Location: Crops Pathology and Genetics ResearchTitle: The bias of a 2D view: Comparing 2D and 3D mesophyll surface area estimates using non-invasive imaging
|Theroux-rancourt, Guillaume - University Of California|
|Earles, J - Yale University|
|Gilbert, Matthew - University Of California|
|Zwieniecki, Maciej - University Of California|
|Boyce, C - Stanford University|
|Brodersen, Craig - Yale University|
Submitted to: New Phytologist
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/5/2017
Publication Date: 7/10/2017
Citation: Theroux-Rancourt, G., Earles, J.M., Gilbert, M., Zwieniecki, M.A., Boyce, C.K., Mcelrone, A.J., Brodersen, C. 2017. The bias of a 2D view: Comparing 2D and 3D mesophyll surface area estimates using non-invasive imaging. New Phytologist. doi: 10.1111/nph.14687.
Technical Abstract: The surface area of the leaf mesophyll exposed to intercellular airspace per leaf area (Sm) is closely associated with CO2 diffusion and photosynthetic rates. Sm is typically estimated from two-dimensional (2D) leaf sections and corrected for the three-dimensional (3D) geometry of mesophyll cells, leading to potential differences between the estimated and real cell surface area. Here, we examined how existing 2D methods used for estimating Sm compare to 3D values obtained from high-resolution X-ray computed tomography (microCT) for 23 species, with broad phylogenetic and anatomical coverage. Relative to 3D Sm values, uncorrected 2D Sm estimates were 15 to 30% lower on average. Two of the four 2D Sm methods typically fell within 10% of 3D values. For most species, only one 2D slice was needed to accurately estimate Sm within 10% of the whole leaf sample median. However, leaves with high vein density and divergent veins (e.g. eudicots) often required multiple sections. These results provide the first comparison of the accuracy of 2D methods in estimating the complex 3D geometry of internal leaf surfaces. Because microCT is not readily available, we provide guidance for using standard light microscopy techniques, as well as recommend standardization of reporting Sm values.