Quantification of the 2018 drought for European forests and impacts - PowerPoint PPT Presentation

Quantification of the 2018 drought for European forests and impacts of stomatal and non stomatal limitation of photosynthesis

European 2018 drought European Drought Observatory, combined drought indicator (CDI) Drought taskforce -> Philosophical transaction of the royal society B

A bit of theory C c Non-stomatal At light saturation : g m 𝑑 = 𝑊 𝑑𝑛𝑏𝑦 (𝐷 𝑗 − Γ ∗) 𝐻𝑄𝑄 (𝐷 𝑗 + 𝐿 𝑛 ) C i 𝐷 𝑗 = 𝐷 𝑏 − 𝐻𝑄𝑄 g s Stomatal 𝑕 𝑡,𝑑𝑝2 C a √𝑊𝑄𝐸 ) 𝐻𝑄𝑄 𝑕 1 𝑕 𝑡,𝐼2𝑃 = 𝑕 0 + 1.6(1 + Medlyn et al.,2011 𝐷 𝑏 gs can be obtained from PM equation Figure from Zhou et al., 2019 LEG a γ g s,𝐼2𝑃 = s R n − G − S + ρC p C a VPD a − LE(s + γ) 𝑕 𝑡,𝐼2𝑃 = 𝑕 𝑡,𝐷𝑃2 1.6

Stomatal and non stomatal limitation of photosynthesis : models 𝑕 𝑡 𝐻𝑄𝑄 = 𝑊 𝑑𝑛𝑏𝑦 (𝐷 𝑗 − Γ ∗) (𝐷 𝑗 + 𝐿 𝑛 ) 𝑕 1 𝐷 𝑗 = 𝐷 𝑏 − 𝑯𝑸𝑸 𝒉 𝒅 𝐻𝑄𝑄 √𝑊𝑄𝐸) 𝐻𝑄𝑄 𝑕 1 𝑕 𝑡 = 𝑕 0 + 1.6(1 + 𝑊𝑄𝐸𝐷 𝑏 𝐷 𝑏 Stomatal limitation Non stomatal limitation Changes in C i which are associated Changes in apparent V cmax with with changes in g1 measured C i values (changes in the GPP-gc slope) g1 is inversely proportional to iWUE

Quantification of drought • In lack of soil and pre-dawn leaf water potential at flux tower sites , Relative Extractable Water (REW): REW t = SWC t − SWC WP SWC FC − SWC WP REW varies from 1 (Field capacity) and 0 (wiliting point) Soil humidity sensors Cumulated over the root zone

Ecosystem stations Site ID Country Latitute Longitude Dominating species Soil texture Precipitation Mean 2018 Difference BE-BRA Belgium 51.308 4.52 Pinus Syvlestris sand 819 724 95 Fagus BE-VIE Belgium 50.305 5.998 sylvatica/Pseudotsuga silty clay loam 898 592 305 Menziesii Quercus robur/Fraxinus 445 CZ-LNZ Czech 48.682 16.946 angustifolia/Carpinus sandy loam 438 7 betulus/Tilia cordata CZ-RAJ Czech 49.444 16.697 Picea abies sandy loam 622 474 148 CZ-STN Czech 49.036 17.97 Fagus Sylvatica sandy loam 644 457 187 DE-HAI Germany 51.079 10.453 Fagus Sylvatica clay loam 734 511 223 DK-SOR Denmark 55.486 11.645 Fagus Sylvatica sandy clay loam 837 399 438 FR-BIL France 44.494 -0.956 Pinus pinaster sand 892 884 8 FR-HES France 48.674 7.065 Fagus Sylvatica silty clay loam 827 737 90 IT-SR2 Italy 43.732 10.291 Pinus Pinea sandy 1078 908 170

Results : stomatal limitation  No consistant stomatal behavior across ecosystems Recall : g1 is inversely proportional to iWUE

Results : non stomatal limitation  Non stomatal limitations are observed at almost all sites where REW felt < 0.4

Degree of limitation We quantify the degree of limitation by : • Fixing V cmax at unstressed value and computing GPP with observed C i • Fixing G 1 at unstressed value and compute GPP with observed V cmax values Compute the ratio of GPP modelled /GPP observed

Degree of stomatal and non stomatal reduction  In most ecosystems, non-stomatal limitation is the dominant mechanism  Decrease of apparent Vcmax could be the result of both diffusional effects (mesophyll conductance) or biochemical effects

Focus on 3 beech forests • FR-HES, DK-SOR and DE-HAI are 3 beech forests • We observe non-stomatal limitation at all 3 sites In term of water use efficiency (iWUE) we observe : • Constant g1 at DK-SOR (constant iWUE) • Decreasing g1 at FR-HES (increased iWUE) which has a visible impact on GPP • Increasing g1 at DE-HAI (decreased iWUE) but with no visible impact on GPP (GPP is already too low) -> unsolved question !

Implications for drought modeling √𝑊𝑄𝐸) 𝐻𝑄𝑄 𝑕 1 𝑕 𝑡,𝐼2𝑃 = 𝑕 0 + 1.6(1 + 𝐷 𝑏 How should plante regulate stomata ? (Cowan & Farquhar, 1977) Stomata regulate both photosynthesis and transpiration Stomata should maximise : 𝐵 − λE 1 where λ is the carbon cost of water. g1 ~ λ If λ = 𝜀𝐵 𝜀𝐹 = 𝑑𝑝𝑜𝑡𝑢𝑏𝑜𝑢 (water spent now can’t be spent later) but does not apply when water availbility decrease ! -> when soil water depletes, the cost should increase ( λ ↗ and g1 ↘) Makëla et al., 1996 Results from this study do no support this !  the costs of stomatal opening are probably not well identified Ideas : - Loss of hydraulic conductivity Dewar et al., 2018 - Limit non-stomatal limitation

Conclusions • Non stomatal limitation was the dominant short term mechanism limiting GPP in forest at flux tower sites • Apparent V cmax has proven a useful way of modeling these NSL • Future optimal conductance models should take NSL into accounts • REW has proven a very useful index of edaphic drought at flux tower sites

Thank you !