Introduction
Crop production is very sensitive to climatic change, as modified rainfall and air temperature patterns can seriously affect plant physiology with negative impact on productivity and yield (Gornall et al., 2010). On the other hand, agriculture contributes to global change with the emission in atmosphere of greenhouse gases (GHGs).
Agricultural soil management practices can contribute for about 60% of the total nitrous oxide (N2O) emissions from the agricultural sector (Smith et al., 2007). The increase of nitrogen fertilisation to satisfy the demand of high nitrogen consuming crops has enhanced the biogenic N2O emissions due to nitrification and denitrification processes, driven by temperature, nitrogen availability and soil moisture, this latter influenced by irrigation.
In this scenario, it is essential to define sustainable management practices to minimise the impact of cropping systems on the global environment while maintaining the yield. In this frame, several different approaches have been proposed in terms of appropriate management of irrigation (Vallejo et al., 2014; Meijjde et al., 2016; Wolff et al., 2016) as well as of fertilisation, such as the use of fertilisers added with nitrification inhibitors (NIs) (Vitale et al., 2013; Abalos et al., 2016; Hube et al., 2016; Huerfano et al., 2016; Vitale et al., 2017) and organo or organomineral fertilisers (Ball et al., 2004; Guardia et al., 2016; Vitale et al., 2017). The latter release nitrogen (N) more gradually over the course of season, compared to conventional mineral fertilisers, thus improving synchrony between soil N availability and crop N demand and potentially increasing overall efficiency of fertiliser use. NIs delay nitrate production by depressing the activity of Nitrosomonas bacteria, thus allowing nitrogen to accumulate in the soil, making it available for plant growth in longer time period. Organo and organo-mineral fertilisers also release nutrients progressively throughout the late stages of the crop development, due to the slow process of mineralisation of the organic matter.
Both fertilisation types thus potentially improve crop growth and yield but have different effects on N2O production in the soil. The use of NIs mitigates soil N2O emission by suppressing microbial activity for a period of time depending on soil type moisture and temperature (Soares et al., 2015). On the contrary, the application of organic manure or sludge to soil has been showed to enhance N2O emission as a consequence of the increased soil moisture and the degradable labile carbon addition. These soil conditions might favour microbial growth and metabolism, thus resulting in N2O production (Kaiser and Ruser, 2000; Sanchez-Martin et al., 2008; Yao et al., 2015).
Among factors involved in crop management, irrigation regime also deserves considerable attention, as it has been shown to foster GHG emissions (Tost et al., 2013). In particular, water supply can also increase N2O emissions favoring conditions for nitrification and denitrification (Maris et al., 2015), while a rational management of irrigation can enhance crop yield and mitigate the impact of cropping systems on environment and global climate by favoring Carbon sequestration and reducing GHGs emission. Most of the studies on the effects of NIs and organic and organomineral fertilisers on crop production and N2O emission has been carried out under optimal irrigation conditions (Vitale et al., 2013; Guardia et al., 2016; Huerfano et al., 2016; Vitale et al., 2017), while little information is available for open field crops under reduced water supply (Albalos et al., 2016).
The main objective of this study was to investigate the efficiency of three fertiliser types on crop growth and soil N2O emission under optimal and reduced water supply. In particular, we tested a mineral fertiliser added with 3,4-dimethylpyrazole phosphate (DMPP) - a new generation nitrification inhibitor, and an organomineral fertiliser. Our study moves from the consideration that organic matter mineralisation and the resulting release of nutrients in the soil requires adequate soil moisture conditions, so that its decomposition could be limited under low soil water content affecting crop growth and N2O emissions. On the other hand, laboratory studies have shown that DMPP is efficient to mitigate NH4+ oxidation also under reduced soil water potential (Barth et al., 2008). This should benefit plant growth and yield while mitigating nitrous oxide production also under reduced soil moisture.
Materials and methods
Experimental site, crop management and experimental set-up
The field trial was carried out in Ponticelli (province of Naples, southern Italy, at 50 m a.s.l.; 40°86’N, 14°33’E), a location characterised by Mediterranean climate conditions with warm dry summer and mild wet winter (Figure 1).
