Estimating soil nitrate leaching of nitrogen fertilizer from global meta-analysis.

Global estimates of soil nitrate leaching of applied nitrogen (N) in agricultural production systems are not imprecise; however, the results of some field experiments have suggested that nitrate leaching responds exponentially rather than linearly to increasing N inputs. In this study, we compiled field data on nitrate leaching for 324 site-year combinations extracted from 86 peer-reviewed articles to test the hypothesis that in response to N fertilizer addition, soil nitrate leaching emission factors (EF) do not remain constant, but rather increase rapidly with increasing rates of N application. The averaged overall results showed that the proportional change in the EF response to increasing N input (ΔEF, %) was 0.042. Because this value was positive and significantly different from zero, our results demonstrate that EF is not constant, but rather increases with N addition. We compared our ΔEF response pattern with other constant EF patterns, and found that the 30% EF estimate of the Intergovernmental Panel on Climate Change and the 19% EF used in another published model were too high. A global estimate of soil nitrate leaching in arable uplands calculated with our ΔEF pattern was substantially lower than other estimates calculated with other procedures. In conclusion, our ΔEF pattern provided a globally applicable procedure for more accurate estimation of leaching loss and potential environmental costs incurred by various N application gradients.