=\langle ({\Delta}_{C\left(F\right),a}+1)\hat{\omega},\hat{\omega}{\rangle}_{a}=\left\rightN\left(d\right)\hat{\omega}{}_{a}^{2}+\leftN\right({\delta}_{a}\left)\hat{\omega}\right{}_{a}^{2}+\left\right\hat{\omega}{}_{a}^{2};$$
all boundary terms vanish. This gives
$\hat{\omega}=0$
, as desired.
When
$\eta =0$
, the problem reduces to showing that
${\Delta}_{C\left(F\right),a}$
has no nullspace in
${t}^{a+1}{L}^{2}$
on the entire cone
$C\left(F\right)$
, which is even more easily verified to be true (e.g. by separation of variables).
Following the discussion from the end of §4.1, if
$(a(f+1)/2,a(f1)/2)$
contains no indicial roots for
${D}_{a}$
,
$N\left(x{D}_{a}\right)$
is injective on
${s}^{a}{L}^{2}{\Omega}^{*}$
.
There are no forms in the nullspace of the normal operator which lie in
${t}^{a}{L}^{2}{\Omega}^{*}\cap \mathcal{D}\left({d}_{max,a}\right)\cap \mathcal{D}\left({\delta}_{min,a}\right)$
or
${t}^{a}{L}^{2}{\Omega}^{*}\cap \mathcal{D}\left({d}_{min,a}\right)\cap \mathcal{D}\left({\delta}_{max,a}\right)$
. From this it is possible to show that
$\omega \in {\mathcal{\mathscr{H}}}_{abs/rel}^{*}(M,g,a)\subset {\mathcal{A}}^{*}{\Omega}^{*}$
, i.e. such harmonic forms are conormal. Unfortunately, these results rely on a slightly more elaborate parametrix construction than is available in the literature, so at present we are only asserting this informally. At the end of the next subsection, however, we show that forms in
${\mathcal{\mathscr{H}}}_{min}^{*}(M,g,a)$
are conormal.
4.3 Closed extensions of
${D}_{a}$
It is a general fact that closed extensions of
$d$
on
${x}^{a}{L}^{2}{\Omega}^{*}$
are in bijective correspondence with the selfadjoint extensions of
${D}_{a}$
on this space. To see this, first note that if
$\overline{d}$
is any closed extension of
$d$
, and if
${\overline{{d}_{a}}}^{*}$
is its Hilbert space adjoint, then
$\overline{{D}_{a}}=\overline{d}+{\overline{{d}_{a}}}^{*}$
is a selfadjoint extension of
${D}_{a}$
. Conversely, any selfadjoint extension of
${D}_{a}$
determines an associated closed extension for
$d$
, cf. [
2,Lemma2.3]
. We summarize this in the
Proposition 11.
If
$d$
has more than one closed extension on
${x}^{a}{L}^{2}{\Omega}^{*}(M,g)$
, then
${D}_{a}$
has more than one selfadjoint extension on this space; equivalently, if
${D}_{a}$
is essentially selfadjoint, then
${d}_{max,a}={d}_{min,a}$
.
Notice that since
$d$
always has closed extensions,
${D}_{a}$
always has selfadjoint extensions.
However, if
${D}_{a}$
is not essentially selfadjoint, then it will have closed extensions which are not necessarily selfadjoint; the relationship between these and the closed extensions of
$d$
is somewhat more complicated, and we shall not attempt to describe it.
Theorem 7.
The symmetric operator
${D}_{a}$
is essentially selfadjoint on
${x}^{a}{L}^{2}{\Omega}^{*}(M,g)$
if and only if there exists no indicial root for
${D}_{a}$
in the interval
$(a(f+1)/2,a(f1)/2)$
. As explained earlier, this is equivalent to the nonexistence of small eigenvalues
${\lambda}^{2}$
for
${\Delta}_{F}$
on
$k$
forms such that
$k\in (f/2a\frac{1}{2}\sqrt{14{\lambda}^{2}}),f/2a+\frac{1}{2}\sqrt{14{\lambda}^{2}})$
. In this case
$d$
also has a unique closed extension.