The soil has a sandy-loam texture and relative content for the 0.0-0.9 m soil layer are: sand 80%, silt 12%, clay 8%, bulk density 1.37 g cm–3; Chemical characteristics for the layer 0.0-0.1 m are: 2.54% organic matter content, total carbonate absent, 0.149 dS m–1 electrical conductivity, pH 7.08 and 1.86 g kg–1 nitrogen. The experimental site is equipped with a meteorological station that allowed estimating the reference evapotranspiration (ET) according to the Hargreaves equation that was locally calibrated, tested and validated.
Tomato seedlings (Solanum lycopersicum L.) were transplanted on May 3rd 2012 in 12 m2 plots at a spacing of 1.0x0.3 m. A total of 18 plots were arranged in a split plot experimental design with two water regimes (factor A) and three fertilisation treatments (factor B). Three different fertilisation treatments were applied as follows: ammonium nitrate (Control, M plots), ammonium sulfate nitrate added with nitrification inhibitor (3,4 dimethylpyrazole phosphate, Entec 26) (DMPP plots), and organo-mineral fertilisation using dried pellets (NPK + animal waste, Olivas) (OM plots). Different fertiliser amounts were supplied, to allow for balancing N amount among treatments. A total 120 Kg N ha–1 was applied to all plots, split in two times: 50% at transplanting and 50% 30 days later. Phosphate and potash supplies were also balanced by supplying supplemental inputs of superphosphate and potassium-sulphate fertilisers to treatments M and OM, one week after the initial N fertilisation event.
The experimental area was watered by drip irrigation. One meter spaced 4 L h–1 drip lines were installed, with drippers spaced 0.30 m. All plots were well watered until 13 June (41 Days After Transplanting, DAT) receiving a water volume of 176 m3 ha–1. Thereafter, 9 plots were well watered by replenishing the 100% of ETc (100% ET), with total water supply of 1083 m3 ha–1, whereas the other plots received only the 50% of ETc (50% ET), with a total water supply of 541.5 m3 ha–1. Kc coefficients used to calculate ETc were 0.8, 1.2, and 0.9, respectively for initial, middle and end season (Rinaldi and Rana, 2004).
Fruits were periodically hand-picked, as they reached maturity, and final harvest took place on July 27 (82 DAT).
Biometrical and physiological measurements
Plant biomass and leaf area were determined at 82 DAT, by sampling three plants per plot. Due to the sandy texture of the soil, it was possible to recover most of the root system when pulling up. Plants were therefore partitioned in root, leaf, stem, and fruits. All plant parts were transferred in an oven at 60°C up to constant weight. Green leaf area was determined by means of an area mater (Li-3000, Licor Inc. Lincoln, NE, USA). Nitrogen (N) content was determined in stems and leaves on three plants per plot. Samples were grounded into a fine powder by an agate mortar and pestle (Fritsch pulverisette). N concentration was measured by gas chromatography (CNS analyser - Thermo Finning).
Nitrogen use efficiency (NUE) was estimated as the ratio between crop yield and applied N, whereas N uptake was estimated by multiplying total biomass weight by N tissue content. Maximal photochemical efficiency of photosystem 2, i.e. Fv/Fm, related to functionality of photosystem 2 (Maxwell and Johnson, 2000), was determined on 30 min darkened leaves by using a fluorometer (Junior PAM, Walz GmbH, Effeltrich, Germany).
Cumulative soil N2O emissions
N2O emissions were measured by means of the static chamber technique; 20 cm diameter and 10 cm heigh chambers were placed on collars previously inserted 3 cm into the soil the previous day or before, to minimise disturbance. Air samples, collected before and three times following chambers closure in a time window of 30 min, were picked up by means of a polypropylene syringe, and stored in 20 mL vials. Gas samples were then analysed by gas chromatography (SRI 8610C, Gas Chromatograph, Torrance, USA) using a 63Ni electron-capture detector. Gas fluxes were calculated by linear regression of gas concentrations over time. Only curves where the regression slope did not change sign over the observation period, i.e. dC/dt|t=0 / dC/dt|t=30 > 0 (Stolk et al., 2009), were considered in results. Cumulative fluxes were calculated by summing the products of the average of two neighboring measurement fluxes by their interval time (Maucieri et al., 2016).
