Leaf vessel networks play a significant role in determining whole plant hydrodynamics, carbon balance, and leaf development. Unfortunately due to their small size and large number of bifurcations, few exhaustive descriptions of the geometry of entire leaf networks exist. We have developed a series of automated image segmentation and network extraction algorithms that identify the geometry of veins, and the areoles they surround, in leaves. We have combined these algorithms in a graphical user interface (to be made publicly available). These algorithms were used to analyze ontogenetic and interspecific patterns from hundreds of leaf networks from dozens of species. Our combined data set also includes measurements of whole leaf properties such as surface area, petiole diameter, dry mass and tissue density. Together these measures serve as a benchmark for understanding how leaf vascular networks impact the scaling of whole leaf form and function.
Results/Conclusions
Across species, total network length and volume behave allometrically when regressed against whole leaf dry mass, wet mass or surface area. In contrast, 2D network surface area is isometric. In addition, several network properties appear statistically invariant with increases in leaf size, including mean vein length, and network volume fraction. Intraspecifically, significant variation in network scaling exists. For example, network length, area and volume exhibit negative allometry, positive allometry, and isometry, depending on species identity and leaf shape. The intraspecific allometric slopes are also correlated with other well known measures of leaf investment such as specific leaf area and tissue density. Our combined results provide a more detailed picture of the scaling of leaf form and investment than has been heretofore possible. We discuss these results in light of existing theories of branching networks. In addition, the automated network analysis algorithms and graphical user interface will allow investigators greater ease in analyze complex leaf network data, considerably expanding our knowledge of the ecological and evolutionary variability in these important photosynthetic organs.
