The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here:

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

Default user image.

Lars Eklundh


Default user image.

Modelling and upscaling ecosystem respiration using thermal cameras and UAVs : Application to a peatland during and after a hot drought


  • Julia Kelly
  • Natascha Kljun
  • Lars Eklundh
  • Leif Klemedtsson
  • Bengt Liljebladh
  • Per Ola Olsson
  • Per Weslien
  • Xianghua Xie

Summary, in English

Field-based thermal infrared cameras provide surface temperature information at very high spatial and temporal resolution and could complement existing phenological camera and spectral sensor networks. Since temperature is one of the main drivers of ecosystem respiration (ER), field-based thermal cameras offer a new opportunity to model and upscale ER in unprecedented detail. We present such an approach based on manual chamber CO2 flux measurements and thermal imagery from a tower-based camera and from Unmanned Aerial Vehicle (UAV) flights. Data were collected over two growing seasons, including the hot drought of 2018, for the two main vegetation microforms (hummock and hollow) of a hemi-boreal peatland in Sweden. Thermal imagery proved suitable for modelling ER in this ecosystem: ER model accuracies were similar when air, soil or surface temperature measurements were used as input. Our findings allowed us to upscale ER using UAV-derived thermal images and we present maps of ER at sub-decimeter resolution (<7 cm). The significantly different ER measured for each microform highlighted the importance of modelling their ER separately. Not accounting for these differences and the microforms' spatial distribution across the peatland led to a bias in upscaled ER of up to 18%. As a result of the severity and timing of the hot drought in 2018, we observed reductions in the ER of both microforms, but more so for hummocks (-48%) than for hollows (-15%), and modelled ER leveled off at high temperatures. These findings indicate that peatland carbon loss during hot droughts may be lower than expected and strongly relates to vegetation composition. The presented upscaling approach offers a new method to analyse how ER varies across a peatland or within a flux-tower footprint, and to interpret biases that occur when using coarse resolution satellite data to upscale chamber or tower-based flux measurements.


  • MERGE: ModElling the Regional and Global Earth system
  • BECC: Biodiversity and Ecosystem services in a Changing Climate
  • Centre for Environmental and Climate Science (CEC)
  • Dept of Physical Geography and Ecosystem Science
  • eSSENCE: The e-Science Collaboration

Publishing year





Agricultural and Forest Meteorology



Document type

Journal article




  • Climate Research


  • 2018 drought
  • Heatwave
  • Remote sensing
  • Spatial heterogeneity
  • Surface temperature
  • Thermal imaging




  • GREEN GAP: GREENhouse GAs flux uPscaling - improved understanding of key ecosystem processes using remote sensing and ground-based measurements


  • ISSN: 0168-1923