Nonlinear models based on leaf architecture traits explain the variability of mesophyll conductance across plant species

Mesophyll conductance ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>g</mml:mi> <mml:mi>m</mml:mi></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0001" wiley:location="equation/pce15059-math-0001.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>g</mi><mi mathvariant="normal">m</mi></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> ) describes the efficiency with which <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>CO</mml:mi> <mml:mn>2</mml:mn></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0002" wiley:location="equation/pce15059-math-0002.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>CO</mi><mn>2</mn></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> moves from substomatal cavities to chloroplasts. Despite the stipulated importance of leaf architecture in affecting <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>g</mml:mi> <mml:mi>m</mml:mi></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0003" wiley:location="equation/pce15059-math-0003.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>g</mi><mi mathvariant="normal">m</mi></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> , there remains a considerable ambiguity about how and whether leaf anatomy influences <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>g</mml:mi> <mml:mi>m</mml:mi></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0004" wiley:location="equation/pce15059-math-0004.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>g</mi><mi mathvariant="normal">m</mi></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> . Here, we employed nonlinear machine-learning models to assess the relationship between 10 leaf architecture traits and <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>g</mml:mi> <mml:mi>m</mml:mi></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0005" wiley:location="equation/pce15059-math-0005.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>g</mi><mi mathvariant="normal">m</mi></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> . These models used leaf architecture traits as predictors and achieved excellent predictability of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>g</mml:mi> <mml:mi>m</mml:mi></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0006" wiley:location="equation/pce15059-math-0006.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>g</mi><mi mathvariant="normal">m</mi></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> . Dissection of the importance of leaf architecture traits in the models indicated that cell wall thickness and chloroplast area exposed to internal airspace have a large impact on interspecific variation in <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>g</mml:mi> <mml:mi>m</mml:mi></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0007" wiley:location="equation/pce15059-math-0007.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>g</mi><mi mathvariant="normal">m</mi></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> . Additionally, other leaf architecture traits, such as leaf thickness, leaf density and chloroplast thickness, emerged as important predictors of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>g</mml:mi> <mml:mi>m</mml:mi></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0008" wiley:location="equation/pce15059-math-0008.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>g</mi><mi mathvariant="normal">m</mi></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> . We also found significant differences in the predictability between models trained on different plant functional types. Therefore, by moving beyond simple linear and exponential models, our analyses demonstrated that a larger suite of leaf architecture traits drive differences in <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>g</mml:mi> <mml:mi>m</mml:mi></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0009" wiley:location="equation/pce15059-math-0009.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>g</mi><mi mathvariant="normal">m</mi></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> than has been previously acknowledged. These findings pave the way for modulating <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:mrow><mml:msub><mml:mi>g</mml:mi> <mml:mi>m</mml:mi></mml:msub> </mml:mrow> </mml:mrow> <mml:annotation><math altimg="urn:x-wiley:01407791:media:pce15059:pce15059-math-0010" wiley:location="equation/pce15059-math-0010.png" display="inline" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mrow><msub><mi>g</mi><mi mathvariant="normal">m</mi></msub></mrow></mrow></math></mml:annotation></mml:semantics> </mml:math> by strategies that modify its leaf architecture determinants.