Monitoring Nitrogen Uptake and Mineralization by Brassica Cover Crops in Maryland.
Amy Kremen and Ray R. Weil. Univ of Maryland, 1103 H.J.Patterson Hall, College Park, MD 20742
Cover crops that capture residual nitrogen from the soil profile in fall may reduce fertilizer need and minimize nitrate leaching if N mineralization from the cover crop residues in spring is synchronous with main crop demand. This research focuses on forage radish (Raphanus sativus L.) and rape (Brassica napus L.) as part of a larger study on several effects of using Brassicas as winter cover crops. Frost kills forage radish in December/January under Maryland (USA) conditions and we killed rape with glyphosate or tillage in April before summer crop planting. The ability of these species to capture N and then release plant-available N into soils was compared to that for rye, currently the most commonly planted cover crop in Maryland. Four cover crop treatments (rye, forage radish, rape and no cover) were applied in randomized complete block field experiments at four research stations in Maryland using a cover/soybean/cover/corn rotation from 2003-05. Two experiments used no-till, conventional tillage was used in one experiment, and no-till was compared to spring disking in one experiment. Surface soil (0-15 and 15-30 cm) mineral N (NO3--N and NH4+-N)and Soluble Organic N (SON)were monitored using monthly composite samples. Soil and air temperatures were monitored throughout the study and soil water content, bulk density, and texture were determined at each sampling date. Cover crop root and shoot biomass was estimated by harvesting two 0.25 m2 quadrats per plot just prior to cover crop termination in fall and spring. Dry matter production and total N content was determined (LECO high temperature combustion) for this biomass as well as early season corn (5-leaf stage) and pre-nodule soybeans (two true leaf stage). We used a lab incubation to evaluate N mineralization rates in contrasting soils (Evesboro loamy sand, Psammentic Hapludult or Mattapex silt loam, Aquic Hapludult) amended with forage radish, rape, or rye root or shoot residues. Amended soil was incubated at 25.0 ± 0.2 °C for 1, 2, 4, 8, 16, 32, and 48 days before determining the CO2 evolved and the NH4+ and NO3- produced. We hypothesized that the rapid decay of forage radish residues in winter would result in a large flush of mineral N in early spring and that leaching would move this flush from the 0-15 cm layer to the 15 to 30 cm layer soon thereafter. However, mineral N levels were higher at 0-15 cm than at 15-30 on all sampling dates. At all research sites and most sampling dates, cover crop had no effect on surface soil NH4+ or SON. There was a significant cover crop treatment effect on surface soil NO3- on most sample dates (Fig 1). Forage radish plots were significantly higher in NO3- than the rape, rye and no cover plots as early as February (~6 weeks after forage radish frost-killed), but concentrations did not exceed 8 mg kg-1. In May, NO3- in rape plots exceeded that of rye at most sites, following kill of both in late April. Soil in Brassica plots generally exceeded that in Rye plots in June, though NO3- levels overall were moderate to low (<20 mg kg-1). Fall N uptake by roots and shoots of forage radish (190 kg N ha-1) significantly exceeded N uptake by rape (135 kg N ha-1). N uptake by rape shoots did not change from fall to spring, and significantly exceeded N uptake by rye shoots during both periods. Young corn plants (5-leaf stage) following decomposition of forage radish and rape had higher dry matter N content compared to those following rye at most sites. Brassica root and shoot material decomposed more quickly than rye root material, based on CO2 evolution over 8 days of incubation (Fig. 2, data from duplicate samples of each of two soils). A related study is evaluating the N leaching risk presented by winter-killed, decomposing forage radish in late winter and early spring months. Additional research should investigate whether a forage radish and rye interplanted combination might be most ideal for successful N budgeting and management.