Mapping the forest/tree line in Norway with ICESat-2 space laser data
With climate change, Norway's forests are moving to higher elevations and latitudes - and thus altering local ecosystems. To understand the potential implications this has and to model further changes, we need to know at what elevations the forest/tree line is now.
Traditionally, human observations are used to determine the forest/tree line, such as with the ongoing citizen science project "Natur i endring" (https://www.naturiendring.no/). Using data from the new space laser satellite ICESat-2, it should be possible to automatically detect the forest line and maybe even the tree line in Norway.
First results from the American Geoscience annual meeting (AGU 2018) show that ICESat-2's elevation profiles are very well able to distinguish between canopy/vegetation height and ground surface: https://fallmeeting.agu.org/2018/files/2018/12/AguICESat2_1210_Final.pdf
The aim of this master thesis is to automatically detect the forest line, or possibly even tree line, in Norway by processing and analysing ICESat-2 laser altimetry data from multiple satellite overpasses. We anticipate detecting "tree line points" for all individual elevation profiles by analysing the canopy height along with the profile.
The so-generated tree line points can be verified by fieldwork and/or compared to existing forest/tree line data from the "Natur i endring"- database which contains over 700 human observations across entire Norway.
Primary research questions are:
- Is it possible to use ICESat-2 data for tree line detection - even in steep mountainous terrain as in Norway?
- Are there differences between tree line measurements from automated detections and human observations? E.g., in terms of elevation (for example, a systematic bias because ICESat-2 does not "hit" the highest trees) or spatial distribution (for example, clumping of human observations in easily accessible regions)?
About the ICESat-2 laser system that records 6 parallel elevation profiles:
A description of ICESat-2 vegetation applications: Neuenschwander&Pitts, 2019, Remote Sensing of Environments:
Student desirable background
Remote sensing, ecology, physical geography or similar, good knowledge of remote sensing/GIS methods, scripting/programming skills.
Recommended master courses: GEO4520