Constraining the spatial and temporal patterns of heterotrophic respiration using observations of terrestrial carbon fluxes and carbon use efficiency
Zhi hua Liu, University of Montana, liuzh833@126.com
Ashley Ballantyne, University of Montana, ashley.ballantyne@umontana.edu (Presenter)
John S Kimball, University of Montana, john.kimball@mso.umt.edu
Ben Bond-Lamberty, Pacific NW National Lab, bondlamberty@pnnl.gov
Rodrigo Vargas, University of Delaware, rvargas@udel.edu
Kevin Arthur Endsley, University of Montana, arthur.endsley@ntsg.umt.edu
Elizabeth Smith, University of Delaware, smithem@udel.edu
Heterotrophic respiration (Rh) is one of the most difficult carbon fluxes in the global terrestrial carbon cycle to constrain and creates tremendous uncertainties in understanding climate-carbon feedbacks. Although the magnitude and trend of global Rh can be estimated from bottom-up in situ observations of soil respiration with assumptions, these observations are often geographically biased. In this study, we took advantage of remote sensing of terrestrial carbon fluxes and carbon use efficiency (CUE), defined as the ratio between net primary productivity and gross primary productivity, to estimate monthly Rh in a top-down approach at 1-degree spatial resolution. We first produced a CUE map using a suite of observational biophysical variables and machine learning (ML) techniques trained with extensive in situ CUE observations (n = 353 site-years). Ecosystem respiration was then separated into autotrophic and heterotrophic components based on satellite observations of gross primary productivity, net ecosystem productivity, and CUE. The ML-derived CUE map ranged between 0.2 to 0.8, which is consistent with ecosystem-scale measurements and meta-analyses, exhibiting a latitudinal pattern controlled by climate, land cover, and soil moisture. Our top-down approach estimates the annual Rh as 52.76±10.43 PgC yr-1, consistent with many bottom-up estimates of Rh. This work offers an observational-constrained approach to understanding the spatial and temporal patterns of Rh, thus illustrating the usefulness of data-driven estimates of Rh by remote sensing of terrestrial carbon fluxes.
Associated Project(s):
Poster Location ID: 1-41
Presentation Type: Poster
Session: Poster Session 1
Session Date: Tue (May 9) 5:00-7:00 PM
CCE Program: TE