Dimensional Visualization and Modelling of Rice Root System Architecture in Soil

Introduction

United Nations FAO (Food and Agricultural Organization) has predicted that food production must increase by 50% in 2030 and double by 2050 to meet the demands of a growing population. This has to be accomplished sustainably under the challenges of decreasing land availability (due to desertification, urbanization, and competition from biofuels) and water resources (depleting ground and surface water reserves, increasing salinity), low soil fertility and ineffective use of fertilizers, and climate change induced extreme events such as heat stress. Focus is on the root system of plants to achieve this. Rice, which requires nearly three times more water than other cereal crops such as wheat and maize, is India’s number one crop in terms of tonnage and one of the world’s top five major cereal crops. Here, we focus on investigating water and nutrient use efficiency of rice plants by studying it root system architecture (RSA).

The phenotype of RSA is influenced by its genotype and the biotic and abiotic environment. Advancements in imaging of roots from ex-situ, invasive methods like soil coring and shovelomics to in-situ methods like minirhizotrons, transparent gel-based growth system and hydroponics help us characterize the 3D architecture of the roots. Non-invasive imaging techniques such as X-ray Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) are leveraged to obtain whole plant in-situ 3D RSA with micron scale resolution. A 3D root growth model combined with high resolution 3D imaging will help us understand the structure-functional relationship of the RSA. Understanding the root growth and function will be greatly beneficial for agricultural gains and food security.

Motivation

The distribution of root water uptake in soil is influenced by spatio-temporal root distribution and soil properties. 1D root water uptake models, using simplified root growth and root distribution simulations, fail to capture the spatial heterogeneity of physiological processes in the root system [1, 2]. They also fail in modelling the horizontal water content variability that arises due to contrasting uptake by different root segments and negative feedbacks like inter-root competition for water and nutrients. The errors (arising due to the above-mentioned reasons) scale up resulting in inefficient model-based recommendations for water and nutrient use impacting crop yield. Hence, we need a 3D root water uptake model that considers the 3D RSA explicitly.

Objectives

• Obtain the 3D RSA of rice non-destructively in soil medium over time in laboratory controlled ideal and stress (drought, low nutrient availability) conditions.
• Model the spatio-temporal water and nutrient uptake by roots under various stress conditions and validate it using MRI technique.
• Based on the above objectives and SimRoot [2] (a root growth model developed for maize and common beans), develop a 3D structure-function root growth model of rice. The illustrations are shown for different crops representing the processes) (b) Single cross-section of Wheat obtained from x-ray CT [4] (c) Segmented roots of Wheat obtained from Root1 (e) Simulated roots of Common bean from SimRoot [2]

1. Rice plants will be grown in an experimental rooftop greenhouse facility at the IIT Madras campus, Chennai, India under controlled conditions.
2. In-situ non-destructive imaging modalities such as x-ray CT and MRI will be employed to obtain the 3D RSA. X-ray CT technique gives us 3D soil structure along with 3D RSA and MRI gives us water content in the root zone and soil.
3. The obtained image will be segmented and reconstructed with RooTrak and Root1 to remove non-root matter (such as soil, water and air pockets) and obtain the 3D RSA. The required root phenotypes such as root length, number of branches, branching angles, diameter of fine roots and root density will be measured as a function of time using above-mentioned software.
4. The resulting dynamic 3D RSA will be employed to model the spatial and temporal water uptake by the root zone using a 3D Richard’s equation model that will be developed as a part of the dissertation.
5. A structure-functional 3D root growth model for rice plants will be developed from the 3D RSA (structure) and water uptake model (function). The model will be simulated under various conditions to provide guidelines for optimal application of fertilizers and other resources.

Expected Outcome

The dissertation will include the development of a 3D root growth model for rice and model simulations under resource constraints. Root growth models exist for specific plants such as maize and common beans. Since rice is a major crop in the Indian sub-continent, developing a 3D root growth coupled with 3D water and nutrient uptake models will be helpful for studying the root traits for resource acquisition under stress. This will help us in increasing the nutrient uptake efficiency of the roots, determine the optimal water and fertilizer input and cropping density to reduce root competition which can severely reduce the overall efficiency of the root system. This will be a step in the direction of achieving increased crop yield sustainably.