
A detailed numerical study of the long time behaviour of dispersive shock
waves in solutions to the KadomtsevPetviashvili (KP) I equation is presented.
It is shown that modulated lump solutions emerge from the dispersive shock
waves. For the description of dispersive shock waves, Whitham modulation
equations for KP are obtained. It is shown that the modulation equations near
the soliton line are hyperbolic for the KPII equation while they are elliptic
for the KPI equation leading to a focusing effect and the formation of lumps.
Such a behaviour is similar to the appearance of breathers for the focusing
nonlinear Schrodinger equation in the semiclassical limit.

Temperature and field dependent $^1$H, $^{19}$F, and $^{79,81}$Br NMR
measurements together with zero  field $^{79,81}$BrNQR measurements on
polycrystalline samples of barlowite, Cu$_4$(OH)$_6$FBr are conducted to study
the magnetism and possible structural distortions on a microscopic level. The
temperature dependence of the $^{79,81}$Br NMR spinlattice relaxation rates
1/$T_1$ indicate a phase transition at $T_{\rm N}\simeq$15 K which is of
magnetic origin, but with an unusually weak slowing down of fluctuations below
$T_{\rm N}$. Moreover, 1/$T_1T$ scales linear with the bulk susceptibility
which indicates persisting spin fluctuations down to 2 K. Quadupolare resonance
(NQR) studies reveal a pair of zerofield NQR lines associated with the two
isotopes of Br with the nuclear spins of $I$ = 3/2. Quadrupole coupling
constants of $\nu_Q\simeq$ 28.5~MHz and 24.7~MHz for $^{79}$Br and $^{81}$Br
nuclei are determined from BrNMR and the asymmetry parameter of the electric
field gradient was estimated to $\eta \simeq 0.2$. The BrNQR lines are
consistent with our findings from BrNMR and they are relatively broad, even
above $T_{\rm N}$. This broadening and the relative large $\eta $ value
suggests a symmetry reduction at the Br site reflecting the presence of a
local distortion in the lattice. Our densityfunctional calculations show that
the displacements of Cu2 atoms located between the kagome planes do not account
for this relatively large $\eta$. On the other hand, full structural
relaxation, including the deformation of kagome planes, leads to a better
agreement with the experiment.

We present a detailed numerical study of various blowup issues in the
context of the focusing DaveyStewartson II equation. To this end we study
Gaussian initial data and perturbations of the lump and the explicit blowup
solution due to Ozawa. Based on the numerical results it is conjectured that
the blowup in all cases is self similar, and that the time dependent scaling
is as in the Ozawa solution and not as in the stable blowup of standard
$L^{2}$ critical nonlinear Schr\"odinger equations. The blowup profile is
given by a dynamically rescaled lump.

The defocusing DaveyStewartson II equation has been shown in numerical
experiments to exhibit behavior in the semiclassical limit that qualitatively
resembles that of its onedimensional reduction, the defocusing nonlinear
Schr\"odinger equation, namely the generation from smooth initial data of
regular rapid oscillations occupying domains of spacetime that become
welldefined in the limit. As a first step to study this problem analytically
using the inversescattering transform, we consider the direct spectral
transform for the defocusing DaveyStewartson II equation for smooth initial
data in the semiclassical limit. The direct spectral transform involves a
singularlyperturbed elliptic Dirac system in two dimensions. We introduce a
WKBtype method for this problem, prove that it makes sense formally for
sufficiently large values of the spectral parameter $k$ by controlling the
solution of an associated nonlinear eikonal problem, and we give numerical
evidence that the method is accurate for such $k$ in the semiclassical limit.
Producing this evidence requires both the numerical solution of the
singularlyperturbed Dirac system and the numerical solution of the eikonal
problem. The former is carried out using a method previously developed by two
of the authors and we give in this paper a new method for the numerical
solution of the eikonal problem valid for sufficiently large $k$. For a
particular potential we are able to solve the eikonal problem in closed form
for all $k$, a calculation that yields some insight into the failure of the WKB
method for smaller values of $k$. Informed by numerical calculations of the
direct spectral transform we then begin a study of the singularlyperturbed
Dirac system for values of $k$ so small that there is no global solution of the
eikonal problem.

The aim of this paper is to study, via theoretical analysis and numerical
simulations, the dynamics of Whitham and related equations. In particular we
establish rigorous bounds between solutions of the Whitham and KdV equations
and provide some insights into the dynamics of the Whitham equation in
different regimes, some of them being outside the range of validity of the
Whitham equation as a water waves model.

