Articles | Volume 21, issue 3
Hydrol. Earth Syst. Sci., 21, 1339–1358, 2017
Hydrol. Earth Syst. Sci., 21, 1339–1358, 2017

Research article 07 Mar 2017

Research article | 07 Mar 2017

Estimation of surface energy fluxes in the Arctic tundra using the remote sensing thermal-based Two-Source Energy Balance model

Jordi Cristóbal1,2, Anupma Prakash1, Martha C. Anderson3, William P. Kustas3, Eugénie S. Euskirchen4, and Douglas L. Kane2 Jordi Cristóbal et al.
  • 1Geophysical Institute, University of Alaska Fairbanks, Fairbanks, Alaska 99775, USA
  • 2Institute of Northern Engineering, Water Environmental Research Center, University of Alaska Fairbanks, Fairbanks, Alaska 99775, USA
  • 3Hydrology and Remote Sensing Laboratory, United States Department of Agriculture, Agriculture Research Service, Beltsville, Maryland 20705, USA
  • 4Institute of Arctic Biology. University of Alaska Fairbanks, Fairbanks, Alaska 99775, USA

Abstract. The Arctic has become generally a warmer place over the past decades leading to earlier snow melt, permafrost degradation and changing plant communities. Increases in precipitation and local evaporation in the Arctic, known as the acceleration components of the hydrologic cycle, coupled with land cover changes, have resulted in significant changes in the regional surface energy budget. Quantifying spatiotemporal trends in surface energy flux partitioning is key to forecasting ecological responses to changing climate conditions in the Arctic. An extensive local evaluation of the Two-Source Energy Balance model (TSEB) – a remote-sensing-based model using thermal infrared retrievals of land surface temperature – was performed using tower measurements collected over different tundra types in Alaska in all sky conditions over the full growing season from 2008 to 2012. Based on comparisons with flux tower observations, refinements in the original TSEB net radiation, soil heat flux and canopy transpiration parameterizations were identified for Arctic tundra. In particular, a revised method for estimating soil heat flux based on relationships with soil temperature was developed, resulting in significantly improved performance. These refinements result in mean turbulent flux errors generally less than 50 W m−2 at half-hourly time steps, similar to errors typically reported in surface energy balance modeling studies conducted in more temperate climatic regimes. The MODIS leaf area index (LAI) remote sensing product proved to be useful for estimating energy fluxes in Arctic tundra in the absence of field data on the local biomass amount. Model refinements found in this work at the local scale build toward a regional implementation of the TSEB model over Arctic tundra ecosystems, using thermal satellite remote sensing to assess response of surface fluxes to changing vegetation and climate conditions.

Short summary
Quantifying trends in surface energy fluxes is crucial for forecasting ecological responses in Arctic regions. An extensive evaluation using a thermal-based remote sensing model and ground measurements was performed in Alaska's Arctic tundra for 5 years. Results showed an accurate temporal trend of surface energy fluxes in concert with vegetation dynamics. This work builds toward a regional implementation over Arctic ecosystems to assess response of surface energy fluxes to climate change.