Proof.
Assume that there are no small eigenvalues, as described in the statement of the theorem. Fix a parametrix
$G$
for
${D}_{a}$
relative to the space
${x}^{a}{L}^{2}{\Omega}^{*}$
. This is an element of order
$1$
in the edge calculus. If
$\omega \in \mathcal{D}\left({D}_{max,a}\right)$
, then
$f={D}_{a}\omega \in {x}^{a}{L}^{2}{\Omega}^{*}(M,g)$
, and applying
$G$
gives that in fact
$\omega \in {x}^{a+1}{H}_{e}^{1}{\Omega}^{*}$
. We recall that in general,
$Gf$
would be the sum of two terms, the first corresponding to these small indicial roots and the second an error term in
${x}^{a+1}{H}_{e}^{1}{\Omega}^{*}$
, but by our hypothesis, the former of these is absent. It is now straightforward to check that
$\omega $
may be smoothly approximated in the
${D}_{a}$
graph norm, i.e. that there exists a sequence of smooth compactly supported forms
${\phi}_{j}$
such that
${\phi}_{j}\to \omega $
,
${D}_{a}{\phi}_{j}\to {D}_{a}\omega $
in
${x}^{a}{L}^{2}{\Omega}^{*}$
. This shows that
$\omega \in \mathcal{D}\left({D}_{min,a}\right)$
.
Conversely, if there do exist indicial roots in the critical range, then these may be used to construct nontrivial elements in
$\mathcal{D}\left({D}_{max,a}\right)\backslash \mathcal{D}\left({D}_{min,a}\right)$
, and by the general abstract theory, there will be more than one selfadjoint extension of
${D}_{a}$
. Because this is not central to our discussion, we leave details to the reader (and refer to [
10]
for a thorough discussion of the conic case).
In the conic case (when
$a=0$
) this result is due to Cheeger [3] ; cf. also [16] . The analysis needed in that case is simpler than the edge analysis used here, though this is not apparent `on the surface'.
We turn now to a description of
$\mathcal{D}\left({D}_{max,a}\right)$
in the more general case where this vanishing condition is no longer satisfied. By definition, this domain is the set of all
$\omega \in {x}^{a}{L}^{2}{\Omega}^{*}$
such that
${D}_{a}\omega \in {x}^{a}{L}^{2}{\Omega}^{*}$
. We have already remarked that the mapping ( 20 ) has closed range, which we denote
${\mathcal{\mathcal{R}}}_{a}$
. Its cokernel is finite dimensional since, by Proposition ( 10 ), the adjoint mapping ( 21 ) has a finite dimensional nullspace. Hence we can choose a generalized inverse
$${G}_{a}:{x}^{a1}{L}^{2}{\Omega}^{*}(M,g)\u27f6{x}^{a}{L}^{2}{\Omega}^{*}(M,g);$$
this is a pseudodifferential edge operator of order
$1$
which satisfies
${D}_{a}{G}_{a}=IP$
on
${x}^{a1}{L}^{2}{\Omega}^{*}$
, where
$P$
is the orthogonal projector onto the cokernel. The condition
$f\in {\mathcal{\mathcal{R}}}_{a}$
is equivalent to
$\langle f,\gamma \rangle =0$
for all
$\gamma $
in the nullspace of ( 21 ). Notice also that
${\mathcal{\mathcal{R}}}_{a}\cap {x}^{a}{L}^{2}{\Omega}^{*}$
is dense in
${\mathcal{\mathcal{R}}}_{a}$
. Therefore,
$$\mathcal{D}\left({D}_{max,a}\right)={G}_{a}\left({\mathcal{\mathcal{R}}}_{a}\cap {x}^{a}{L}^{2}{\Omega}^{*}\right)+{\mathcal{N}}_{a},$$
where
${\mathcal{N}}_{a}$
is the nullspace of ( 20 ).
Finally, if
$\gamma \in {\mathcal{\mathscr{H}}}_{min}^{*}(M,g,a)$
, then
$\langle {D}_{max,a}\eta ,\gamma \rangle =0$
for all
$\eta \in \mathcal{D}\left({D}_{max,a}\right)$
, i.e.
$\langle f,\gamma \rangle =0$
for all
$f\in {\mathcal{\mathcal{R}}}_{a}\cap {x}^{a}{L}^{2}{\Omega}^{*}$
. Using the density statement above, this shows that
$\gamma $
lies in the nullspace of ( 21 ), and is thus conormal.
We note in conclusion that any
$\omega \in \mathcal{D}\left({D}_{max,a}\right)$
has a `weak' asymptotic expansion of the form
$$\omega \sim {\omega}_{0}^{}(y,z){x}^{{\gamma}_{0}^{}}+{\omega}_{0}^{+}(y,z){x}^{{\gamma}_{0}^{+}}+{\omega}^{\prime}$$
where
${\omega}^{\prime}=\mathcal{O}\left({x}^{a+1}\right)$
, again in a suitable weak sense. We refer to [
18]
for more details.
4.4 The minimal Hodge cohomology
We are now in a position to prove the
Theorem 8.
Let
$M$
be a manifold with an incomplete edge metric
$g$
. Then
$${\mathcal{\mathscr{H}}}_{min}^{k}(M,g,a)=\{\begin{array}{ccccc}\text{Im}(I{H}_{\underline{\mathfrak{m}}+<a>}^{k}(X,B)& \to & I{H}_{\overline{\mathfrak{m}}+\ll a1\gg}^{k}(X,B))& & f\text{odd}\\ \text{Im}(I{H}_{\underline{\mathfrak{m}}+<a1/2>}^{k}(X,B)& \to & I{H}_{\overline{\mathfrak{m}}+\ll a1/2\gg}^{k}(X,B))& & f\text{even}.\end{array}$$
In particular, when
$a=0$
,
$${\mathcal{\mathscr{H}}}_{min}^{k}(M,g,0)=\text{Im}\left(I{H}_{\underline{\mathfrak{m}}}^{k}\right(X,B)\u27f6I{H}_{\overline{\mathfrak{m}}}^{k}(X,B\left)\right).$$

