Sowing crops into dry water repellent soils can increase the severity of repellency and reduce crop performance

Context. Dry-sowing, the practice of planting crops before autumn rains, is increasingly used to expedite cropping programs and enable crops to germinate promptly after rainfall. However, growers have reported poor crop emergence on water-repellent soils following this practice, even after subsequent substantial rainfall. Aims. To test hypotheses that, compared to wet-sowing, dry-sowing on water-repellent soils disrupts soil physical properties, increases repellency and reduces crop performance. Methods. Field experiments were conducted at Pingrup, Western Australia, from 2015 to 2017 on a water-repellent Petroferric Brown Sodosol. In 2015, an unreplicated demonstration trial was dry-sown and a similar sized area was wet-sown 5 days later after rainfall the previous day. In 2016 and 2017, rainout shelters were used to manipulate soil water at seeding and create dry- and wet-sowing conditions simultaneously in a replicated experiment established in a randomised complete block design. Crop emergence, grain yield, 1000-grain weights, soil water repellency and soil water contents were measured. Laboratory experiments assessed changes in soil properties (repellency, water infiltration, bulk density) in soils that were wetted, compacted while wet, then dried to simulate field conditions before disturbance. Key results. In 2015, the demonstration trial yielded 22% less barley (Hordeum vulgare) under dry-sowing (1.4 t ha-1; standard error of the mean (SEM) 0.02) than wet-sowing (1.8 t ha-1, SEM 0.04). In field experiments in 2016 (barley) and 2017 (lupin (Lupinus angustifolius)), dry-sowing reduced crop emergence by 42% (59 vs. 101 plants m-²; P 0.001) and 54% (16 vs. 35 plants m-²; P 0.001) respectively but yield impacts were small (2.1 t ha-1 vs 2.4 t ha-1; P 0.05 in 2016; 2.4 t ha-1 vs 2.3 t ha-1, not significant in 2017), likely due to frost at vulnerable crop stages and wetter than average seasons. Compared to wet-sowing, dry-sowing significantly increased soil water repellency (mean MED of 0.6 vs 0.2; P 0.001 in 2016 , and 0.7 vs 0.3; P 0.001 in 2017) and reduced soil water contents in the top 0.05 m shortly after sowing (7 vs 10 % v/v; P 0.001 in 2016, and 8 vs 10 % v/v; P 0.001 in 2017). There were no significant differences in surface soil water content later in the season in either year. Laboratory experiments revealed that disturbing dried water-repellent Pingrup soil immediately increased water repellency (as measured by both the Molarity of Ethanol Drop (MED) test (P 0.001) and the Water Drop Penetration Time test (P = 0.004)) and increased soil bulk density by 15% (P 0.001). Similar changes were also observed in two additional water repellent soils collected from different locations in WA. These changes were attributed to disruption of hydrophilic pore networks needed for water infiltration. Conclusions. On water-repellent soils, sowing should occur after autumn rains commence to preserve soil pore structure and water infiltration for optimal crop performance, even where severity of repellence is regarded as low. Implications. Further research into strategies that conserve soil structure may enable successful dry-sowing on water repellent soils.