Binning data inside irregular 2d shape

I have three vectors of data, two representing x-y space (tempV and precV) and one representing z direction (acV). I also have an irregular shape and I want to bin all z data inside the shape (into more than one bin). I also have a logical vector for the datapoints that fall inside the shape. I want to plot the shape and the binned z data as bars. I have tried a number of options but am stuck because my bins in some cases fall outside the shape as the outline is not a rectangular. I'm guessing I need to change the x (P1) and y (A1) location of the z (ac1) data when I plot it, but havent figured out a good way to do so (like bin centers or similar). Below is what I have so far. In the attached figure, the five greenish bars that are closest to the x and y axes are in the wrong location, I want to force them to locations inside the same area as the other four (i.e., the are vaguely delineated by the black lines). Thanks.

load(data.mat)

%plot the outline of the shape
figure
plot(xv_1,yv_1) % polygon

%bin data and find bin edges
h1 = histogram2(tempV(in_1),precV(in_1),3);

p1bin = discretize(precV(in_1),h1.YBinEdges);
t1bin = discretize(tempV(in_1),h1.XBinEdges);
[P1,A1]=ndgrid(h1.XBinEdges(1:end-1),h1.YBinEdges(1:end-1));
ac1 = accumarray([t1bin,p1bin],acV(in_1),[],@sum);

cm1 = [0.650 0.612 0.192];
figure
scatterbar3(P1,A1,ac1,1) %function from file exchange
colormap(cm1)

[Image Analyst]

Attach screenshot please. Use the green and blue frame icon to insert your image. Don't attach a .fig file.

[Elin Jacobs]

Edited above.