Modeling cotton ( Gossypium spp.) leaves and canopy using computer aided geometric design (CAGD)

Abstract

The research presented here develops a geometrically accurate model of cotton crop canopies that can be used to explore changes in canopy microenvironment and physiological function with leaf structure. We develop an accurate representation of the leaves, including changes in three-dimensional folding and orientation with age and cultivar. Photogrammetrical analysis of leaf surfaces is used to generate measured points at known positions. Interpolation of points located on the surface of the cotton leaves is then performed with a tensor product interpolants model that generates a generic leaf shape. Dynamic changes in leaf shape and canopy position over the growing season are based on measurements of cotton canopies in the field, and are used to modulate the generic leaf shape. The simulated leaves populate a canopy element based on statistical distributions from measured crop canopies. The simulation is found to give a good representation of cotton canopy leaves, adequately capturing the three-dimensional structure of the leaves and changes in leaf shape and size over the growing season. The simulated canopy accurately estimates leaf area index, except for the earliest measurement period prior to canopy closure. The application of the CAGD algorithm for representing cotton leaf and canopy geometry, and the technique for changing the leaves’ spatial position, size and shape through time of four representative cotton canopies is found to be a useful tool for developing a realistic crop canopy. We use leaf area index (LAI) as a measure of the accuracy of model-predicted LAI values in comparison to LAI in crop canopies in situ, obtaining r 2 values ranging from 0.82 to 0.92. The level of detail captured in the model could contribute greatly to future studies of physiological function and biophysical dynamics within a crop canopy.