What is root scanning?

(And why does it take so long?!)

Our lab has been using the root scanner as a tool for root trait analysis for a few years now. The scans are analysed with WinRHIZO software which provides (among other things) structural root trait information such as total root length, average root diameter and root volume. As in this paper by Alex Williams.

In a previous experiment, I scanned a great many samples comprised of a single plant still joined at the root crown. This made the cleaning process easier as you had something to hold on to while washing all other material away. Our current experiment, which is investigating community effects, has 12 plants in each pot. For this reason, we chose to take three soil cores instead of root washing the whole pot.

These three cores meant we had less soil to wash away, and less root material in total, but it also meant that we generally had no anchor points. All of our roots are free agents, and vary greatly in size. Root washing our cores without losing any root material proved to be a tricky task. We used a system of sieves, beakers to dip, and medium power spray head. These samples were then stored in ethanol over Christmas.

In January, the laborious task of the final picking and scanning began. This is the really up close and personal cleaning process. Each sample is rinsed then a section at a time is picked over with tweezers to remove any remaining bits of grit, above-ground material, moss or clearly already dead plant material (because we want a snapshot of the root system at the moment of harvest, not any decaying material from before).

This is done in a series of takeaway style or transparent tupperware tubs in front of our large lab windows, to take advantage of the light. Many of my colleagues pity me for having my nose in a tub of rooty water all day most days, but luckily I find it quite zen. There’s a sense of satisfaction hunting out the bits of debris that shouldn’t be there.

The amount and type of root, and amount of unwanted material varies greatly between samples, as does the time they each take but it is common for one sample to take many hours. I have found small paintbrushes can be helpful in catching moss fragments, and have made a range of tiny sieves out of the plastic mesh we use to make litter bags, to aid in catching the roots we want.

Once the sample is satisfactorily clean, it is gently poured into a perspex tray on the scanner bed. Additional DI water is added until the roots are covered. Only ever using plastic tools, the roots are carefully teased apart and spread out until they form a layer one root deep (or as close as we can get it). The scan itself usually takes about five minutes. In that time, we prepare a paper bag, blue roll to pat the sample dry, and a sieve to catch it when we drain the DI water. We record the fresh weight then place the sample in an oven to dry.