Statistical analysis
Statistical analysis of data was performed by means of the Sigma-Plot package (Sigma-Plot 12.2). Differences among fertilisation treatments and water regimes were checked by two-way ANOVA followed by the Duncan’s test. Spearman’s correlations were performed to assess the soil-related independent variables on soil gas emissions.
Results and discussion
Plant growth and yield
In Table 1 are summarised the ANOVA results. Shoot biomass of well-watered (100% ET) plants treated with organo-mineral (OM) fertiliser was greater (P<0.05) compared to plants fertilised with mineral (M and DMPP) fertiliser (Figure 2A). No significant differences in root dry matter production was observed (Figure 2B). The higher carbon gain in OM plants under well-watered conditions is likely due to the higher (P<0.01) total leaf area, which increased light interception (Figure 2D) as previously reported by Vitale et al. (2017) on tomato gown upon same conditions, rather than to an enhanced unit photosynthesis rate or to higher photosynthetic pigments and/or proteins content. In fact, the leaf-nitrogen content (Table 2), also related to Rubisco content (Makino, 2003) and Fv/Fm values (Table 3), related to functionality of photosystem 2 (Maxwell and Johnson, 2000), were comparable among treatments, thus suggesting that the unit photosynthetic rate was the same for all treatments.
At 50% ET, M and OM plants showed a reduction (P<0.01) in shoot biomass, that was more pronounced in OM (Figure 2A). On the contrary, shoot biomass was not affected by limited water supply under DMPP fertilisation while a significant (P<0.05) increase in root biomass was detected (Figure 2B). As a consequence, the shoot/root ratio (SRR) decreased in M and OM plants but was unaffected in DMPP plants (Figure 2C). SRR offers an estimation of the distribution of dry matter among the different plant organs. The reduction in SRR in M and OM plants indicates that the proportion of dry matter allocated to shoots was lower compared to those diverted to roots. Reduced shoot-root ratio is a common adaptive reaction of plants to withstand water stress (Boutraa et al., 2010), and is aimed at reducing water loss by transpiration while exploring deeper layers of the substrate in search of water. A reduction (P<0.01) in leaf area in both M and OM plants was observed, particularly evident in OM plants (Figure 2D), that represents an effective mean to control water loss by transpiration under limited water supply. On the other hand, the reduced leaf area in response to water shortage also explains the reduced carbon gain (i.e. the lower dry matter production) in M and OM plants as a consequence of a reduced light interception.
Water shortage is known to determine also a reduced nutrient uptake by roots; it is well known that the capacity of roots to absorb nutrient declines in water stressed plants, due to either a decrease in transpiration rate (passive uptake) and/or to an impairment of the transport and root membrane permeability (active uptake) (Alam, 1999). A reduction in passive nutrient uptake by roots, likely as a consequence of reduced plant transpiration rates (Youssef et al., 2012) due to a reduced total leaf area, may be assumed for M and OM plants, as indicated by the lower N content in water stressed plants compared to well-watered plant (Table 3), leading to a reduction in fruit yield (Figure 2E). Instead, the high efficacy of DMPP in mitigate NH4+ oxidation during the fast growth of plants, when wheatear conditions were unfavorable for a fast degradation of 3,4-dimethylpyrazole phosphate molecular (soil temperature below 20°C, data not shown), allowed to maintain a good plant growth under limited water supply. Barth et al. (2008) reported, also on a sandy soil, as DMPP efficiently inhibits nitrification under limited soil water content, thus allowing high NH4+ accumulation in the soil that will be readily available for plant growth (Peet et al., 1985). As a consequence, DMPP plants enhanced crop N uptake (i.e. total biomass weight multiplied by N tissue content) and NEU (i.e. the ratio between crop yield and applied N) leading, under water shortage, to a higher fruit yield compared to M and OM plants (Figure 2E).
