Effects of controlled drainage on N and P losses and N dynamics in a loamy sand with spring crops

Abstract The effects of controlled drainage on N and P losses from soil were examined in a 4-year field drainage experiment on a loamy sand in Southern Sweden. Of the three plots (0.2 ha each), one was drained by conventional subsurface drainage (CD), and two by controlled drainage (CWT1 and CWT2). The groundwater level in the CWT plots was naturally drained to at least 70 cm below the soil surface during the vegetation period between early spring and harvest but allowed to rise to 20 cm below the soil surface during the rest of the year. Measurements of precipitation, drain outflow, weir depths and air and soil temperatures were carried out hourly. Groundwater levels were measured and samples of drain outflow for analyses were collected twice a month. Mineral N contents in soil were measured three times a year and grain yields and N uptake in crops after harvest. Controlled drainage significantly lowered N and P loads in drain outflow and altered N dynamics in soil. The relative decrease in N and P loading in drain outflow from CWT plots, compared with CD, was of the same magnitude as the reduction in drain outflow rate (60–95%). The high-risk periods for N losses coincided with periods of high outflow rates and high mineral N content in soil, for all years and for all treatments. Positive correlations in N concentrations in drain outflow were found between drainage systems. In contrast, high-risk periods of P losses from CD did not exclusively occur in months with highest outflow rates. In 3 out of 4 years, peak P loads occurred in late winter and P concentrations were positively correlated to soil temperature. In CWT, peak P loads coincided with highest outflow rates and no correlations were found between P concentrations and soil temperature. CWT plots, compared to CD, improved N efficiency for applied fertiliser due to lower N loads in drain outflow and higher N uptake by crop. The yields in CWT were 2–18% larger and the crop uptake of N increased by 3–14 kg ha −1 . The net changes of N (changes in mineral N content in soil, NO 3 -N loading in drain outflow, N fertiliser application, N uptake in crops) showed a surplus during autumn in all plots, with the highest surplus in CWT due to higher mineralisation rates and lower N loads. During the winter season, a surplus in measured net changes of N was found in CD in all years. In contrast, the net changes of N in CWT showed a deficit, probably due to a lower mineralisation rate and non-measured N sinks such as denitrification, immobilisation and deep seepage loads.