Plant water content integrates hydraulics and carbon depletion to predict drought-induced seedling mortality

Gerard Sapes, Beth Roskilly, Solomon Dobrowski, Marco Maneta, William R.L. Anderegg, Jordi Martinez-Vilalta, Anna Sala

Research output: Contribution to journalArticlepeer-review

78 Scopus citations


Widespread drought-induced forest mortality (DIM) is expected to increase with climate change and drought, and is expected to have major impacts on carbon and water cycles. For large-scale assessment and management, it is critical to identify variables that integrate the physiological mechanisms of DIM and signal risk of DIM. We tested whether plant water content, a variable that can be remotely sensed at large scales, is a useful indicator of DIM risk at the population level. We subjected Pinus ponderosa Douglas ex C. Lawson seedlings to experimental drought using a point of no return experimental design. Periodically during the drought, independent sets of seedlings were sampled to measure physiological state (volumetric water content (VWC), percent loss of conductivity (PLC) and non-structural carbohydrates) and to estimate population-level probability of mortality through re-watering. We show that plant VWC is a good predictor of population-level DIM risk and exhibits a threshold-type response that distinguishes plants at no risk from those at increasing risk of mortality. We also show that plant VWC integrates the mechanisms involved in individual tree death: hydraulic failure (PLC), carbon depletion across organs and their interaction. Our results are promising for landscape-level monitoring of DIM risk.

Original languageEnglish
Pages (from-to)1300-1312
Number of pages13
JournalTree Physiology
Issue number8
StatePublished - Aug 1 2019


  • Pinus ponderosa
  • carbon starvation
  • drought
  • hydraulic failure
  • non-structural carbohydrates


Dive into the research topics of 'Plant water content integrates hydraulics and carbon depletion to predict drought-induced seedling mortality'. Together they form a unique fingerprint.

Cite this