Contribution of crop morphological characteristics and density of selected crops to weed species composition and suppression

Abstract

A field study was conducted at Lincoln University during the 1999-2000 growing season to investigate the effect of crop species and sowing density on weed dynamics, productivity and species composition. Crops with a spreading (narrow-leafed lupin, Lupimus augustifolius and dwarf French bean, Phaseolus vulgaris), rosette (turnip, Brassica campestris and forage rape, Brassica napus) and upright (maize, Zea mays and ryecorn, Secale cereale) growth habits were sown at 0.0. 0.5. 1.0. .2.0 and 4.0 times their optimum population. No other weed control measures were applied.

The six crops were sown on 8 September 1999 (early spring): narrow-leafed lupin, ryecorn and forage rape on and 4 November 1999 (early summer): dwarf French bean, maize and turnip. The weed seed bank prior to sowing was predominated by Coronopus didymus.

There were significant differences in the suppressive ability of the different crop species and different crop populations (p < 0.001). Weed dry matter (DM) was lowest in turnip (0.58 g/m²) and highest in dwarf French bean (123.50 g/m²) at final harvest.

By final harvest crop density had a marked effect on weed DM production. There was decreased weed DM at higher plant populations in all crop species, except turnip, at all plant populations. However, there were differences between 0.5 and 4.0 x optimum populations in bean (397 and 12 g/m²), rape (189 and 26 g/m²), lupin (125 and 7 g/m²), maize (106 and 0 g/m²) and ryecorn (51 and 18 g/m²).

Weed suppression as affected by the different crop treatments was directly related to leaf area index (LAI), radiation interception and radiation use efficiency (RUE). The highest LAI’s were recorded in turnip (4.1) at 60 DAS. While other crops such as bean and maize attained LAI's of only 0.66 and 1.1 respectively by the same time. Leaf area index increased with increased plant population in all crops. Leaf area index in turnip at 60 DAS ranged from 3.5 at 0.5 x optimum population to 5.1 at 4.0 x optimum population. In maize LAI ranged from 0.4 to 2.0 at the same plant populations at the same time.

Canopy closure occurred at 50 DAS in turnip at 4.0 x optimum population, at 60 DAS for lupin, ryecorn and rape and at 83 DAS for maize and bean. Canopy closure was never attained at 0.5 x optimum population in lupin, rape, ryecorn, bean and maize.

Turnip intercepted the most solar radiation (SR) at 1068 MJ/m² which was 354 MJ/m² more than bean, which intercepted the least photosynthetic active radiation (PAR). Total intercepted PAR also increased with increased plant population. There was a strong linear relationship between cumulative intercepted PAR and cumulative DM yield in all crops. Maize produced more DM per MJ of intercepted PAR than all the other crops at 3.4 g DM MJ PAR⁻¹ whereas lupin, ryecorn, rape, bean and turnip produced 1.7. 1.2. 0.98 and 0.37 g DM PAR⁻¹ respectively.

Crops with large leaf size and rapid growth were effective in reducing the weed seed bank, weed species and numbers in the following growing season. The most effective reduction occurred with turnip followed by maize (55 and 66 x 103 seed/m² respectively) compared with lupin and rape which contained the highest (158 and 130 x 103 seeds/m² respectively). Weed seed production was markedly affected by plant population (p < 0.05). Higher plant populations (2.0 and 4.0 x optimum population) of lupin, rape, ryecorn and maize effectively suppressed weed seed production. Lower plant populations (0.5 and 1.0 x optimum population) contained higher weed seed numbers/m² in lupin, rape, ryecorn, bean and maize plots. Coronopus didymus was the most abundant species in the weed seed bank in the 2000-growing season.

Weed seedling emergence in the 2000-growing season also reflected previous crop treatments. Bean contained the highest weed seedling density (1.163 weed seedlings/m²) and turnip the least (109 weed seedling/m²). As plant population increased from 0.0 to 4.0 x optimum population weed density decreased. The decrease was most pronounced in lupin (1.128 to 466 weed seedlings/m²), rape (1.082 to 319 weed seedlings/m²) and ryecorn (1.308 to 362 weed seedlings/m²). Chenopodium album was the most abundant weed species to emerge during the 2000 growing season.

A mechanistic model of crop and weed growth was tested for its ability to simulate DM accumulation of weed and crop in lupin, rape and ryecorn and used SR and LAI data. The model accurately predicted crop and weed DM for lupin, rape and ryecorn grown at the 4 crop densities. The model could be of practical value in predicting the potential weed DM yield under different crop species at different plant densities.

The results suggested that inclusion of large leaf size and rapid growth in selection of crop as competitors to suppress weeds by using crops such as turnip, maize and ryecorn should be feasible in a weed management program.