Proof.
Recalling that
${\mathcal{\mathscr{H}}}_{min}^{*}$
is quasiisometry invariant, we may as well assume that
${\Delta}_{F}$
has no small nonzero eigenvalues, in the sense of the preceding subsections. If there are no small eigenvalues at all, i.e. either when
$(a(f+1)/2,a(f1)/2)\cap \mathbb{N}=\varnothing $
(which holds, for example, when
$a=0$
and
$f$
is odd), or else if there exists
${q}_{a}\in (a(f+1)/2,a(f1)/2)$
but
${H}^{{q}_{a}}\left(F\right)=\left\{0\right\}$
, then the result follows directly from what we have already done, since then
${H}_{max/min}^{k}(M,g,a)$
and and
${\mathcal{\mathscr{H}}}_{max/min/abs/rel}^{k}(M,g,a)$
are all equal, cf. Corollary 3 . Thus we suppose that there exists
${q}_{a}\in (a(f+1)/2,a(f1)/2)$
such that
${H}^{{q}_{a}}l\left(F\right)\ne \left\{0\right\}$
.
According to Theorem
4 , the space appearing on the right in the statement of this theorem, for
$f$
even or odd, is identified with
$\text{Im}\left({H}_{min}^{k}(M,g,a)\to {H}_{max}^{k}(M,g,a)\right)$
. For simplicity, we denote it as
${\mathcal{J}}^{k}(M,g,a)$
.
We claim first that there is a natural injective map
$${\mathcal{\mathscr{H}}}_{min}^{k}(M,g,a)\u27f6{\mathcal{J}}^{k}(M,g,a).$$
To see this, recall that any form
$\omega \in {\mathcal{\mathscr{H}}}_{min}^{k}(M,g,a)={\mathcal{\mathscr{H}}}_{abs}^{k}(M,g,a)\cap {\mathcal{\mathscr{H}}}_{rel}^{k}(M,g,a)$
naturally represents a class in
${H}_{min}^{k}(M,g,a)$
. If
$\left[\omega \right]=0$
in
${H}_{max}^{k}(M,g,a)$
, then
$\omega =d\zeta $
for some
$\zeta \in \mathcal{D}\left({d}_{max,a}\right)$
. But this would imply that
$\left\right\omega {}_{a}^{2}=\langle \omega ,d\zeta {\rangle}_{a}=0$
since
$\omega \in ker\left({\delta}_{min,a}\right)$
.
This proves the claim.
The issue, then, is to prove that any class
$\left[\eta \right]\in {\mathcal{J}}^{k}(M,g,a)$
is represented by an element of
${\mathcal{\mathscr{H}}}_{min}^{k}(M,g,a)$
. Choose a representative
$\eta \in {\mathcal{A}}^{\gamma}{\Omega}^{k}$
,
$\gamma >a(f1)/2$
, for this class. Now use a generalized inverse
$G$
for
${D}_{a}:{x}^{a}{L}^{2}{\Omega}^{*}(M,g)\to {x}^{a1}{L}^{2}{\Omega}^{*}(M,g)$
(acting on its maximal domain). According to Propositions 8 and 9, this gives
$\zeta \in {x}^{a}{L}^{2}{\Omega}^{*}\cap {\mathcal{A}}^{{\gamma}_{0}^{}}$
and an element of the cokernel, which by duality corresponds to an element
$\omega \in ker{D}_{a}\cap {x}^{a+1}{L}^{2}{\Omega}^{*}={\mathcal{\mathscr{H}}}_{min}^{k}(M,g,a)$
, such that
$\eta ={D}_{a}\zeta +\omega $
. (Recall that
${\gamma}_{0}^{}$
is the lower of the two indicial roots in the critical interval.) The theorem will follow once we show that
${\delta}_{a}\zeta =0$
. Now
$$\left\right{\delta}_{a}\zeta {}_{a}^{2}=\langle {\delta}_{a}\zeta ,\eta d\zeta \omega {\rangle}_{a}.$$
Integrating by parts formally this should vanish, so it remains to show that each integration by parts is valid. First,
$\langle {\delta}_{a}\zeta ,\eta {\rangle}_{a}=0$
since both terms are conormal and
$d\eta =0$
. Similarly
$\langle {\delta}_{a}\zeta ,\omega {\rangle}_{a}=0$
since
$\omega \in ker{d}_{min,a}$
. To show that the remaining term vanishes, observe that
$\zeta ={x}^{{\gamma}_{0}^{}}{\zeta}_{0}^{}+{x}^{{\gamma}_{0}^{+}}{\zeta}_{0}^{+}+{\zeta}^{\prime}$
where
${\zeta}^{\prime}\in {\mathcal{A}}^{\gamma}{\Omega}^{*}$
,
$\gamma >a(f1)/2$
. Writing
${\zeta}_{0}^{\pm}={\mu}_{0}^{\pm}+dx\wedge {\nu}_{0}^{\pm}$
, then
${\mu}_{0}^{\pm},{\nu}_{0}^{\pm}\in {\mathcal{C}}^{\infty}$
and in the nullspace of
${\Delta}_{F}$
. A closer inspection of the equation
${D}_{a}\zeta =\eta \omega $
shows that
$$\mathfrak{d}{\mu}_{0}^{\pm}=\mathfrak{d}{\nu}_{0}^{\pm}={\mathfrak{d}}^{*}{\mu}_{0}^{\pm}={\mathfrak{d}}^{*}{\nu}_{0}^{\pm}=0.$$
(The operators
$\mathfrak{d}$
,
${\mathfrak{d}}^{*}$
are the differential and codifferential for the (fibreharmonic) projected complex for the Riemannian submersion metric
$(Y,\stackrel{~}{g}(x\left)\right)$
.) In any case, the identity
$\langle d\zeta ,{\delta}_{a}\zeta {\rangle}_{a}=0$
is now immediate. This proves the remaining assertion, and hence the theorem.
5 Hodge theory for complete edge metrics
We are also able to determine the dimensions of the spaces of the spaces
${L}^{2}{\mathcal{\mathscr{H}}}^{k}(M,g)$
when
$(M,g)$
is a manifold with a complete edge metric. Unlike the incomplete case, in certain degrees this Hodge cohomology may be infinite dimensional, i.e. there is an infinite dimensional space of
${L}^{2}$
harmonic forms. The simplest example of this is when
$(M,g)$
is the
$n$
dimensional hyperbolic space, or indeed any conformally compact manifold, and
$k=n/2$
, cf. [
20]
.
Theorem 9.
Let
$({M}^{n},g)$
be a manifold with a complete edge metric. Let
$X$
be the compact stratified space defined in §3. Suppose that
$k$
is not of the form
$j+(b+1)/2$
where
${\mathcal{\mathscr{H}}}^{j}\left(F\right)\ne \left\{0\right\}$
. Then
$${L}^{2}{\mathcal{\mathscr{H}}}^{k}(M,g)\sim =I{H}_{f+\frac{b}{2}k}^{k}(X,B).$$
In this case, the
${L}^{2}$
signature theorem for
$M$
is the same as the
${L}^{2}$
signature theorem for
$M$
endowed with the conformally equivalent incomplete edge metric
${x}^{2}g$
. In all other cases, where
$k$
does have this form,
${L}^{2}{\mathcal{\mathscr{H}}}^{k}(M,g)$
is infinite dimensional.

