Research Focus Areas

3D Computer-Aided Design and Analysis of Plant Systems

Helios Plant and Environmental Modeling Framework

Helios is a model coupling framework designed to provide maximum flexibility in integrating and running arbitrary 3D environmental system models. Users interact with Helios through a well-documented open-source C++ API. Version 1.0 comes with model plug-ins for radiation transport, the surface energy balance, stomatal conductance, photosynthesis, solar position, and procedural tree generation. Additional plug-ins are also available for visualizing model geometry and data and for processing and integrating LiDAR scanning data. Many of the plug-ins perform calculations on the graphics processing unit, which allows for efficient simulation of very large domains with high detail.

Publication: Bailey (2019).
GitHub: www.github.com/PlantSimulationLab/Helios
Documentation: Helios Documentation

Computer-aided Design and Management of Agricultural Systems

3D computer-aided design (CAD) tools have revolutionized design and analysis of products across a wide range of industries, which allow for rapid design and evaluation of complex systems in a simulated environment before expending the time and money associated with prototyping. Agricultural systems have analysis have similar design challenges, with ``prototyping" (i.e., field trials) being a critical bottleneck that can take years to decades to complete.

Our lab is working to utilize Helios as a 3D CAD tool for agriculture in order to design and manage more efficient and sustainable production systems. One challenge associated with design and management of perennial cropping systems is that geometries are complex and heterogeneous, and have high variation in systems design and management practices. A strength of Helios is that it is able to explicitly resolve these complexities within the model, while still being efficient enough to simulate field-scale problems.

Our focus to date has been on designing vineyard production systems to mitigate excess fruit temperatures driven by a changing climate, and on increasing system-level water use efficiency in woody perennial cropping systems.

Development and Evaluation of Simplified Plant Systems Models

Although Helios is highly efficient when compared against models of comparable complexity, very large scale problems may necessitate very simplified model formulations (e.g., land surface models, dynamic vegetation models, ecosystem models). However, even simple models can be difficult to validate over large scales. Because Helios is based on physical conservation equations, it can be used to evaluate and develop simplified models in an idealized environment to determine whether they are theoretically consistent.

Publications: Ponce de Léon and Bailey (2019); Bailey et al (2020).

LiDAR Sensing of Plant Canopy Architecture

Terrestrial (ground-based) LiDAR scanning

LiDAR scanning is rapidly growing in popularity as a means for generating detailed measurements of plant canopy structure, which can, among other things, provide inputs for detailed models. However, the challenge with utilizing LiDAR datasets is the difficulty in processing raw data into useful information describing canopy structure. Raw LiDAR data consists of millions of (x,y,z) Cartesian points in space, which by themselves are really only useful in deriving distance measurements (e.g., plant height, width, etc.). But they don't directly provide information useful for model-based studies such as leaf area, biomass, normal vectors, etc. We have developed a number of algorithms that can extract 3D distributions of leaf area, the leaf angle distribution, and ultimately perform leaf-by-leaf reconstructions from ground-based LiDAR data. This allows us to measure the geometry of actual plant systems, and feed them into 3D biophysical models (see Helios above) to analyze canopy structure and function.

Publications: Bailey and Mahaffee (2017a); Bailey and Mahaffee (2017b); Bailey and Ochoa (2018).

Aerial LiDAR scanning

Aircraft-based LiDAR scanning data provides much lower resolution than ground-based data, but can cover a much wider spatial extent. High-resolution, continuous aerial LiDAR datasets are becoming readily available, such as the USGS 3DEP program that aims to generate a continuous across the conterminous U.S. The recent California state budget also allocates $80M to the generation of high-resolution LiDAR data across the state. These data are most commonly used to generate maps of elevation and vegetation height, but advancements are needed in algorithm development to convert raw aerial LiDAR data into inputs relevant to applications in wildfire spread, ecological modeling, among others. We have developed a plug-in within Helios process aerial LiDAR datasets into common raster maps of elevation, canopy height, and ground cover fraction, along with leaf area index (LAI) and the 3D distribution of leaf area density (LAD). The implementation improves on currently available software for processing aerial LiDAR data such as lastools in that 1) software is fully open-sourced and documented, and 2) calculations are accelerated by performing calculations on graphics processing units (GPUs) using NVIDIA CUDA.

Plant System Productivity and Water-Use Efficiency

Temperature-Based Measurement of Plant Water Status

Measurement of leaf or canopy temperatures have long been touted as a means for infering plant water status, and ultimately for guiding irrigation management decisions. The theoretical premise behind this approach is straight-forward: as plants become water stressed, stomatal pores on leaves begin to close to stop evaporation of water which leads to a reduction in evaporative cooling and an increase in temperature. When properly normalized to remove the effect of ambient weather, these temperature measurement can thus theoretically be related to water status. Our recent work using novel experimental and analytical techniques, which focuses on almond orchard applications, has revealed critical shortcomings with such measurements. Theoretical analyses have shown that the common temperature normalization procedure (the crop water stress index, or CWSI) does a poor job of removing weather effects, and is as sensitive to wind speed as water status. Experimental work has shown that, at least in almond, trees must be highly stressed before temperature-based measurements can detect a statistally significant decline in water status, rendering it ineffective for irrigation management.

Publications: Poirier-Pocovi and Bailey (2020); Poirier-Pocovi et al. (2020).

Turbulence and Energy Transport Processes

Turbulence in Plant Canopies

The character of turbulence in plant canopies is unique in that transport of mass and momentum to and from the canopy is highly efficient and dominated by large-scale coherent turblence structures with length scales on the order of the canopy height iteself. Our previous work has investigated how heterogeneity in canopy architecture affects this character of turbulent transport, and more fundamentally the mechanisms behind the origin and evolution of these coherent structures.

Publications: Bailey and Stoll (2013); Bailey et al. (2014); Bailey and Stoll (2016).

Lagrangian Particle Dispersion Modeling

Movement of particulates by turbulence is often most intuitively viewed from a Lagrangian frame of reference, which follows "packets" of particles as individuals. This approach can be used to perform unique analyses and visualizations of turbulent transport that can be difficult in an Eulerian framework, since the integrative nature of Lagrangian models damps chaotic turbulent fluctuations.

Publications: Bailey et al. (2014); Bailey (2017); Bailey et al. (2018).

Department of Plant Sciences
One Shields Ave.
Davis, CA 95616