Journal of Integrative Agriculture 2017, 16(0): Available online at ScienceDirect

Transcription

1 Journal of Integrative Agriculture 2017, 16(0): Available online at ScienceDirect RESEARCH ARTICLE Nitrogen mobility, ammonia, and estimated leaching loss from long-term manure incorporation in red soil HUANG Jing 1, 2, 3, DUAN Ying-hua 2, XU Ming-gang 2, ZHAI Li-mei 2, ZHANG Xu-bo 4, WANG Bo-ren 2, 3, ZHANG Yang-zhu 1, GAO Su-duan 5, SUN Nan 2 1 College of Resources and Environment, Hunan Agricultural University, Changsha , P.R.China 2 Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences (CAAS)/National Engineering Laboratory for Improving Quality of Arable Land, Beijing , P.R.China 3 Red Soil Experimental Station of CAAS in Hengyang/National Observation and Research Station of Farmland Ecosystem in Qiyang, Qiyang , P.R.China 4 Key laboratory of Ecosystem Network Observation and Modeling/Yucheng Comprehensive Experiment Station, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing , P.R.China 5 USDA Agricultural Research Service, San Joaquin Valley Agricultural Sciences Center, CA , USA Abstract Nitrogen (N) loss from fertilization in agricultural fields has an unavoidable negative impact on the environment and a better understanding of the major pathways can assist in developing the best management practices. The aim of this study was to evaluate the fate of N fertilizers applied to acidic red soil (Ferralic Cambisol) after 19 years of mineral (synthetic) and manure fertilizer treatments under a cropping system with wheat-maize rotations. Five field treatments were examined: control (CK), chemical nitrogen and potash fertilizer (NK), chemical nitrogen and phosphorus fertilizer (NP), chemical nitrogen, phosphorus and potash fertilizer (NPK) and the NPK with manure (NPKM, 70% N from manure). Using the soil total N changes in 0100 cm depth, ammonia ( ), nitrous oxide (N 2 O) emission and N uptake by crops, the potential N leaching loss was estimated using a mass balance approach. In contrast to the NPKM, all mineral fertilizer treatments (NK, NP and NPK) showed increased nitrate (NO 3 ) concentration with increasing soil depth, indicating higher leaching potential. However, total loss was much higher in the NPKM (19.7%) than other mineral fertilizer treatments ( 4.2%). The N 2 O emissions were generally low (0.20.9%, the highest from the NPKM). Total gaseous loss accounted for 1.7, 3.3, 5.1, and 21.9% for NK, NP, NPK, and NPKM treatments, respectively. Estimated N leaching loss from the NPKM was only about 5% of the losses from mineral fertilizer treatments. All data demonstrated that manure incorporation improved soil productivity, increased yield, and reduced potential leaching, but with significantly higher, which could be reduced by improving the application method. This study confirms that manure incorporation Received 12 June, 2016 Accepted 28 October, 2016 HUANG Jing, Tel: , huangjing@caas. cn; Correspondence ZHANG Yang-zhu, zhangyangzhu2006@163.com; SUN Nan, sunnan@caas.cn 2017, CAAS. All rights reserved. Published by Elsevier Ltd. doi: /S (16)

2 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): is an essential strategy in N fertilization management in dryland red soil cropping system. Key words: soil NO 3 N, ammonia, nitrogen leaching, long-term field experiment, mass balance, nitrous oxide emission 1. Introduction Overuse of synthetic nitrogen (N) fertilizers due to intensification of agricultural productions has led to high losses from agricultural soils and caused damage to the environment (Erisman et al. 2007). China has become the largest N fertilizer consumer accounting for about 30% of the world s consumption since 2002 (FAO 2010). However, for a popular cropping system with winter wheat-summer maize rotation, the N use efficiency (NUE) is less than 30% (Zhao et al. 2006). Nitrogen loss has led to surface water eutrophication, ground water contamination, air quality degradation, and contributed to global warming by forming greenhouse gases. Nitrate leaching losses from soil into water not only reduces soil fertility but also causes threat to environment and human health (Cameron et al. 2013). Groundwater contamination has been reported as a major concern in northern China. A survey of NO 3 concentration in 600 groundwater samples showed that about 45% of the samples exceeded the drinking water standard of 50.0 mg L -1 proposed by major developed countries with the highest reported concentration reaching 113 mg L 1 (Zhang et al. 2004). Increased concentration and mobility of NO 3 in soil profile indicates high risk of leaching and groundwater contaimination. In wheat-maize fields, NO 3 N in the 0 90 cm soil layer was found to accumulate above 200 kg N ha 1 at N application rate of 553 kg N ha 1 yr 1 (Ju et al. 2006). The N leaching loss in subtropical areas such as the red soil region in southern China was expected to be worse than northern China because of higher