11 September, 2020 (Green, nature, SMC->NBP)

Green, J.K., Seneviratne, S.I., Berg, A.M. et al. Large influence of soil moisture on long-term terrestrial carbon uptake. Nature 565, 476–479 (2019). https://doi.org/10.1038/s41586-018-0848-x

Overview

The authors concluded that soil moisture has substantial effect on long-term net biosphere productivity (NBP, a.k.a. NEE?) using a global satellite product of SIF and TWS (GRACE) and multi-model ensemble from the GLACE-CMIP5 project. They explained that the SMC effect was attributable to non-linear response of NBP to SMC. Also, they emphasized the need for better improvement representations of CO₂ fertilization effect in models, as it is a basis of their results.

SMC reduction and water-limited ecosystem

The authors decomposed changes in the total NBP into changes due to:

• dNBP_SMvar: soil moisture variations from the climatological annual cycle (i.e. seasonal mean)
• dNBP_trend: longer-term soil moisture trends (i.e. 30-year running mean)
• dNBP_others: CO₂ fertilization and temperature effects
• error: all other limiting factors

Figure 1:

• dNBP_SMvar + dNBP_trend (purple) remained negative during the period: SMC reduces NBP.
• Nevertheless NBP_land increased until it met a peak: dNBP_others has significant effect.
  • this means that model representations of CO₂ fertilization effect is crucially important for the projection.
• The contribution of dNBP_trend (dNBP_SMvar) gradually increased (decreased).
  • Once an ecosystem put in a water-limited condition, the intra-annual variability becomes less effective.
• ESMs agreed that there is a peak point for the global NBP.

Figure 2:

• All ESMs and observations (i) showed that actual GPP exponentially (?) decreases when SMC becomes less than a certain amount (i.e. turning to the water-limited condition).
  • SMC decreases –> E decreases –> less cooling effect (temperature increases)
  • SMC decreases –> E decreases –> VPD increases –> stomatal closure –> photosynthesis decreases –> NBP decreases
  • SMC decreases –> (Ra / GPP) ratio increases + decreased photosynthesis –> NBP decreases
  • SMC decreases + increased temperature –> more fire occurrences –> NBP decreases
• Therefore, it is also important to have realistic representations of vegetation responses to SMC reduction.

Etc

• increased temperature lengthens the growing season length in mid- and northern latitude +
  reduced cloud coverage increases available radiation in energy-limited regions
  = remains as a CO₂ sink

Methodology

Three experimental runs: a reference (CTL), ExpA, and ExpB

1. CTL: CMIP5 historic run + RCP8.5 scenario (SMC and CO₂ fertilization are accounted)
2. ExpA: CTL + mean climatological SMC forcing (seasonal mean SMC) –> remove soil moisture variability
3. ExpB: CTL + 30-year running mean SMC –> SMC trend

dNBP_SMvar = dNBP_{CTL - ExpB}: exclude the effect of long-term SMC trend (?)

dNBP_SMtrend = dNBP_{ExpB - ExpA}: exclude seasonal variability to remain the long-term variability only (?)