|Smucker A J M|
|Ferguson J C|
Submitted to: Plant and Soil
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/20/1997
Publication Date: N/A
Citation: N/A Interpretive Summary: To understand the vital role roots play in plant growth, it is essential to know the size and distribution of the plant root system. Research in this area has been limited severely by labor intensive needs to separate manually non-root materials that are left with the roots when they are washed out of the soil. We developed a technique to carry out this separation task electronically with image analyses. Labor requirements were cut by 50%. In addition, accuracy was improved greatly. Researchers in root investigations benefit directly, since this breakthrough will expand their ability to carry out root studies.
Technical Abstract: Root lengths and diameters offer important clues for understanding root dynamics and their multiple functions. Although other techniques have been described, soil coring methods are the most accurate for quantifying the distribution and biomass of replicated root samples taken from the field. Historically, destructive root sampling has been labor intensive and requires manual separation of extraneous organic debris recovered along with the hydropneumatic elutriation of roots from soil cores. Quantification of root distributions by public domain (NIH-Image) and Vicom-based image processing algorithms eliminated labor-intensive manual separation. This was accomplished by determining a threshold length to diameter ratio for each object from length and diameter data generated during image analyses. Objects with a length to diameter ratio less than this threshold value were considered non-root materials and electronically separated by computer algorithms. Iterative analyses of length to diamete ratios of 16:1 provided the best threshold ratio for separating root images of subsurface maize (Zea mays L.) from associated organic debris. In contrast, 12:1 ratios were the best threshold value for separating roots from debris in surface soils. Using this combination of threshold ratios for a set of 24 soil cores, a near perfect fit was obtained between calculated total root length per core and actual root length. A linear relationship was observed between root lengths determined by NIH-Image analyses and lengths determined independently by a Vicom image processor, using the same root samples. This high correlation suggests that computer image processing provides opportunities for comparing root length densities between laboratories.