Takes a numpy array of bathymetry points from a grid ( exportable from a raster in ArcGIS using ArcPy ) and writes a tab separated xyz file, for import into gis or in this case it was written for importing bathymetry as scatter data into DHI MIKE21. Requires a hdr file for width, height, origin and cell size information.

grid_bathy_xyz.py

# Initial data obtained from hdr file produced with flt file when running 
# Raster to Float in ArcToolBox and numpy file ( npy ) from RasterToNumpyArray 
# in ArcPy

# Numpy uses top left as origin and geopgraphic coordinate systems treat the 
# origin as lower left, you can either flip the incoming bathymetry maxtrix
# vertically or create a descending range of y_coordinates

import numpy as np
bathy = np.load('D:\\numpy_bathy.npy')
#bathy = np.flipud(bathy)
cell_size = 3
no_of_columns = 18325
no_of_rows = 13303
lower_left_x = 607512
lower_left_y = 6455598
lower_right_x = lower_left_x + ( cell_size * no_of_columns )
x_coordinates = np.arange(lower_left_x, lower_right_x, cell_size)
upper_left_y = lower_left_y + ( cell_size * no_of_rows )
y_coordinates = np.arange(upper_left_y, lower_left_y, -1 * cell_size)
xyz_file = open('D:\\bathymetry_output2.xyz','w')
for index_y,y in enumerate(y_coordinates):
    print index_y
    for index,x in enumerate(x_coordinates):
        depth = bathy[index_y,index]
        if depth > -1000.0 :
            xyz_file.write(str(x) + '\t' + str(y) + '\t' + str(depth) + '\n')
xyz_file.close()