Proof.
There are several viable ways to proceed: one could use a parametrix construction based on the edge calculus to do a global Hodge theoretic argument as in [20] ; one could also, as in the incomplete case, use sheaf theory, calculations of local cohomologies, etc., ab initio; we take a shorter intermediate route, reducing to the incomplete edge case using the conformal invariance of the space of middle degree
${L}^{2}$
harmonic forms.
It suffices to consider forms of degree
$k\ge n/2$
. Define
$k=(n+r)/2$
with
$0\le r\le n$
.
For later use, we also set
$\sigma =(f+r)/2$
,
${\sigma}^{\prime}=(fr)/2$
. Now, since
$2k=n+r$
,
$k$
is the middle degree on the manifold
$\stackrel{~}{M}=M\times {S}^{r}$
. Endow
$\stackrel{~}{M}$
with the product metric
${G}_{r}=g+{\kappa}_{r}$
(where
${\kappa}_{r}$
is the standard metric on
${S}^{r}$
); in a neighbourhood of
$\partial \stackrel{~}{M}$
$${G}_{r}=\frac{d{x}^{2}+h}{{x}^{2}}+(\kappa +{\kappa}_{r}).$$
In other words,
${G}_{r}$
is still a complete edge metric with the same base
$(B,h)$
as
$g$
, but with fibre
$(F\times {S}^{r},\kappa +{\kappa}_{r})$
. There is a corresponding incomplete edge metric
$\hat{g}={x}^{2}\stackrel{~}{g}$
on
$\stackrel{~}{M}$
. We denote its compactification, obtained by pinching the fibres
$F\times {S}^{r}$
at the boundary, by
$\widehat{X}$
.
We first claim that
$$\begin{array}{c}{L}^{2}{\mathcal{\mathscr{H}}}^{k}(\stackrel{~}{M},\stackrel{~}{g})={\mathcal{\mathscr{H}}}_{max}^{k}(\stackrel{~}{M},\hat{g},0).\end{array}$$ 
(23)

