Updated respiration routines alter spatio-temporal patterns of carbon cycling in a global land surface model

Ethan E. Butler, Kirk R. Wythers, Habacuc Flores-Moreno, Ming Chen, Abhirup Datta, Daniel M. Ricciuto, Owen K. Atkin, Jens Kattge, Peter E. Thornton, Arindam Banerjee, Peter B. Reich

Research output: Contribution to journalArticlepeer-review


We updated the routines used to estimate leaf maintenance respiration (MR) in the Energy Land Model (ELM) using a comprehensive global respiration data base. The updated algorithm includes a temperature acclimating base rate, an updated instantaneous temperature response, and new plant functional type specific parameters. The updated MR algorithm resulted in a very large increase in global MR of 16.1 Pg (38%), but the signal was not geographically uniform. The increase was concentrated in the tropics and humid warm-temperate forests. The increase in MR led to large but proportionally smaller decreases in global net primary production (19%) and in average global leaf area index (15%). The effect on global gross primary production (GPP) was a more modest 5.7 Pg (4%). A detailed site level analysis also demonstrated a wide range of effects the updated algorithm can have on the seasonal cycle of GPP. Output from the updated and old models did not differ markedly in how closely they matched a suite of benchmarks. Given the substantial impact on the land surface carbon cycle, a neutral influence on model benchmarks, and better alignment with empirical evidence, an MR algorithm similar to the one presented here should be adopted into ELM.

Original languageEnglish (US)
Article number104015
JournalEnvironmental Research Letters
Issue number10
StatePublished - Oct 2021


  • carbon cycle
  • global
  • land surface model
  • respiration

ASJC Scopus subject areas

  • Renewable Energy, Sustainability and the Environment
  • Environmental Science(all)
  • Public Health, Environmental and Occupational Health


Dive into the research topics of 'Updated respiration routines alter spatio-temporal patterns of carbon cycling in a global land surface model'. Together they form a unique fingerprint.

Cite this