Species identity improves soil respiration predictions in a semiarid scrubland

Carme Estruch*, Petr Macek, Cristina Armas, Nuria Pistón, Francisco I. Pugnaire

*Corresponding author for this work

Research output: Contribution to journalArticleResearchpeer-review

11 Citations (Scopus)

Abstract

Soil respiration accounts for ca. three quarters of total ecosystem respiration and is sensitive to temperature and moisture. Plants can influence soil CO2 emissions through specific effects on soil humidity, soil temperature and soil microbial communities. These plant-soil effects mostly come via litter production and root exudates, enhancing soil autotrophic and heterotrophic respiration. We explored how plant species affected soil CO2 emissions in an arid environment. We altered soil temperature in bare soil and under the canopy of four plant species differing in functional type, and measured monthly fluxes to establish seasonal patterns of CO2 release along a 20-month period. We found that soil temperature explained 69% of the annual soil respiration (SR) variance, while soil water content explained 71% of SR variance. When we included plant species identity in the model, soil temperature and soil water content explained 76% and 81% of SR variance, respectively, exemplifying how plant species modulate SR responses as a function of temperature and water availability. Our results demonstrate that plant species influence soil carbon balance and emphasize that species identity matters in dry ecosystems. SR dynamics in dry ecosystems can be accurately modelled with soil water and temperature as predictors, but models are more efficient if plant species identity is considered.

Original languageEnglish
Article number114153
JournalGeoderma
Volume363
DOIs
Publication statusPublished - 1 Apr 2020

Keywords

  • Arid plant communities
  • Climate change
  • Seasonal CO emissions
  • Soil respiration models

Fingerprint

Dive into the research topics of 'Species identity improves soil respiration predictions in a semiarid scrubland'. Together they form a unique fingerprint.

Cite this