The verification is straightforward; by conformal invariance of the
${L}^{2}$
condition and the operator
$\delta =\pm *d*$
in the middle degree, harmonic forms on the complete manifold are in the maximal domains of
$d$
and
$\delta $
(and indeed their nullspaces) on the incomplete manifold, and conversely.
By the
${L}^{2}$
Künneth theorem,
$${L}^{2}{\mathcal{\mathscr{H}}}^{k}(\stackrel{~}{M},\stackrel{~}{g})={L}^{2}{\mathcal{\mathscr{H}}}^{k}(M,g)\oplus {L}^{2}{\mathcal{\mathscr{H}}}^{kr}(M,g).$$
On the other hand, while we have not shown how to compute
${\mathcal{\mathscr{H}}}_{max}^{k}(\stackrel{~}{M},\hat{g},0)$
in general, and indeed have noted that it is sometimes infinite dimensional, we now show that under certain hypotheses, it equals
${\mathcal{\mathscr{H}}}_{min}^{k}(\stackrel{~}{M},\hat{g},0)$
; thus we can then apply Theorem 3 to calculate the right side of ( 23 ) as
$I{H}_{\mathfrak{m}}^{k}\left(\stackrel{~}{X}\right)$
. (Under these hypotheses, the intersection cohomologies with upper or lower middle perversity are the same, so we just write
$\mathfrak{m}$
.) So, let us suppose that either
$b$
is even, or else if
$b$
is odd then
${H}^{\sigma}\left(F\right)$
(and hence
${H}^{{\sigma}^{\prime}}\left(F\right)$
) is trivial; note this last condition is automatic when
$r>f$
. We claim that under these conditions, the minimal and maximal de Rham complexes on
$\stackrel{~}{M}$
coincide, and thus
${\mathcal{\mathscr{H}}}_{max}^{k}(\stackrel{~}{M},\hat{g},0)={\mathcal{\mathscr{H}}}_{min}^{k}(\stackrel{~}{M},\hat{g},0)$
. Indeed, this follows directly from Corollary 3 : we have
$dim\stackrel{~}{M}=\stackrel{~}{n}=2k$
even; if
$b$
is even, then
$\stackrel{~}{f}=f+r=\stackrel{~}{n}b1$
is odd, while if
$b$
is odd, then
$${H}^{\stackrel{~}{f}/2}(F\times {S}^{r})={H}^{\sigma}\left(F\right)\oplus {H}^{{\sigma}^{\prime}}\left(F\right)=\left\{0\right\}.$$
Taking these facts together, and assuming this vanishing of the fibre cohomology when
$b$
is odd, we have proved that
$$\begin{array}{c}{L}^{2}{\mathcal{\mathscr{H}}}^{k}(M,g)\oplus {L}^{2}{\mathcal{\mathscr{H}}}^{kr}(M,g)=I{H}_{\mathfrak{m}}^{k}\left(\stackrel{~}{X}\right).\end{array}$$ 
(24)

