Influence of drip and furrow irrigation systems on nitrogen oxide emissions from a horticultural crop

https://doi.org/10.1016/j.soilbio.2008.02.005Get rights and content

Abstract

Irrigation management has an important influence on emissions of nitrous oxide (N2O) and nitric oxide (NO) from irrigated agricultural soils. In order to develop strategies to reduce the emission of these gases, a field experiment was carried out to compare the influence of different irrigation systems: furrow (FI) and drip-irrigation (DI), on N2O and NO emissions from a soil during the melon crop season. Two fertilizer treatments were evaluated for each irrigation regime: ammonium sulphate (AS) as a mineral N fertilizer, at a rate of 175 kg N ha−1; and a control without any N fertilizer (Control). On plots where the AS treatment was applied, drip irrigation reduced total N2O and NO emissions (by 70% and 33% respectively) with respect to values for furrow irrigation. This was probably due to the lower amount of water applied and the different soil wetting pattern associated with DI. Dry areas of the drip-irrigated plots emitted a similar amount of N2O to the wet areas (0.45 kg N2O-N ha−1) in the Control and greater quantities in the AS treatment (0.92 kg N2O-N ha−1 for dry and 0.70 kg N2O-N ha−1 for wet areas). We suggest that the N oxide pulses observed throughout the irrigation period on DI plots could have been the result of frequent increases in the soil wetting volume after the addition of water. Denitrification losses (from depths of 0–10 cm) were estimated at 11.44 kg N2O- N ha−1 for the AS treatment under FI and at 4.96 kg N2O-N ha−1 for DI. Under DI, nitrification was an important source of N2O, whereas denitrification was the most important source under FI. The addition of NH4+ and the use of DI enhanced the N2O/N2 ratio of gases produced through denitrification. The quantity of dissolved organic C (DOC) in the soil generally decreased with addition of NH4+.

This work showed that, in comparison with furrow irrigation, drip irrigation is a method that can be used to save water and mitigate emissions of the atmospheric pollutants NO and N2O.

Introduction

The water regime is one of the key factors that influences nitrous oxide (N2O) and nitric oxide (NO) emissions from agricultural soils. Soil moisture controls the biotic and abiotic processes involved in the production, consumption and diffusion of N2O and NO within the soil. These gases are predominantly produced by microbial processes and are by-products of nitrification or intermediate products of denitrification (Firestone and Davidson, 1989). Although these processes generally take place in the soil at the same time, denitrification rates rapidly increase when water filled pore space (WFPS) exceeds 60%, predominating over nitrification, due to a decrease of O2 supply. On the other hand, nitrification may contribute significantly when WFPS is within the range 30–70% (Davidson, 1991, Granli and Bockman, 1994). Furthermore, the water regime controls the amount of N leached, N mineralization, and other processes such as nitrate movement within the soil, all of which influence the spatial distribution of mineral N in the soil and consequently the production of N oxides.

In irrigated soils, the amount of water applied and its distribution onto the soil are factors that affect WFPS temporally and spatially and consequently the processes producing emissions of N oxides. Under rainfall or sprinkler irrigation, infiltration is mainly vertical, and water distribution is quite homogeneous onto the soil (Mualem and Assouline, 1996). However, other irrigation systems, such as furrow (FI) or drip irrigation (DI), strongly affect the soil water distribution (Allen et al., 1998) because vertical and lateral infiltration are produced at the same time. Furrow irrigation is one of the most common surface irrigation practices because it is suitable for many row crops and for crops that cannot stand in water for long periods (e.g. 12–24 h). Under FI, water is applied at a high rate on furrows to promote an adequate lateral infiltration of water, trying to attain a homogeneous soil moisture distribution within the soil (Allen et al., 1998). Water flows through the macropores, temporarily creating a high proportion of anaerobic microsites, which probably favours denitrification. In practice, after irrigation most of the soil volume is wetted, although a small proportion of the ridge may remain dry.

In arid and semiarid areas, surface drip irrigation (DI) has gained widespread popularity as an efficient and economically viable alternative (Vázquez et al., 2006), because it offers the potential to increase water and N use efficiency (Vázquez et al., 2005). As dripping water is applied at very low rates (2–10 l h−1), the wetting front moves by capillarity action and only a small soil volume around the emitters has a water content close to saturation during irrigation (Vázquez et al., 2005). After the first irrigation a significant soil volume remains dry, although during each irrigation event the wetting front advances slightly into the soil. The volume and shape of the soil wetting zone mainly depends on the soil hydraulic properties and the amount and rate of water application (Li et al., 2003). Increasing the quantity of water increases horizontal and vertical infiltration and consequently the volume of wetted soil. However, a moisture gradient is always produced in the wetted area, from the emitter (higher than field capacity) to the wetting front (lower than field capacity), independent of the amount of water applied (Li et al., 2003, Vázquez et al., 2005). Therefore, we suspect that this wide range of WFPS may favour denitrification and nitrification at the same time in the wetted areas of soil. Although the effect of WFPS in these processes has been very well documented (Davidson, 1991, Granli and Bockman, 1994), the influence of this special soil wetting pattern on N oxide emission is not known. To date, no information exists about the emission of N oxides and the relative importance of each process in drip-irrigated soils.

