Combining rasters with rasterio

I've recently been using LiDAR derived Digital Terrain Models, available from DataMapWales as 1km x 1km tile downloads. Although this keeps the file size small, being able to work with a single raster file is easier.

We can use Rasterio's merge method to combine the rasters into a single image.

First, install the requirements into a Python environment:

$ pip install rasterio

Create an iterable of the raster file paths and pass it to the merge method; this will return a tuple with two elements. The first, dest, is the raster in the form of a numpy ndarray, and the second, out_transform is an affine transformation that we will add to the Geotiff's metadata when writing to the file system.

from pathlib import Path

import rasterio

path = Path(".")
dest, output_transform = rasterio.merge(list(path.glob("**/*.tif")))

Next, get a copy of the metadata from one of the original raster files, and update the height, width and transform values:

with rasterio.open("original1.tif") as src:
    out_meta = src.meta.copy()
    out_meta.update(
        {
            "driver": "GTiff",
            "height": dest.shape[1],
            "width": dest.shape[2],
            "transform": output_transform
        }
    )

And then simply write the new, merged raster to a tif file:

with rasterio.open("merged.tif", "w", **out_meta) as fp:
    fp.write(dest)

The whole process looks like this:

from pathlib import Path

import rasterio

path = Path(".")
dest, output_transform = rasterio.merge(list(path.glob("**/*.tif")))

with rasterio.open("original1.tif") as src:
    out_meta = src.meta.copy()
    out_meta.update(
        {
            "driver": "GTiff",
            "height": dest.shape[1],
            "width": dest.shape[2],
            "transform": output_transform
        }
    )

with rasterio.open("merged.tif", "w", **out_meta) as fp:
    fp.write(dest)