Image analysis for non-destructive and non-invasive quantification of root growth and soil water content in rhizotrons

Studies aiming at quantification of roots growing in soil are often constrained by the lack of suitable methods for continuous, non-destructive measurements. A system is presented in which maize (Zea mays L.) seedlings were grown in acrylic containers - rhizotrons - in a soil layer 6-mm thick. These thin-layer soil rhizotrons facilitate homogenous soil preparation and non-destructive observation of root growth. Rhizotrons with plants were placed in a growth chamber on a rack slanted to a 45degrees angle to promote growth of roots along the transparent acrylic sheet. At 2- to 3-day intervals, rhizotrons were placed on a flatbed scanner to collect digital images from which root length and and root diameters were measured using RMS software. Images taken during the course of the experiment were also analyzed with QUACOS software that measures average pixel color values. Color readings obtained were converted to soil water content using images of reference soils of known soil water contents. To verify that roots observed at the surface of the rhizotrons were representative of the total root system in the rhizotrons, they were compared with destructive samples of roots that were carefully washed from soil and analyzed for total root length and root diameter. A significant positive relation was found between visible and washed out roots. However, the influence of soil water content and soil bulky density was reflected on seminal roots rather than first order laterals that are responsible for more than 80% of the total root length. Changes in soil water content during plant growth could be quantitified in the range of 0.04 to 0.26 cm(3) cm(-3) if image areas are of 500 x 500 pixel were analyzed and averaged. With spatial resolution of 12 x 12 pixel, however, soil water contents could only be discriminated below 0.09 cm(3) cm(-3) due to the spatial variation of color readings. Results show that this thin-layer soil rhizotron system allows researchers to observe and quantify simultaneously the time courses of seedling root development and soil water content without disturbance to the soil or roots.