The Digital Forestry stream takes people who already understand forests and gives them the analytical toolkit — machine learning, computer vision, advanced geospatial analysis — to turn remotely sensed data into decisions.
Applicants arriving from a computer science or geomatics background will be assessed individually. The forestry context can be built; what matters is that the analytical work has somewhere real to land. We will have an intake in Fall 2027 designed just for you. Contact foremprograms@unb.ca to be added to the mailing list.
If you have a BScF or comparable degree, the only additional requirement is basic working proficiency in Python — enough to read a script, write a loop, and manipulate a data frame.
You do not need a computer science degree, and you do not need to secure a supervisor before you apply.
Twenty-six credit hours of coursework across the Fall and Winter terms, with the capstone carrying through to the following summer — organized so each new method has somewhere to land, first in the geospatial fundamentals, then in the machine learning and computer vision that run on top of them.
Advanced GIS in ArcGIS Pro applied to real inventory and landscape problems on the UNB Woodlot; LiDAR point cloud processing through to enhanced forest inventory; active and passive sensor theory and practice.
Supervised and unsupervised methods, neural networks and deep learning, and the workflow design skills to move a model from notebook to something that runs on real data.
Image matching, segmentation, classification and object detection, then applied deep learning — transfer learning, autoencoders, and the ethical questions that come with automated interpretation.
Forestry & Environmental Management, Engineering – Geodesy & Geomatics, and Computer Science each contribute courses across Fall and Winter term.
FOR*6996 isn't a course you take at the end. You register in it every term, from your first week, carrying a single land-based digital forestry project through from question to defensible answer — applying each new method as you acquire it.
Graduates leave with a finished piece of applied work, on a real land base rather than a teaching dataset, that they can put in front of an employer.
Derive enhanced forest inventory attributes from airborne laser scanning.
Generate and interpret terrain and canopy surfaces from point cloud data.
Train, validate and critically evaluate machine learning and deep learning models on imagery and inventory data.
Design analytical workflows that hold up on operational-scale data rather than a teaching subset.
The forestry foundation is what makes those skills useful. A BScF graduate knows what a plausible stand structure looks like, how an inventory attribute will be used in a management plan, and when a modelled result is telling them something real rather than reproducing an artefact of the data.
That judgment is the part a purely technical background does not supply — and it's what employers doing enhanced inventory, monitoring and geospatial analysis are short of.
Applications go through UNB's School of Graduate Studies. When you apply, you'll be selecting the program:
To expedite your application for the first cohort, apply under the non-degree category instead:
Then email foremprograms@unb.ca to let us know. We'll work through the rest of the process with you, including the move into the Master of Forestry.