precipitation. However, we have insufficient data to validate (Xu et al. 2010). Sun et al. (2008) found about 16.8% of N fertilizer applied at 150 kg N ha 1 yr 1 was leached in a rain-fed peanut rape rotation system in an acidic red soil. Long et al. (2012) found that pig manure applied at 150 kg N ha 1 yr 1 did not result in elevated NO 3 concentrations in soil, and addition of lime with high manure incorporation rate (600 kg N ha 1 yr 1 ) in surface soil layer (0 15 cm) had no significant effect to reduce NO 3 concentrations in soil below ( cm). This potentially indicates that N leaching can be significant, if manure application rate is too high. Ammonia ( ) is one of the major N losses from soil fertilization. High loss is caused by a chemical reaction shifting from ammonium (NH 4+ ) to at high ph (NH 4+ +OH +H 2 O). The worldwide losses range from 10 to 19% (average 14%) of the used N fertilizers (Ferm 1998). In China, high has been reported from calcareous soil (high soil ph) in the Northern China Plain. The losses were 30 39, 11 48, and 1 20% of total N applied to rice (urea or ammonium bicarbonate), maize (urea) and wheat (urea), respectively (Cai et al. 2002). Measurements in paddy soils (i.e., under flooded conditions) showed that accounted for % of urea-n applied from different N fertilizers and application methods (Zhang et al. 2011). Ammonia losses from rice paddies were generally lower ( %) under different combined irrigation systems, including non-flooding and wetdry cycles. The losses were also lower under different nutrient managements, including compound fertilizer treatment with ammonium bicarbonate or urea and/or control released urea (Xu et al. 2012). Limited studies have assessed under dryland farming (e.g., wheat, maize) in red soil. One of the studies reported was in the range of % when kg ha 1 urea was applied in red soil under dryland farming with the crop rotation of Smooth Crabgrass (Digitaria ischaemum) in spring and Winter Radish (Raphanus sativus) in autumn (Zhou et al. 2007). These values were much lower than those from the calcareous soils in different regions and paddy soils in the same region. More accurate assessment on the loss is needed from dryland red soil. In addition to, other volatile forms of N include nitrous oxide (N 2 O), dinitrogen (N 2 ), and nitrogen oxides (NO x, nitric oxide (NO) and nitrogen dioxide (NO 2 )). N 2 O is a potent greenhouse gas with the warming potential ~300 times greater than the equivalent mass of CO 2. It also contributes to the destruction of stratospheric ozone (Cicerone 1987). NO is a precursor of NO 2. It reacts with O 2 and further with water to form nitric acid (HNO 3 ), which is a major component of acid rain. The NO x can also react with volatile organic compounds under sunlight to form ozone, a ground level air pollutant that is a respiratory hazard and a greenhouse gas (Williams et al. 1992). Understanding N loss in gaseous forms has significance in N cycling and developing management practices. More attention has been paid to and N 2 O because of their role in mass and global warming. The other forms of gaseous N are minor except N 2, which could become significant under anaerobic

3 4 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): conditions such as flooded paddy soil, but insignificant in dry land production (Mosier et al. 1989). The N 2 O and NO are intermediate products of nitrification and denitrification in soil (Sahrawat and Keeney 1986). N 2 O is primarily produced from nitrification process at low and moderate soil moistures and denitrification processes at soil moisture containing over 70% water-filled pore space due to a decreased O 2 supply (Ruser et al. 2006). Thus, in non-irrigated dryland conditions, nitrification is expected to be the dominant process in N 2 O production, which results in much lower N 2 O emissions than those from high soil moisture conditions in the laboratory (Cai et al. 2016). Field data also indicate that the gas loss from denitrification (mainly as N 2 and N 2 O) can be as high as 33% of applied N in paddy field (Zhu et al. 1989), but substantially lower in dryland production of maize and wheat (0.8 2% of applied N) (Cai et al. 2002). Annual loss of N 2 O and NO from wheat-maize rotation system under different fertilization (urea or organic fertilizers) treatments was reported in the range of 0 1.7% (Akiyama et al. 2004; Cui et al. 2012). The red soils (equivalent to Ferralic Cambisol in the World Soil Classification by FAO) cover about 2 million ha in tropical and subtropical regions in southern China (Xu et al. 2003). The productivity of the red soils is low especially in dryland crop production (e.g., wheat and maize) due to low soil ph. To increase grain yield, high amount of chemical fertilizers have been used in the last few decades, which resulted in serious environmental problems, including intensified acidification and reduction in soil productivity. To address