Modelling and Linking Radar Backscatter and Stem Water Content in a Boreal Forest
Licentiatavhandling, 2025

In the context of climate change, forests serve as crucial mitigators but are also increasingly threatened by disturbances such as wildfires, insect outbreaks, and windthrows. Monitoring forest vitality is essential, with tree water status serving as an important indicator of forest health.
Spaceborne synthetic aperture radar (SAR), known for its sensitivity to vegetation water content (VWC), offers a promising tool for providing forest VWC on a global scale with high spatial resolution. However, the processes linking SAR backscatter to forest VWC are not fully understood. This thesis addresses this gap by establishing a link between radar backscatter and stem water content in boreal forests, using data from two tower-based radars, BorealScat and BorealScat-2, as an intermediary platform.
In Paper I, for the first time, the relevance of tower-based radar observations for spaceborne SAR was examined. This was done by comparison of time series of BorealScat radar backscatter and backscatter observed by Sentinel-1 and ALOS-2 PALSAR-2 at C- and L-band, respectively. A good agreement was found for a healthy forest, with an RMSE of 0.9 dB and 0.6 dB at C-band and  L-band, respectively. At C-band, tree death was visible in tomographic tower-based backscatter but not in Sentinel-1 or BorealScat full-forest backscatter, indicating the potential value of tomographic spaceborne SAR missions for forest vitality monitoring.
In Paper II, for the first time, forest backscatter was linked to high quality stem water content measurements. A modified version of the water cloud model was used to relate stem water content time series to canopy backscatter and attenuation time series observed at L-band by BorealScat-2. Modelled stem water content closely matched in situ stem water content with a relative RMSE of 4% and 2% for the inversion of the canopy backscatter model and the inversion of the attenuation model, respectively. These findings demonstrate the feasibility of using L-band SAR backscatter to monitor VWC and the value of tomography in future spaceborne SAR missions.

C-band

stem water content

vegetation water content

Remote sensing

L-band

tomography

forest

synthetic aperture radar

HA1, Hörsalsvägen 4
Opponent: Dr. Heather Reese, Department of Earth Sciences, University of Gothenburg, Sweden

Författare

Theresa Leistner

Chalmers, Rymd-, geo- och miljövetenskap, Geovetenskap och fjärranalys

COMPARISON OF TOWER-BASED AND SATELLITE L- AND C-BAND RADAR BACKSCATTER FROM A BOREAL FOREST

IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM,;(2023)p. 8319-8322

Paper i proceeding

T. Leistner, A. R. Monteith, J. Gutierrez Lopez, L. M. H. Ulander. "Estimation of Stem Water Content from Tower-Based L-Band Radar Backscatter of a Boreal Forest". Submitted.

Etablera samband mellan trädens vatteninnehåll och satellit-SAR i boreal skog

Rymdstyrelsen (2021-00109), 2022-01-01 -- 2025-12-31.

Ämneskategorier (SSIF 2025)

Skogsvetenskap

Naturgeografi

Klimatvetenskap

Utgivare

Chalmers

HA1, Hörsalsvägen 4

Online

Opponent: Dr. Heather Reese, Department of Earth Sciences, University of Gothenburg, Sweden

Mer information

Senast uppdaterat

2025-04-11