The effects of root modification on growth, water use, and yield of Vica faba L.

Dense subsoil layers in some Canterbury soils often provide mechanical resistance sufficient to prevent root growth into subsoil layers. This can increase crop susceptibility to drought stress, resulting in a reduction in yield potential. It was hypothesised that loosening the subsoil would increase root distribution at depth, and hence increase transpiration by the crop. Provided transpiration is proportional to crop dry matter production, and harvest index is stable, increases in seed yields would occur. It was also hypothesised that drought-avoiding strategies such as irrigation and early spring sowing would reduce the benefit of subsoiling. The hypotheses were tested by conducting two field experiments. In the first experiment (1983/1984), subsoiling and irrigation were used to assess the effect of root modification on water use and yield of Vicia faba. In the second experiment (1984/1985), the use of early sowing and subsoiling as a drought-avoiding strategy for Vicia faba was investigated. All crops were spring sown. The soil was a Templeton silt loam (Udic Ustochrept), with an incipient fragipan at 0.4 - 0.6 m depth. A single-tine winged subsoiler loosened the subsoil to a depth of 50 cm. Measurements of penetrometer resistance showed that the subsoiling operations were very effective in loosening the subsoil. The rooting density in subsoil layers and average rooting depth, measured by mini-rhizotrons, were generally increased in subsoiled treatments compared to conventionally cultivated treatments. However, total root length and root mass were increased only in the October sowing of the 1984/1985 experiment. Total root length of irrigated plants was significantly smaller than unirrigated plants, but there was no effect on average rooting depth. The cultivation treatments did not affect transpiration in the 1983/1984 experiment (average 352 mm), or the September sowing of the 1984/1985 experiment (average 244 mm). However, where total root length was also increased, in the October sowing of the 1984/1985 experiment, transpiration was increased by 18%. Total dry matter and shoot dry matter yields in subsoiled treatments were generally greater than those in conventionally cultivated treatments over both seasons. At maturity, shoot dry matter yields in subsoiled treatments averaged 10% greater than conventionally cultivated treatments. In irrigated treatments, shoot dry matter yields were 20% greater than unirrigated treatments, while early sown treatments out yielded late sown treatments by 93%. Within these treatments, however, subsoiling was still effective in increasing shoot dry matter yields. The harvest index was very variable within and between seasons, and thus increases in dry matter were not often translated to seed yield. Irrigation had the greatest influence on seed yields, averaging 1.71 and 1.17 t ha-1 in unirrigated and irrigated treatments respectively. The present study showed that dense subsoil layers restricted root growth at depth. Subsoiling improved root distribution, and led to increases in dry matter yields. For crops with a more stable harvest index, this would probably have also increased grain yields. Unexpectedly, while transpiration was little affected, water us efficiency was increased by root modification. Thus there were substantial physiological differences between plants grown on subsoiled and conventionally cultivated treatments. The potential for increased crop production by root modification in this area is considerable. These results have much use for crop modelling exercises. Suggestions for further research are provided.