It remains to compute the final term on the right of ( 24 ). We decompose
$$\stackrel{~}{X}=M\times {S}^{r}{\bigsqcup}_{Y\times {S}^{r}}\stackrel{~}{Z},$$
where
$\stackrel{~}{Z}$
is the cone bundle over
$B$
with fibre
$F\times {S}^{r}$
and boundary
$Y\times {S}^{r}$
. Of course,
$${H}^{k}(M\times {S}^{r})={H}^{k}\left(M\right)\oplus {H}^{kr}\left(M\right).$$
On the other hand,
$$I{H}_{\mathfrak{m}}^{*}\left(\stackrel{~}{Z}\right)=I{H}_{\mathfrak{m}}^{*}\left(Z\right)\oplus I{H}_{\mathfrak{m}}^{*r}\left(Z\right).$$
To prove this, we return to the sheaftheoretic description. For product neighbourhoods
$\mathcal{U}=\mathcal{V}\times {C}_{1}(F\times {S}^{r})$
, we have
$$I{H}_{\mathfrak{m}}^{j}(\mathcal{U};\mathcal{\mathcal{L}})=I{H}_{\mathfrak{m}}^{j}\left({C}_{1}\right(F\times {S}^{r}\left)\right)=\{\begin{array}{ccc}{H}^{j}\left(F\right)\oplus {H}^{jr}\left(F\right),& & j\le (f+r1)/2\\ 0& & \text{otherwise.}\end{array}$$
Note that we have combined the conditional inequality in this last step, which should depend on the parity of
$f+r$
, into one condition. This condition is correct as stated if
$f+r$
is odd; if
$f+r$
is even, then one would expect the condition
$j\le (f+r)/21=\sigma 1$
or
$j\le (f+r)/2=\sigma $
, depending on whether one was using upper or lower middle perversity. However, the hypothesis
${H}^{\sigma}\left(F\right)=0$
guarantees that we get the same result in either case. Thus we see that the spectral sequence whose hypercohomology computes the intersection cohomology of
$\stackrel{~}{Z}$
decouples into two noninteracting pieces.
We have now proved that for a fixed
$k$
, assuming the hypotheses above,
$${L}^{2}{\mathcal{\mathscr{H}}}^{k}\left(M\right)\oplus {L}^{2}{\mathcal{\mathscr{H}}}^{kr}\left(M\right)=I{H}_{\mathfrak{m}}^{k}\left(X\right)\oplus I{H}_{\mathfrak{m}}^{kr}\left(X\right).$$
We would, of course, like the summands to be equal separately; this can be seen simply by noting that the correspondence takes place on the level of forms, and we can separate out the terms with like degree. It remains to show that in the remaining cases, i.e. when
$b$
is odd and
${H}^{\sigma}\left(F\right)\ne \left\{0\right\}$
,
$\sigma =(2kb1)/2$
,
${L}^{2}{\mathcal{\mathscr{H}}}^{k}\left(M\right)$
is infinite dimensional. This follows from two assertions:
that
$0$
is in the essential spectrum of
${\Delta}_{k}$
, and that there is a spectral gap at
$0$
for
${\Delta}_{k}$
. We content ourselves with sketching the proofs briefly. The first step relies on the observation that up to quasiisometry, some neighbourhood of infinity looks like the product of half of a hyperbolic space and a compact manifold, specifically
${\mathbb{H}}_{+}^{b+1}\times F$
. Since
$b+1$
is even, there is an infinite dimensional family of
${L}^{2}$
harmonic forms on the first factor. Since
$k=(b+1)/2+\sigma $
, we can take suitable truncations of these, coupled with harmonic forms of degree
$\sigma $
on
$F$
, to produce a Weyl sequence on
$M$
. For the second step, we construct a parametrix in the edge calculus for
${\Delta}_{k}$
. Its normal operator is given by
$$N\left({\Delta}_{k}\right)={\sum}_{j=0}^{k}N\left({\Delta}_{{\mathbb{H}}^{b+1},j}\right)+{\Delta}_{F,kj}.$$
Both of these operators have spectral gaps at zero (even when
$j=(b+1)/2$
), so it is possible to construct a parametrix for
${\Delta}_{k}\lambda $
with compact remainder when
$\lambda $
is small but nonzero. We refer to [
17]
and [
20]
for more complete descriptions of such proofs in a slightly simpler context.
References

