How to interpolate a 3d surface with the different size of squares?

I have a surface between a range -5 to 5 on both x and y axis. I would like to interpolate this surface with a new meshgrid data which has different size for each iteration. However, I cannot get the entire surface with interpolated new data. I only get the different size of squares. Is there a way to interpolate the original surface with different grid spacing ? Here is my code as well. Figure 1 is my original surface. and Figure 4 is interpolated surface which is missing the original surface itself. What I would like to have is Figure 4 in Figure 1.

L1 = 4;
L2 = 3;
L1 = L1;
[xP,yP] = meshgrid(n1:n2);
    zP  = real( L3*sqrt(1-((xP.^2)/ (L1^2)) - ((yP.^2)./ (L2^2))));
    
 for i =1:length(CoordsX(:,1))
    sqx=[min(CoordsX(i,:)), max(CoordsX(i,:))] ;
    sqy=[min(CoordsY(i,:)), max(CoordsY(i,:))];
    [x,y,z] =rotation(180-alpha,xP,yP,zP);
    [Xq,Yq] = meshgrid(sqx,sqy);
    Vq = interp2(xP,yP,z+z0_el,Xq,Yq,'cubic');
    surf(Xq,Yq,Vq)
    hold on
    daspect([1 1 1])
 end

[KSSV]

n1, n2 is not defined.

[Nilly]

sorry. n1 = -5 , n2= 5

z0_el = 6;

[Asad (Mehrzad) Khoddam]

The lower an upper limits of sqx and sqy are less than the original limits of xP and yP. So, you miss some data points.