Our hypothesis was that the irrigation system will affect the relative importance of nitrification and denitrification and thereby N oxide emissions, due to the different water patterns of furrow and drip irrigation. Under drip irrigation conditions, the application of water at localized points at a low water flow probably favours more aerobic than anaerobic microsites, and nitrification may be an important source of N oxides. The advance of the wetting front may also produce pulses of N oxides from the drip-irrigated areas, similar to those observed after rainfall when soil was dry (Scholes et al., 1997). Although the amount of water applied is often lower with drip irrigation than with furrow irrigation, which could affect denitrification and nitrification processes, we suspect that the difference in the distribution of water between systems may produce the most important differences in the pattern of N oxide emission between the irrigation systems. The objectives of this study were, therefore, to compare NO and N2O emissions from different irrigation systems (furrow and drip) applied to a melon crop fertilized with mineral N, and to detect how these different methods affected the emission pathways (nitrification and denitrification) of these gases. The results of this experiment would help to establish strategies to mitigate N oxide emissions from irrigated agricultural soils.

Section snippets

Soil characteristics

The field experiment was carried out at the ‘El Encín’ Field Station, near Alcalá de Henares (Madrid) (latitude 40° 32′N, longitude 3° 17′W), in the middle of the Henares river basin. The soil was a Calcic Haploxerepts (Soil Survey Staff, 1992) with a loamy texture in the upper (0–28 cm) horizon. Some physico-chemical properties of the top 0–28 cm of the soil layer, measured by conventional methods, were: total organic C, 8.2 ± 0.4 g kg−1; pHH2O, 7.3; bulk density, 1.4 ± 0.1 g cm−3; CaCO3, 13.1 ± 0.3 g kg

Environmental conditions, evolution of mineral N and soluble organic carbon

The WFPS of the upper 10 cm of the soil profile was 16% immediately after the application of fertilizers (Fig. 2). During the irrigation period (15th June to 14th September), mean WFPS remained between 65 and 83% under the FI system and between 56 and 75% for the wet soil areas under the DI system (Fig. 2). Mean WFPS was generally higher in FI than DI plots. Emitters generated circular wet areas on plots, which increased in size during the irrigation period. After the second irrigation event,

Effect of irrigation type on mineral N and dissolved organic carbon in soil

The drip irrigation system reduced the amount of water required (45%) with respect to the furrow irrigation system. Sharmasarkar et al. (2001) also observed a 43% reduction in the application of irrigation water when DI was used instead of FI to irrigate a sugar beet crop. This reduction was possible because a significant area of the soil surface remained temporarily dry when drip irrigation was used (Fig. 1), and also because more deep drainage probably occurs under furrow than drip irrigation

Conclusion

This study demonstrated that selection of irrigation regime had a major influence on N2O and NO emissions. Drip irrigation saves water and also helps to mitigate the production of N oxides, especially N2O. It can therefore be used as an efficient tool for reducing the gases lost from fertilized soils in arid and semiarid regions. We noted that the different patterns of water distribution of furrow and drip irrigation directly affected nitrification and denitrification, the main pathways leading

Acknowledgments

The authors are grateful to the Spanish Commission of Science and Technology (CICYT) and the EU (through the Nitroeurope IP Project) for financing this research and to the IMIA (Instituto Madrileño Investigación Agraria) for permitting the use of the experimental field. It is a pleasure to acknowledge Roberto Sainz for his help with installing the different irrigation systems, Ana Ros, Silvia Dominguez, Paloma Martín and Ana María Gomez for their technical assistance and Miguel Quemada and Mark

References (31)

  • I.C. Anderson et al.

    Relative rates of nitric oxide and nitrous oxide production by nitrifiers, denitrifiers and nitrate respirers

    Applied and Environmental Microbiology

    (1986)
  • L. Cárdenas et al.

    Effect of soil moisture, temperature, and inorganic nitrogen on nitric oxide emissions from acidic tropical savannah soils

    Journal of Geophysical Research

    (1993)
  • E.A. Davidson

    Fluxes of nitrous oxide and nitric oxide from terrestrial ecosystems

  • J. Dick et al.

    The effect of rainfall on NO and N2O emissions from Ugandan agroforest soils

    Phyton: Annales Rei Botanicae

    (2001)
  • J.M. Estavillo et al.

    Short term effect of ploughing a permanent pasture on N2O production from nitrification and denitrification

    Plant and Soil

    (2002)
  • Cited by (109)

    View all citing articles on Scopus
    View full text