these issues and develop nutrient management strategies in this region, a long-term field experiment was established in 1990 in southern China to determine the effects of various fertilization (including synthetic fertilizer and manure) regimes on crop response and soil nutrients in a wheat-maize rotation system in red soil. Previous studies have reported crop yield, N uptake, N 2 O emissions and soil properties (Duan et al. 2011; Zhai et al. 2011). A recent research has used a mass balance approach to estimate the total N loss (including both and leaching loss) to the environment (Duan et al. 2016). In the present study, we adopted this approach to further quantify N losses via specific pathways. Specifically we aim to: 1) determine amount during the maize and wheat growing seasons; 2) examine NO 3 -N distribution or mobility in soil profiles; and 3) estimate total gaseous loss and the potential leaching loss by integrating all available data. 2. Materials and methods 2.1. Experimental site This study was conducted on a red soil at the Qiyang Experimental Station ( N, E) of the Chinese Academy of Agricultural Sciences, Qiyang, Hunan Province, China. The annual rainfall, sunshine hours, and average temperature were mm, h and 18.6 C, respectively. The top soil (0 20 cm) of the experimental field at the beginning of the long-term experiment (i.e., in 1990) had a soil organic carbon of 8.5 g kg 1, total N of 1.1 g kg 1, total P of 0.5 g kg 1, total K of 13.3 g kg 1, available N of 79 mg kg 1, available P of 11 mg kg 1, available K of 122 mg kg 1, and bulk density (BD) of 1.26 g cm Field treatment and experimental design Five fertilizer treatments investigated in this study were: control (CK), inorganic N with K (NK), inorganic N with P (NP), combined inorganic N, K, and P (NPK), and NPK with manure (NPKM, 70% N applied was from manure). All treatments were established in 1990 and repeated annually until 2009 (total 19 years). The soil profiles were sampled for N analysis every year. The treatments were duplicated and plots were laid out in a randomized block design. Each plot size was 196 m 2 (19.6 m 10 m). The N fertilizer was supplied as urea and pig manure at an annual rate of 300 kg N ha 1, P as single superphosphate at 53 kg P ha 1, and K as KCl at 100 kg K ha 1. The amount of pig manure to provide 70% of total N (210 kg N ha 1 from pig manure) was calculated from the total N content in the dry manure (average 16.7 g kg 1 ). The annual application of fresh pig manure was 41.7 t ha 1. For the annual supply, 30% of the total amount of fertilizer was applied for wheat and 70% for maize. All mineral fertilizers and manure were applied as basal fertilizer before crop planting. To simulate local farmers practices, both mineral fertilizers and manure were applied by banding at a depth of 10 cm, followed by sowing of crop seeds, and then covering with soil. Wheat (cultivar Xiangmai 4) was seeded in November and was harvested in early May each year. Maize (cultivar hybrid Yedan 13) was seeded between wheat rows in late March and harvested in middle of July each year. The period after maize harvest before the next season s wheat seeding was referred to fallow period (about 3.5 mon) NO 3 N distribution in soil profile and determination of To evaluate N leaching potential from various fertilization regimes, soil samples were collected from the top 1 m profile in fall 2008 after harvesting maize, and NO 3 N concentration in the profile was determined. To collect a representative sample in the plot, five cores of soil were randomly collected in each plot using a 5 cm i.d. auger and combined. The samples were separated into 020,

4 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): , 4060, 6080 and cm depths. Fresh soil samples were used for soil NO 3 N and NH 4+ N analysis. The well mixed 10 g fresh soil samples were extracted with 50 ml distilled water. After shaking for 30 min, the suspension was filtered and analyzed for NO 3 -N and NH 4+ -N by phenoldisulfonic acid and indophenol blue colorimetry, respectively (Bao 2000). The amount of NH 4+ N in each extract was found in the range of % (average 5.0%) of total extracted mineral N, which means each extract contained >95% NO 3 -N. Ammonia rate was measured from 2009 to 2010 during wheat and corn growing seasons, using a continuous airflow chamber method as described in Tian et al (1998). Immediately following fertilizer applications, a PVC chamber base (20 cm in diameter, and 5 cm in height) was inserted about 5 cm deep into the soil, between crop rows at the center of each plot and the base top was fitted with a water trough. During measurement, a cylindrical chamber (20 cm in diameter, and 10 cm in height) fitted with two plastic tubes was inserted into the water trough at the top of the chamber base. The chamber air inlet was a plastic tube (25 mm i.d.) at a height of 2 m above the soil surface. The background air concentration was monitored with