Cumulative N2O fluxes
During the studied period, ranging from May to July 2012, the highest and lower cumulative N2O fluxes (in absolute value) were found respectively in M plots and DMPP plots under well waterd conditions, corresponding to about 230 kg ha–1 CO2 eq and 80 kg ha–1 CO2 eq. This values do not result dangerous for environment; thus it can be stated that crops managment in this study was so to derterm a low impact on global environmental.
Cumulative N2O emission was lower (P<0.05) in DMPP compared to M and OM treatments under condition of adequate water supply (Figure 3), highlighting the effectiveness of DMPP in mitigating nitrification, the main process involved in N2O production. On the other hand, N2O fluxes were similar between M and OM plots (Figure 3). Field studies report higher N2O emissions in response to animal manure fertilisation with respect to synthetic N fertilisers (Bouwman et al., 2002). In our study, this difference could be related to the higher NEU and N uptake by plants grown under organo-mineral fertilisation compared to those grown under mineral fertilisation, as the latter could suffer from a reduced soil N availability, a critical factor for biological N2O production.
We observed a reduction in cumulative N2O fluxes under conditions of limited water supply in M and OM plots (Figure 3). Soil moisture is one of the main factors affecting N2O production and a reduced water content in the soil has been reported to reduce N2O emission, as also confirmed by the positive relationship between N2O fluxes and Water Filled Pore Spavce (WFPS) (Table 4). The reduction of N2O emissions under water scarsity in M and OM plots could be related to the reduced availability in the soil of N in a form readily available for biological N2O production. On the contrary, a significant increase in cumulative N2O fluxes was observed in DMPP plots (Figure 3) under water shortage, likely due to a reduced capacity of 3,4-dimethylpyrazole phosphate to mitigate NH4+ oxidation, a consequence of increased soil temperature under conditions of reduced soil moisture. It is well known that the 3,4- dimethylpyrazole phosphate (DMPP) degradation is faster under soil temperature above 20°C, as found in this study during late spring and summer (data not shown). In fact, a positive corrrelation between N2O fluxes and soil temperature (Tsoil) as well as a negative corrrelation between N2O fluxes and WFPS in DMPP plots were observed (Table 3), evidencing the reduced power of DMPP in limiting NH4+ oxidation. Nevertheless, soil N2O emission values measured in DMPP plots under water scarsity were lower in absolute term compared to those measured in M and OM plot recorded under well-water irrigation. As expressed in CO2 eq they were about 180 kg ha–1, not dangerous for global environment.
Conclusions
Crops management determined low N2O emissions from soil resulting in a low impact on global environment as refferred to emissions expressed in CO2 eq.
Organo-mineral fertilisation proved a more efficient nutrient source compared to mineral fertilisation, leading to a significant improvement in crop yield. It is well-known that organic colloids strongly holds mineral nutrients, that are progressively released when organic matter is degraded. This provides plants with nutrients at the later stages of the crop development. On the other hand, mineral fertilisers containing nitrates makes N readily available already during the early vegetative growth stages. However, the mineral fertiliser added with nitrification inhibitor (DMPP) reduced marcately the N2O emission.
The reduced performance of organo-mineral fertiliser under water shortage is a likely consequence of unfavourable soil environmental conditions for the mineralisation of the organic component. The environmental conditions during the course of the experiment were such as to limit NH4+ oxidation by DMPP, with consequent benefits on crop N uptake and yield but not to mitigate soil N2O emssions that, however, were lower in absolute term compared to M and OM values recorded under well-water irrigation. Thus, it can be stated that, under reduced water supply, DMPP added-fertiliser could represent a better nitrogen source compared to conventional and organo-mineral fertilisers able to preserve crop yield and to determine soil N2O emissions (as expressed in CO2 eq) not dangerous for global environment.