An important step in the efficient computation of multidimensional theta
functions is the construction of appropriate symplectic transformations for a
given Riemann matrix assuring a rapid convergence of the theta series. An
algorithm is presented to approximately map the Riemann matrix to the Siegel
fundamental domain. The shortest vector of the lattice generated by the Riemann
matrix is identified exactly, and the algorithm ensures that its length is
larger than $\sqrt{3}/2$. The approach is based on a previous algorithm by
Deconinck et al. using the LLL algorithm for lattice reductions. Here, the LLL
algorithm is replaced by exact Minkowski reductions for small genus and an
exact identification of the shortest lattice vector for larger values of the
genus.

We study numerically the evolution of perturbed Kortewegde Vries solitons
and of well localized initial data by the NovikovVeselov (NV) equation at
different levels of the "energy" parameter $ E $. We show that as $ E \to
\infty $, NV behaves, as expected, similarly to its formal limit, the
KadomtsevPetviashvili equation. However at intermediate regimes, i.e. when $ 
E  $ is not very large, more varied scenarios are possible, in particular,
blowups are observed. The mechanism of the blowup is studied.

The dispersionless KadomtsevPetviashvili (dKP) equation
$(u_t+uu_x)_x=u_{yy}$ is one of the simplest nonlinear wave equations
describing twodimensional shocks. To solve the dKP equation we use a
coordinate transformation inspired by the method of characteristics for the
onedimensional Hopf equation $u_t+uu_x=0$. We show numerically that the
solutions to the transformed equation do not develop shocks. This permits us to
extend the dKP solution as the graph of a multivalued function beyond the
critical time when the gradients blow up. This overturned solution is
multivalued in a lip shape region in the $(x,y)$ plane, where the solution of
the dKP equation exists in a weak sense only, and a shock front develops. A
local expansion reveals the universal scaling structure of the shock, which
after a suitable change of coordinates corresponds to a generic cusp
catastrophe. We provide a heuristic derivation of the shock front position near
the critical point for the solution of the dKP equation, and study the solution
of the dKP equation when a small amount of dissipation is added. Using
multiplescale analysis, we show that in the limit of small dissipation and
near the critical point of the dKP solution, the solution of the dissipative
dKP equation converges to a Pearcey integral. We test and illustrate our
results by detailed comparisons with numerical simulations of both the
regularized equation, the dKP equation, and the asymptotic description given in
terms of the Pearcey integral.

The Kerr solution in coordinates corotating with the horizon is studied as a
testbed for a spacetime with a helical Killing vector in the Ernst picture. The
solution is numerically constructed by solving the Ernst equation with a
spectral method and a Newton iteration. We discuss convergence of the iteration
for several initial iterates and different values of the Kerr parameters.

A purely numerical approach to compact Riemann surfaces starting from plane
algebraic curves is presented. The critical points of the algebraic curve are
computed via a twodimensional Newton iteration. The starting values for this
iteration are obtained from the resultants with respect to both coordinates of
the algebraic curve and a suitable pairing of their zeros. A set of generators
of the fundamental group for the complement of these critical points in the
complex plane is constructed from circles around these points and connecting
lines obtained from a minimal spanning tree. The monodromies are computed by
solving the defining equation of the algebraic curve on collocation points
along these contours and by analytically continuing the roots. The collocation
points are chosen to correspond to Chebychev collocation points for an ensuing
ClenshawCurtis integration of the holomorphic differentials which gives the
periods of the Riemann surface with spectral accuracy. At the singularities of
the algebraic curve, Puiseux expansions computed by contour integration on the
circles around the singularities are used to identify the holomorphic
differentials. The Abel map is also computed with the ClenshawCurtis algorithm
and contour integrals. As an application of the code, solutions to the
KadomtsevPetviashvili equation are computed on nonhyperelliptic Riemann
surfaces.

We present the first numerical approach to Dbar problems having spectral
convergence for real analytic rapidly decreasing potentials. The proposed
method starts from a formulation of the problem in terms of an integral
equation which is solved with Fourier techniques. The singular integrand is
regularized analytically. The resulting integral equation is approximated via a
discrete system which is solved with Krylov methods. As an example, the Dbar
problem for the DaveyStewartson II equations is solved. The result is used to
test direct numerical solutions of the PDE.

