Aspects of browntop seed production

Abstract

A three-year old stand of browntop (Agrostis capillaris L.) cv. Grasslands Sefton on the Lincoln University cropping farm was used to investigate the effects of post-harvest management (autumn-stubble treatment and sheep grazing), spring nitrogen and plant growth retardants on dry matter yield, seed yield and seed quality.

Autumn-stubble treatments (control, mowing or burning) were imposed in April, four weeks after harvest. The sheep-grazing experiment was located next to the main experimental site. Spring-nitrogen treatments were 0, 60, 120 or 240 kg N/ha; the total N allocation to each plot was applied in three split applications in early September, early October and early November. The chlormequat-timing treatment was imposed in October, December and January (early, middle or late). Terpal-rate treatment (0, 2 or 4 l a.i./ha) was imposed in December and timing treatment was imposed in October and December. There was also a nil treatment for each experiment.

Green material was analyzed for total N monthly by AgResearch Invermay Soil Fertility Service. Apical development stages were determined by microscopic dissection of tiller apices fortnightly throughout the winter and the spring. Growth stages were determined by field observation and lodging was assessed during the growing season. Seed was hand harvested in February, hand threshed and machine cleaned, and seed yield components recorded. The thousand-seed weight (TSW) and second-seed percentage (%) were recorded. Seed germination was measured using two different methods. The fresh and dry weight of seedlings, an indication of seed vigour were measured at the interim and final stage.

Data collected were analyzed using both univariate and multivariate statistical methods by software Minitab, Genstat or SAS.

There was a linear response in browntop seed yield to increasing the amount of N applied up to 240 kg/ha in spring. The increase in yield resulted from an increase in yield/head and thousand seed weight which compensated for a significant reduction in heads/m² with increasing N. The size of seed head appeared to be an important component in yield formation and this was further confirmed using multivariate statistical methods.

By monitoring apex development in browntop, it was found that the marked increases in the size of seed head in response to N application occurred at the critical period of floret initiation in November/December. Examination of nitrogen-use efficiency demonstrated that the highest N absorption occurred in December after 120 kg/ha N fertilizer was split-applied in September, October and November.

Herbage tissue N percentage (%) was significantly (p<0.001) correlated with seed yield at all samplings although the highest correlation occurred in December. This was supported by results analyzed using multivariate statistical methods, as canonical correlation analysis showed that among N-related variables, N absorption in November/ December had the highest correlation with seed yield-related parameters. Therefore it might be possible to influence seed yield by adjusting herbage N early in the season to ensure optimum herbage N% in November/December.

Autumn-stubble treatment had no effect on seed yield and there were no interactive effects between autumn-stubble treatment and spring-N application on seed yield. The sheep-grazing experiment showed that seed areas can be grazed by sheep after seed harvest until mid-winter without decreasing browntop seed yield. Autumn-stubble treatment, sheep grazing and PGR application did not reduce dry matter yield at harvest.

Nitrogen application markedly improved quality parameters (seed-N concentration (%), TSW, seed germination (%), seed vigour and second-seed percentage). Seed vigour of browntop was significantly (P<0.01) increased by N applied during the growing season, as N had a positive and highly significant effect on fresh and dry seedling weight. In addition, both TSW and seed-N content were closely related to seedling weight. Autumn-stubble treatment, sheep-grazing and PGR application had little effect on seed quality parameters.

Further experimental work might be conducted to investigate the long term effects of autumn-stubble treatment and sheep-grazing, validate the model created based on herbage tissue N analysis and further evaluate Terpal and chlormequat.