A. Borel et al., Intersection Cohomology, Birkhauser (1994).

J. Brüning and M. Lesch, Hilbert Complexes, J. Func. Anal. 108, 88132, (1992).

J. Cheeger, On the spectral geometry of spaces with conelike singularities, Proc. Nat. Acad. Sci. 76 (1979), 21032106.

J. Cheeger, On the Hodge theory of Riemannian pseudomanifolds in Geometry of the Laplace operator Proc. Symp. Pure Math Vol. XXXVI, Amer. Math. Soc., Providence, (1980),91146.

J. Cheeger and X. Dai,
${L}^{2}$
cohomology of a nonisolated conical singularity and nonmultiplicativity of the signature. Preprint.

J. Cheeger, M. Goresky and R. McPherson,
${L}^{2}$
cohomology and intersection homology of singular algebraic varieties in Seminar on Differential Geometry, Annals of Mathematical Studies 102, Princeton Univ. Press, Princeton, N.J., (1982), 303–340.

X. Dai, Adiabatic limits, nonmultiplicativity of signature, and the Leray spectral sequence, J. Amer. Math. Soc. 4 (1991), no. 2, 265–321.

G. deRham, Differentiable manifolds. Grundlehren der Mathematischen Wissenschaften, 266, SpringerVerlag, Berlin, (1984).

M. P. Gaffney, A special Stokes's theorem for complete Riemannian manifolds, Ann. Math. 60 no. 2, (1954), 140–145.

J. Gil and G. Mendoza, Adjoints of elliptic cone operators, Amer. J. Math 125 (2003), 357–408.

M. Goresky and R. MacPherson, Intersection homology theory, Topology, 19 (1980), no. 2, 135–162.

M. Goresky and R. MacPherson, Intersection homology II, Invent. Math. 72 no. 1 (1983), 77–129.

T. Hausel, E. Hunsicker, and R. Mazzeo, Hodge cohomology of gravitational instantons. Duke Math. J. 122 no. 3 (2004), 485–548.

C. Hodgsen, S. Kerckhoff Rigidity of hyperbolic conemanifolds and hyperbolic Dehn surgery, J. Diff. Geom. 48 (1998), 1–59.

E. Hunsicker, Hodge and signature theorems for a class of manifolds with fibration boundary, in preparation.

M. Lesch, Differential Operators of Fuchs Type, Conical Singularities, and Asymptotic Methods. Teubner Texte für Mathematik 136, Teubner–Verlag, Leipig (1997).

R. Mazzeo, The Hodge cohomology of a conformally compact metric, J. Diff. Geom. 28 (1988), 309–339.

R. Mazzeo Elliptic theory of differential edge operators I, Comm. PDE. 16 (1991), no. 10, 1616–1664.

R. Mazzeo and R. Melrose, Pseudodifferential operators on manifolds with fibred boundaries in “Mikio Sato: a great Japanese mathematician of the twentieth century.”, Asian J. Math. 2 (1998), no. 4, 833–866.

R. Mazzeo and R. S. Phillips, Hodge theory on hyperbolic manifolds, Duke Math. J. 60 (1990), no. 2, 509–559.

R. Mazzeo and H. Weiss, The Teichmüller theory of conic surfaces, in preparation.

G. Montcouquiol, Rigidité infinitésimale de cônesvariétés Einstein à courbure négative. arXiv:math.DG/0503195

L. Saper
${L}^{2}$
cohomology of locally symmetric spaces I, arXiv:math.RT/0412353

L. Saper On the cohomology of locally symmetric spaces and their compactifications, in Current developments in mathematics 2002, Eds. D. Jerison et al., International Press (2003), 219289.