In this section, we recall some definitions and preliminary results.
  2.1. Approximation Tools
Now, let 
 be a rectangle and 
 denote the space of the continuous functions on 
R. As usual, 
 can be equipped with the uniform norm 
. From now on, the notations 
 and 
 will denote the function 
 as depending on the only variable 
z or 
s, respectively. For smoother functions, we introduce the following Sobolev-type space
        
        where the superscript 
 denotes the 
 derivative of the one-dimensional function 
 or 
, 
 and 
. We equip 
 with the norm
        
Now, let 
 be the space of bivariate algebraic polynomials of total degree 
n and 
 be the space of bivariate polynomials of degree 
n in each variable. Obviously, 
. We denote by 
 and 
 the errors of best polynomial approximation on 
R for bivariate continuous functions by means of polynomials in 
 and 
, respectively, i.e.,
        
From the definitions, it follows that
        
In [
7], the following Favard-type inequality was proved when 
:
        and therefore, due to the linearity, the same inequality holds true for any general rectangle 
R.
		Moreover, by (
2), the same estimate can be stated also for 
, i.e.,
        
  2.2. An Algebraic Cubature Formula on Curvilinear Polygons
In order to describe the cubature formula, which is the base of the proposed Nyström method, we recall some definitions and properties of a planar parametric curve (see [
11]).
We consider 
simple closed curves, i.e., curves parametrized on 
        for which 
 is continuous and injective on 
 and 
, with 
. Moreover, we assume that 
 is 
piecewise , i.e., there is at most a finite number of 
breakpoints , where 
 (
 is considered a breakpoint if 
).
The space 
 of piecewise 
 parametric curves on the partition of 
, generated by a fixed set of “parameter breakpoints” 
, is endowed with the norm
        
        where 
, for 
 is piecewise continuous.
		Let a 
singular point be a point 
 such that 
 or 
, and we define a 
cusp as a breakpoint 
, such that 
, for some 
 (which means, in particular, that the left and right tangents at the point have opposite directions). We say that a curve in 
 is 
generalized regular if it has no singular points and no cusps.
Now, let 
 be a compact domain (the closure of a bounded and simply connected open set), whose boundary is a Jordan piecewise polynomial parametric curve 
, 
, given counter-clockwise by a sequence of polynomial parametric curves 
, with 
, defined in the interval 
, and “breakpoints”
        
        with 
 (i.e., 
). Furthermore, let 
 be the minimal rectangle containing 
H, 
, 
, 
, and 
 are the nodes and weights of the Gauss–Legendre rule in 
. The cubature formula proposed in [
10] for the integral of the bivariate function 
f on the domain 
H, i.e., 
, is defined as follows:
        with
        
        where
        
In [
10], it was proved that this formula is exact for 
, and stable, since it results in
        
We underline that, as discussed in [
10], in general, the nodes of this cubature formula are not contained in the domain 
H, but only in the minimal rectangle 
 with sides parallel to the axes.
However, the nodes are inside the cubature domain when there exists a straight line l such that the following property  holds:
When property 
 holds, a change in coordinates so that 
l becomes parallel to the new 
y-axis implies that all the cubature nodes 
 are in 
H, and that all the weights 
 are nonnegative, taking as 
 (which defines the nodes of the cubature formula) the intersection point of 
l with the new 
x-axis. For a detailed discussion on this, the interested reader can consult [
10] and the references therein.
Go back now to a general domain 
, and assume that it is a compact domain whose boundary is a Jordan (simple and closed) curve, defined parametrically by two piecewise smooth functions 
 that are not polynomials
        
In [
10], it was suggested to use the Chebfun package [
12] in order to provide two piecewise polynomials 
 (interpolating at Chebychev–Lobatto nodes), such that
        
        where 
 is the machine precision and 
 denotes the ordinary maximum norm on 
. Denoting by 
 the domain whose boundary is defined as
        
        in [
10], it was proposed to approximate
        
        with 
, and then approximate 
 with 
.
Obviously, in order to construct the cubature for 
, it is necessary that 
 is still a Jordan curve; that is, essentially a 
simple curve. There exist sufficient conditions (see, for instance, [
11,
13]) such that is reasonable to assume that 
Chebfun constructs a Jordan curve if the original boundary of 
 is piecewise 
 and a generalized regular curve, in the sense that it has no singular points and no cusps.
Concerning the convergence of the cubature rule, the following results were proven in [
10]. For any subset 
 of 
, we denote by 
 the error of best polynomial approximation of the function 
f by means of bivariate polynomials of total degree 
m, with respect to the uniform norm on 
. Moreover, let 
 denote the minimal rectangle containing 
.
Theorem 1 ([
10], Theorem 3)
. Let Ω 
be a compact domain in , whose boundary (9) is a Jordan curve, and let  be the domain with the boundary, as in (11). Assume that ,  are at least piecewise Hölder continuous and satisfy (10). The following cubature error estimate holds for Formula (7) with where ,  and , in the general case, while  when Ω 
satisfies Property , and  denotes the closed disk centered at the origin with radius . In the case when the approximated boundary is not guaranteed to be simple, as for example if the original curve has some singular points, the following error estimate holds true.
Theorem 2 ([
10], Theorem 5)
. Let Ω 
be as in (9), and , be the piecewise polynomial approximating curve (10). Assume that the integrand f is Hölder-continuous with constant C and exponent  on the minimal rectangle  containing , where . Then, the following estimate holds for the error of the cubature formula (7) with :where .