3D Leaf Geometry:

Controls on Photosynthesis and Transpiration

Leaves are sophisticated 3D bioreactors that have been evolutionarily optimized to inhabit Earth’s broad range of environmental conditions. The rate of CO2 and H2O transport between the atmosphere and chloroplasts are key controls on plant photosynthetic capacity, productivity, and survival. Atmospheric CO2 enters and transpirational H2O exits the leaf intercellular airspace via small pores called stomata. This 3D stomatal-intercellular airspace network defines the leaf vapor-phase and is a major biophysical control on CO2 and H2O transport rates. In addition to the vapor-phase network geometry, leaf internal CO2 and H2O transport depends on the mesophyll and vascular tissue distribution (i.e. the liquid-phase), and the liquid-vapor interface. Thus, we are currently exploring biophysical behavior arising from the 3D geometry of the vapor (i.e. stomata and intercellular airspace) and liquid (i.e. mesophyll and vascular tissue) phases, along with the liquid-vapor interface.

Due to instrumental and computational limitations, however, little work has been done to biophysically link leaf 3D geometry to photosynthesis and transpiration. As part of our growing 3D Leaf Atlas, we have collected X-ray microCT images (~1.5 μm3 resolution) for leaves from over 140 plant species spanning a diverse phylogeny, from bryophytes to recently derived angiosperms (Fig. 1A). From these microCT scans, we are innovating novel image processing techniques to quantify 3D leaf geometric traits that control CO2 and H2O transport rates, such as: stomatal distribution, porosity, lateral diffusivity, tortuosity, network disconnectivity, mesophyll surface area, and vascular distribution (Figs. 1B and 1C). Then, we use this 3D geometric data to build biophysical models (via finite element methods) of CO2 diffusion and photosynthesis, and H2O transport (Fig. 1D). These models are then physiologically validated against leaf-level gas exchange and hydraulic measurements. We’re currently applying these methods to 19 metabolically distinct CAM and C3 Bromeliaceae species, 50 conifer species, and more generally to other leaf types found in our 3D Leaf Atlas (Fig. 1E).


relevant publications

Earles, J.M.*, Théroux-Rancourt, G.*, McElrone, A.J. and Brodersen, C.R. (in preparation). A long, tortuous, and disconnected intercellular airspace impedes diffusion in CAM Bromeliaceae leaves. *Authors contributed equally

Théroux-Rancourt, G.*, Earles, J.M.*, Gilbert, M.E., Zwieniecki, M.A., Boyce, C.K., McElrone, A. and Brodersen, C. (2017). The bias of a two-dimensional view: Comparing two-dimensional and three-dimensional mesophyll surface area estimates using non-invasive imaging. New Phytologist, 215(4): 1609-1622. *Authors contributed equally



Kevin Boyce – Stanford University
Craig Brodersen – Yale University
Matthew Gilbert – University of California, Davis
Thorsten Knipfer – University of California, Davis
Andrew McElrone – USDA Agricultural Research Service
Guillaume Théroux-Rancourt – University of California, Davis
Maciej Zwieniecki – University of California, Davis