Assessing Light Use Efficiency of a Temperate Spruce Forest from Environmental Drivers

Light use efficiency (LUE) is an essential property of forest ecosystems and is driven by various intertwined environmental variables. An in-situ assessment of LUE, however, is difficult due to the complexity of measuring absorbed photosynthetically active radiation (APAR) of forest canopies. Here, we use APAR from a network of PAR sensors, eddy covariance derived gross primary productivity (GPP), and Sentinel-2 remote sensing data to assess LUE of green canopy parts of a temperate spruce forest. LUE is then analyzed for the 2021 growing season and in regard to environmental drivers including rare in-situ measurements of the physiologically important red to far-red ratio (R/FR). Results show that LUE of the canopy’s green part was on average 4.0% ± 2.3%. A unimodal relation of LUE to air temperature was found with a maximum LUE around 15 °C, while LUE decreased at vapor pressure deficits above 7.5 kPa. LUE was higher during diffuse light and when the R/FR ratio decreased little from above to below canopy measurements, likely because obscured parts of the canopy received a higher light quantity and quality in this way than under clear skies. The importance of environmental variables for GPP is further quantified using machine learning based variable importance measures. These results show that canopy chlorophyll content and APAR were crucial variables for GPP predictions. Finally, we critically evaluate the uncertainties and limitations of LUE estimates for green parts of a forest canopy.