N2O emissions from urine-treated tropical soil: Effects of soil moisture and compaction, urine composition, and dung addition

Abstract

Increasing attention is being paid to the importance of N2O emissions due to livestock activities in tropical countries. Understanding the key variables driving N2O emission could help minimize impacts of N2O release and improve the accuracy of N2O inventories. We aimed to investigate the effects of soil moisture, soil compaction, urine composition, urine volume, and dung addition on N2O emissions from a urine-treated tropical Ferralsol under controlled conditions. Manipulated soil conditions (e.g., moisture content, compaction, and dung addition) affected N2O emissions when varying quantities of urine-N (p = 0.02) were applied (urine volumes remained equal) and when varying urine volumes (p = 0.04) were applied (quantities of urine-N remained equal). When the amount of urine-N applied was varied, the estimated N2O emission factor (EF) was 3.14 ± 0.70%, 2.29 ± 1.25%, 3.90 ± 0.64%, 4.73 ± 0.88%, and 6.62 ± 1.10% for moist soil, dry soil, compacted soil, plus dung, and plus dung and compacted soil treatments, respectively. While varying the volume of urine, the estimated N2O EF was 4.96 ± 1.66%, 4.27 ± 1.42%, 3.99 ± 1.19%, 6.50 ± 0.35%, and 7.37 ± 0.76% for moist, dry soil, compacted soil, plus dung, and plus dung and compacted soils treatments, respectively. The urine-N concentration influenced N2O emissions (p = 0.02) [which decreased linearly (p = 0.062)] as well the volume of urine (p < 0.01) [which increased linearly (p < 0.01)]. The chemical form of the applied urine-N (urea, nitrate, or ammonium) did not affect N2O emissions and the emissions factor averaged 1.40 ± 0.38%. N2O production was affected by the KCl concentration in the urine (p < 0.01), and the effect was curvilinear. The key driving factor affecting N2O emissions was soil moisture content. The N2O response varied when the urine volume differed (in both moist and dry soil conditions), and with the addition of dung.