The Peregrine breather is widely discussed as a model for rogue waves in deep
water. We present here a detailed numerical study of perturbations of the
Peregrine breather as a solution to the nonlinear Schr\"odinger (NLS)
equations. We first address the modulational instability of the constant
modulus solution to NLS. Then we study numerically localized and nonlocalized
perturbations of the Peregrine breather in the linear and fully nonlinear
setting. It is shown that the solution is unstable against all considered
perturbations.

For equation P$_I^2$, the second member in the P$_I$ hierarchy, we prove
existence of various degenerate solutions depending on the complex parameter
$t$ and evaluate the asymptotics in the complex $x$ plane for $x\to\infty$
and $t=o(x^{2/3})$. Using this result, we identify the most degenerate
solutions $u^{(m)}(x,t)$, $\hat u^{(m)}(x,t)$, $m=0,...,6$, called {\em
tritronqu\'ee}, describe the quasilinear Stokes phenomenon and find the large
$n$ asymptotics of the coefficients in a formal expansion of these solutions.
We supplement our findings by a numerical study of the tritronqu\'ee solutions.

We present a detailed numerical study of solutions to general Kortewegde
Vries equations with critical and supercritical nonlinearity. We study the
stability of solitons and show that they are unstable against being radiated
away and blowup. In the $L_{2}$ critical case, the blowup mechanism by
Martel, Merle and Rapha\"el can be numerically identified. In the limit of
small dispersion, it is shown that a dispersive shock always appears before an
eventual blowup. In the latter case, always the first soliton to appear will
blow up. It is shown that the same type of blowup as for the perturbations of
the soliton can be observed which indicates that the theory by Martel, Merle
and Rapha\"el is also applicable to initial data with a mass much larger than
the soliton mass. We study the scaling of the blowup time $t^{*}$ in
dependence of the small dispersion parameter $\epsilon$ and find an exponential
dependence $t^{*}(\epsilon)$ and that there is a minimal blowup time
$t^{*}_{0}$ greater than the critical time of the corresponding Hopf solution
for $\epsilon\to0$. To study the cases with blowup in detail, we apply the
first dynamic rescaling for generalized Kortewegde Vries equations. This
allows to identify the type of the singularity.

A multidomain spectral method with compactified exterior domains combined
with stable second and fourth order time integrators is presented for
Schr\"odinger equations. The numerical approach allows high precision numerical
studies of solutions on the whole real line. At examples for the linear and
cubic nonlinear Schr\"odinger equation, this code is compared to transparent
boundary conditions and perfectly matched layers approaches. The code can deal
with asymptotically non vanishing solutions as the Peregrine breather being
discussed as a model for rogue waves. It is shown that the Peregrine breather
can be numerically propagated with essentially machine precision, and that
localized perturbations of this solution can be studied.

We survey and compare, mainly in the twodimensional case, various results
obtained by IST and PDE techniques for integrable equations. We also comment on
what can be predicted from integrable equations on non integrable ones.

We present a computational approach to general hyperelliptic Riemann surfaces
in Weierstrass normal form. The surface is either given by a list of the branch
points, the coefficients of the defining polynomial or a system of cuts for the
curve. A canonical basis of the homology is introduced algorithmically for this
curve. The periods of the holomorphic differentials and the Abel map are
computed with the ClenshawCurtis method in order to achieve spectral accuracy.
The code can handle almost degenerate Riemann surfaces. This work generalizes
previous work on real hyperelliptic surfaces with prescribed cuts to arbitrary
hyperelliptic surfaces. As an example, solutions to the sineGordon equation in
terms of multidimensional theta functions are studied, also in the solitonic
limit of these solutions.

We provide a numerical study of various issues pertaining to the dynamics of
the DaveyStewartson systems of the DS II type. In particular we investigate
whether or not the properties (blowup, radiation,...) displayed by the
focusing and defocusing DS II integrable systems persist in the non integrable
case.

We present the first detailed numerical study of the semiclassical limit of
the DaveyStewartson II equations both for the focusing and the defocusing
variant. We concentrate on rapidly decreasing initial data with a single hump.
The formal limit of these equations for vanishing semiclassical parameter
$\epsilon$, the semiclassical equations, are numerically integrated up to the
formation of a shock. The use of parallelized algorithms allows to determine
the critical time $t_{c}$ and the critical solution for these $2+1$dimensional
shocks. It is shown that the solutions generically break in isolated points
similarly to the case of the $1+1$dimensional cubic nonlinear Schr\"odinger
equation, i.e., cubic singularities in the defocusing case and square root
singularities in the focusing case. For small values of $\epsilon$, the full
DaveyStewartson II equations are integrated for the same initial data up to
the critical time $t_{c}$. The scaling in $\epsilon$ of the difference between
these solutions is found to be the same as in the $1+1$ dimensional case,
proportional to $\epsilon^{2/7}$ for the defocusing case and proportional to
$\epsilon^{2/5}$ in the focusing case. We document the DaveyStewartson II
solutions for small $\epsilon$ for times much larger than the critical time
$t_{c}$. It is shown that zones of rapid modulated oscillations are formed near
the shocks of the solutions to the semiclassical equations. For smaller
$\epsilon$, the oscillatory zones become smaller and more sharply delimited to
lens shaped regions. Rapid oscillations are also found in the focusing case for
initial data where the singularities of the solution to the semiclassical
equations do not coincide.

