Poor Fertility, Short Longevity, and Low Abundance in the Soil Seed Bank Limit Volunteer Sugarcane from Seed

Abstract

The recent development of genetically modified sugarcane, with the aim of commercial production, requires an understanding of the potential risks of increased weediness of sugarcane as a result of spread and persistence of volunteer sugarcane. As sugarcane is propagated vegetatively from pieces of stalk and the seed plays no part in the production cycle, the fate of seed in the environment is yet to be studied. In this study, sugarcane seed samples, collected in fields over a 2-year period, were used to determine the overall level of sugarcane fertility, seed dormancy, and longevity of seed under field conditions. A survey of the soil seed bank in and around sugarcane fields was used to quantify the presence of sugarcane seeds and to identify and quantify the weeds that would compete with sugarcane seedlings. We demonstrated that under field conditions, sugarcane has low fertility and produces non-dormant seed. The viability of the seeds decayed rapidly (half-life between 1.5 and 2.1 months). This means that, in Australia, sugarcane seeds die before they encounter climatic conditions that could allow them to germinate and establish. Finally, the soil seed bank analysis revealed that there were very few sugarcane seeds relative to the large number of weed seeds that exert a large competitive effect. In conclusion, low fertility, short persistence, and poor ability to compete limit the capacity of sugarcane seed spread and persistence in the environment.

Keywords: GMO environmental risk assessment; Saccharum spp.; seed dormancy; seed persistence; soil seed bank; sugarcane; weed competition.

Figure 1

Distribution of average number of germinated seed per gram of fuzz from 84 field samples collected in 2012 and 2013. Germination tests were conducted at 30 or 36°C, under constant light and non-limited water conditions with either three or five technical replicates.

Figure 2

Sugarcane seed longevity under field conditions. The effect of site and depth of burial on sugarcane seed longevity was assessed on three field-collected samples. Symbols represent the average number of germinated seed from bags recovered from the fields each month and error bars denote SE (±). Depth (p-value <2.2 e⁻¹⁶), Sample (p-value = 3.8 e⁻¹⁵) and Site (p-value = 9.6 e⁻⁰⁵) had a significant effect on seed longevity over time.

Figure 3

Sugarcane seed longevity under field conditions. The effect of amount of seed, site, and depth of burial on sugarcane seed longevity was assessed on four field-collected samples. Symbols represent the average number of germinated seed from bags recovered each month and error bars denote SE (±). Depth (p-value <2.2 e⁻¹⁶), Site (p-value <2.2 e⁻¹⁶), Sample (p-value = 2.9 e⁻⁸), and Amount (p-value = 0.007) had a significant effect on seed longevity over time.

Figure 4

Estimation of seed longevity half-life based on the exponential decay of seed viability. The decay constant was estimated by fitting a non-linear model and seed longevity half-life was derived by dividing ln(2) by the decay constant.