Summary. Spatial transportation networks underlie organismal performance in many biological systems. These networks reflect evolved solutions to transportation problems under natural selection. Network architecture can be highly variable, reflecting a set of tradeoffs that underlie network performance. There are four possible functions of networks efficiency, damage resistance, damage resilience, mechanical support that must be traded off against cost investment. Prior theory has mostly focused on single functions or costs. Moreover, very few networks have been fully quantified or had their performance measured, due to the difficulty of collecting data and developing conceptual vocabulary for multi-scale variation in architecture. To better identify the rules that link form and function, the project will advance theory for networks and transport processes, then confront this theory with extensive empirical data in a macroevolutionary context. Leaf venation networks are a model empirical system. Leaves are central to plant performance via their roles in carbon gain and water loss, processes mediated by resource transport through their venation networks. These networks have high diversity of form and function and are tractable to phenotyping and functional characterization. Five hundred species with high phylogenetic diversity will be collected from temperate forests, desert, and lowland/montane tropical forests. Performance will be measured in the field with ecophysiology methods, and machine learning methods will be used to extract network architecture from images. The project will 1) quantify network architecture in a phylogenetically broad set of species, 2) determine how network architecture and functions/costs are linked, 3) develop and test theory for these functions/costs of networks based on multi-scale network statistics, and 4) identify macro-evolutionary drivers of network architecture. Intellectual merit. The project will advance theory for the evolution of tradeoffs in spatial transportation networks. It will provide insight into the selective forces and biophysical constraints that have led to the evolution of species with a diverse set of network architectures. It will also generate a large dataset for linking network form and function in leaves, contributing to deeper understandings of plant ecophysiology and macroevolution. Last, it will develop a rule-set for understanding networks that could one day guide the engineering of artificial networks, e.g. for designing solar cells or synthetic organs. Broader impacts. The project will accelerate research efforts in other biological contexts by contributing very large empirical image datasets for network architecture, as well as machine learning algorithms to automatically extract networks from images. The project also will support interdisciplinary training for one postdoctoral researcher and two graduate students, who will gain international fieldwork and collaboration experience with colleagues at the University of Oxford. The project will support career development for four REU students and eight semesterly undergraduate interns via a comprehensive research mentoring program aimed at inclusion of underrepresented minority students. The project will also provide a competitive yearlong residency and international field experience for an artist who will produce a gallery show and several digital pieces communicating our work and its implications to the public.
|Effective start/end date||1/1/19 → 12/31/22|
- National Science Foundation (NSF): $997,795.00
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.