Using a Fourier spectral method, we provide a detailed numerically
investigation of dispersive Schr\"odinger type equations involving a fractional
Laplacian. By an appropriate choice of the dispersive exponent, both mass and
energy sub and supercritical regimes can be computed in one spatial dimension,
only. This allows us to study the possibility of finite time blowup versus
global existence, the nature of the blowup, the stability and instability of
nonlinear ground states, and the long time dynamics of solutions. The latter is
also studied in a semiclassical setting. Moreover, we numerically construct
ground state solutions to the fractional nonlinear Schr\"odinger equation.

We present the first detailed numerical study of the Toda equations in $2+1$
dimensions in the limit of long wavelengths, both for the hyperbolic and
elliptic case. We first study the formal dispersionless limit of the Toda
equations and solve initial value problems for the resulting system up to the
point of gradient catastrophe.
It is shown that the breakup of the solution in the hyperbolic case is
similar to the shock formation in the Hopf equation, a $1+1$ dimensional
singularity. In the elliptic case, it is found that the breakup is given by a
cusp as for the semiclassical system of the focusing nonlinear Schr\"odinger
equation in $1+1$ dimensions. The full Toda system is then studied for finite
small values of the dispersion parameter $\epsilon$ in the vicinity of the
shocks of the dispersionless Toda equations. We determine the scaling in
$\epsilon$ of the difference between the Toda solution for small $\epsilon$ and
the singular solution of the dispersionless Toda system. In the hyperbolic
case, the same scaling proportional to $\epsilon^{2/7}$ is found as in the
small dispersion limit of the Kortewegde Vries and the defocusing nonlinear
Schr\"odinger equations. In the elliptic case, we obtain the same scaling
proportional to $\epsilon^{2/5}$ as in the semiclassical limit for the focusing
nonlinear Schr\"odinger equation. We also study the formation of dispersive
shocks for times much larger than the breakup time in the hyperbolic case. In
the elliptic case, an $L_{\infty}$ blowup is observed instead of a dispersive
shock for finite times greater than the breakup time. The
$\epsilon$dependence of the blowup time is determined.

We provide a detailed numerical study of various issues pertaining to the
dynamics of the Burgers equation perturbed by a weak dispersive term: blowup
in finite time versus global existence, nature of the blowup, existence for
"long" times, and the decomposition of the initial data into solitary waves
plus radiation. We numerically construct solitons for fractionary Kortewegde
Vries equations.

We study the critical behaviour of solutions to weakly dispersive Hamiltonian
systems considered as perturbations of elliptic and hyperbolic systems of
hydrodynamic type with two components. We argue that near the critical point of
gradient catastrophe of the dispersionless system, the solutions to a suitable
initial value problem for the perturbed equations are approximately described
by particular solutions to the Painlev\'eI (P$_I$) equation or its fourth
order analogue P$_I^2$. As concrete examples we discuss nonlinear Schr\"odinger
equations in the semiclassical limit. A numerical study of these cases provides
strong evidence in support of the conjecture.

We present a numerical study of solutions to the generalized
KadomtsevPetviashvili equations with critical and supercritical nonlinearity
for localized initial data with a single minimum and single maximum. In the
cases with blowup, we use a dynamic rescaling to identify the type of the
singularity. We present a discussion of the observed blowup scenarios.

There exist two versions of the KadomtsevPetviashvili equation, related to
the Cartesian and cylindrical geometries of the waves. In this paper we derive
and study a new version, related to the elliptic cylindrical geometry. The
derivation is given in the context of surface waves, but the derived equation
is a universal integrable model applicable to generic weaklynonlinear
weaklydispersive waves. We also show that there exist nontrivial
transformations between all three versions of the KP equation associated with
the physical problem formulation, and use them to obtain new classes of
approximate solutions for water waves.