Working With Elevation Data – A DSM Perspective

How DSM Data Can Help You


Elevation models are something that we are seeing more and more of these days that are enhancing our decision making ability. Many of our clients are leading the way in using their eBee to acquire powerful, accurate 3 dimensional data-sets, like never before. But what exactly kind of data are we actually looking at?

Figure 1: DSM of a bare ground canola field. Note the mounds on the right side are caused by a tree line, or surface element, and not a terrain element like a small hill.

There are three common GIS elevation model types that are referenced in many modern applications Digital Elevation Models (DEM), Digital Terrain Models (DTM), and Digital Surface Models (DSM). You may find there are varying schools of thought as to what the differences are that truly separate a DTM from a DEM (even not all GIS specialists don’t agree), but thankfully our main focus will be on what makes up a DSM.

A DSM is very useful for 3D modelling. It is often used in urban planning, aviation, telecommunications where viewing above ground obstructions, vegetation, buildings etc. is needed. A DSM is created from a dense point cloud generated in several possible ways including that generated in the the post processing stages after a mission with your eBee/Albris. The point cloud is comprised of a large number of georeferenced points in a 3Dimension coordinate system, with each point defined by an x, y and z coordinate. As we are working with UAV’s, the most common resulting elevation model is DSM. The DSM focuses on the surface elements above bare ground and any terrain features

Why You might need to know this?

Understanding the type and quality of data you need for your project important to the r final results. Let’s look at a field drainage project that requires a data-t that is based on the physical terrain as opposed to the items above it, so a DEM or DTM is ideal. Inversely, a DSM elevation model data set is best when looking at projects that depend on surface elements, i.e. biomass calculation, volumetric calculation or building reconstruction.

Working with Imperfect Data

Sometimes there are work arounds that allow you to use one elevation model data set like DSM data, but create a terrain map (DTM) that is accurate enough for your project Let’s use the drainage example again. Alternatively to using a DEM or DTM, you could potentially use a DSM to help along your project. With the proper filtering of outliers or unwanted surface features (buildings, trees, etc.), we can create a bare ground mosaic that is quite similar a DTM/DEM of the site.

Pix4D has the power to help

High quality data-sets like DSM’s are becoming more and more attainable. The challenge going forward will be “how can we collect it…” to “how can we use it…”, and sometimes it just takes a slight change in analytics and time to work with your data to get you the results you need.


DSM prior to truck removal


DSM after truck removal

Stay tuned as we explore how to work with imperfect data to create a DSM through Pix4D in upcoming blogs.