an inlet at a height of 2 m above the soil surface. The volatilized into each chamber was collected through the outlet plastic tubing (10 mm i.d.) immersed into the bottom of a ml glass erlenmeyer flask, which contained 250 ml of 2% boric acid solution to trap. The air flow through a group of multiple chambers was controlled by a flow meter (air flow through each chamber was 0.8 L s 1 ). Sampling time for each measurement was 2 h (between 2:00 and 4:00 pm) when it likely represented the daily average flux according to Tong et al. (2009). The dissolved in the acid solution was determined by titration with sulfuric acid (0.005 mol L 1 H 2 SO 4 ) in the laboratory (Xu et al. 2013). The from all treatments was corrected for the background air concentration in each measurement. Cumulative (kg N ha 1 d 1 ) during each (maize or wheat) growing season was calculated as follows: ( Cumulative N emission = F +F ) j i (t 2 j t i ) 24 (1) Where, F i and F j are the flux (kg N ha 1 h 1 ) in the i and j day, t i and t j are the time (day) between any of the two sampling events (i and j) after fertilizer application. The for each treatment was measured daily following fertilizer application for about two weeks. After two weeks, measurements were made occasionally for other times during the year because little differences were found between any of the fertilizer treatments and the control. The differences in emission between fertilizer treatments and the control were considered to be the result of fertilizer application Nitrogen mass balance and estimate of N leaching loss Nitrogen balance (N input -N export by crop) has been used to estimate the risk of N losses from arable land (Constantin et al. 2010). However, the loss estimated using this method only reflects plant use efficiency. To more accurately assess N loss to the environment from fertilization regime, Duan et al. (2016) incorporated soil N status change into the mass balance equation and estimated the total loss as follows: N loss to the environment =N from fertilizer and manure +N input from the environment N uptake by crops N change in soil (2) Where, N loss to the environment is all losses to the environment including gases and leaching, N from fertilizer and manure is N from fertilizer applications, N input from the environment is N via soil N mineralization, N fixation, precipitation, etc., which was estimated from the plant uptake in the control (CK) that no N fertilizer was applied, N uptake by crops is the nitrogen output from soil by crops uptake, N change in soil refers to changes in total N (TN) content. In this study, soil for total N analysis and soil bulk density (S BD ) were collected in the fall of 2009, i.e., after 19 years of fertilizer treatment; thus N mass balance was conducted for the period of The annual soil N change rate was calculated from: Soil N storage change (kg ha 1 y 1 )= (3) (TN 2009 TN 1990 )/( ) S BD S s S h 10 3 Where, TN 1990 or TN 2009 is the soil N content (g kg 1 ) determined at the beginning of the experiment, i.e., in 1990 or the end of the study period (2009), S BD represents soil bulk density (g cm 3 ), S s is the area (ha) and S h is the soil depth (m). In this study, we chose the boundary of 1 m soil depth to define the rooting zone, i.e., soil N in top 1 m soil is available for crops and the N below this layer is subjected to leaching because the majority of maize and wheat roots are found in the surface soil (Shi et al. 2012). The total N loss to the environment estimated in eq. (2) include all possible pathways including runoff/erosion, leaching,, and all other gaseous (N 2 O, NO x, N 2 ) losses. The field was flat and had little runoff and erosion. In this study, we collected new data on for each crop growing season. Previous investigations have determined N 2 O emissions (Zhai et al. 2011), while other gaseous (N 2, NO x ) losses were not measured. However, these losses were minor based on the studies below. Under the dryland cropping conditions, the N 2 O/(N 2 O+N 2 ) ratio averaged 0.5 in dryland soil (Schlesinger 2009). The NO x (mainly NO) can be estimated in relation to N 2 O with a NO/ N 2 O ratio, which is about 0.4 in fertilized cropland (Stehfest and Bouwman 2006). So other gaseous (N 2, NO x ) losses were estimated as 1.4 times N 2 O emission. Under these assumptions and with all available data, eq. (4) can be rearranged to estimate the potential leaching:

5 6 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): N leaching loss =N from fertilizer and manure +N input from the environment N plant uptake N increase in soil N NH3 (4) N N N2 O emission other gaseous loss (NO x +N 2 ) Where, all units in eq. (4) are kg N ha 1 y 1. All the data were measured in our experiments, except for the minor gas loss such as NO x and N 2 were estimated based on available literature (Stehfest and Bouwman 2006; Schlesinger 2009). The and N 2 O emissions were measured at different years from the total soil N determination. However, we assume they represent annual loss and consider that the same treatment was applied to the same plot and there were no extreme weather conditions in terms of temperature during the entire study period, although climate was relatively drier in 2009 and 2010 (Fig. 1). The fate of N fertilizer applied during the 19 years of applications was evaluated in percent of total N applied. For example, N uptake by plant Plant uptake (%)= 100% (5) N total input Where, total N input was defined in eq. (2). Soil N storage, leaching loss, and (, N 2 O emissions, and other gases) were calculated using the similar expression as eq. (5). Although not all parameters were determined in the same year due to a number of logistic reasons, we assumed that the measurements would represent the annual average. This allowed the estimates of N loss via different pathways from the production system that can assist in developing effective N management practices for sustainable agricultural production. To the best of our knowledge, this is the first attempt to predict N leaching loss with most data measured in the field Data analyses Annual rainfall (mm) Annual rainfall Annual temperature Year Fig. 1 Trends of annual rainfall and temperature over the entire study period ( ) at the long-term experimental site in Qiyang, Hunan Province, China Annual tempeture ( C) Statistical analyses on the data of or N 2 O and soil NO 3 N were performed using SAS 9.2 (SAS Institute 2008). Because there were only two replicates due to limitations on the field size and preference to using large plots, the analysis power is limited. The analysis relies on the fact that if the error variance is small and the treatment effect size is large, significance is still possible (Steel and Torrie 1980; Milliken and Johnson 1989). Thus the purpose of performing such statistical analysis in this study was to provide some useful information on treatment effects with available data. For the or N 2 O data, a model based on the randomized completed block design was used to fit the treatment effects and produce residual diagnostics. The means separation between the different treatments was performed using Tukey s adjustment. For the soil profile NO 3 -N data that indicate N mobility, a SAS PROC MIXED program was used to fit a mixed model with repeated measures. The treatment, soil depth, and their interaction are the fixed effects and the replications and treatments replications are the random effects. This random effect defines the experimental units for incorporating a first order, autoregressive covariance structure among the repeated measures on the soil depths. Log-transformed data was used in order to improve residual diagnostics. The soil depth treatment interaction was significant (P<0.001). So, the corresponding least square means and 95% confidence intervals were produced, which were back transformed to the original units of measurement. These are reported in the results. 3. Results 3.1. Nitrate movement in soil profile To examine soil N mobility after long-term fertilization regimes, the NO 3 contents in the cm soil profile after 18 years of different fertilizer treatments were analyzed and Soil depth (cm) Soil NO 3 -N content (mg kg 1 ) CK NP NK NPK NPKM Fig. 2 Soil NO 3 -N content in soil after 18 years of fertilizer treatments after maize harvest in October CK, control; NP, inorganic nitrogen with phosphorus; NK, inorganic nitrogen with potassium; NPK, inorganic nitrogen, phosphorus and potassium combination; NPKM; NPK with manure. Error bars denote the lower and upper values at the 95% confidence level generated by a PROC MIXED program.

6 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): the results are shown in Fig. 2. Soil NO 3 N concentration in the control (CK) was the lowest in the profile, ranging from 1.4 to 5.6 mg kg 1. It also exhibited some downward movement from surface to 40 cm depth or below. The NO 3 contents in fertilizer treatments at each depth were all significantly higher than those in the CK (P<0.05). All the mineral fertilizer treatments (NK, NP and NPK) showed a trend of increasing NO 3 concentrations with increasing soil depth, although an exceptionally high NO 3 N concentration (70.7 mg kg 1 ) was observed in the NK treatment at cm soil depth. Soil NO 3 contents in NPKM treatment were the lowest among all fertilizer treatments and unlike other treatments, the NO 3 content below 60 cm depth was significantly lower than those above (P<0.05) indicating less downward N movement Dynamics of during wheat and maize growing seasons The, measured following the fertilizer applications during the maize and wheat growing seasons, is shown in Fig. 3. The highest rate was observed on the first day, ranging from 2.3 to 10.4 kg N ha 1 d 1 (Fig. 3-A) with the highest from the NPKM treatment. After the initial high values, the rates dropped substantially (especially for the NPKM treatment) and continued to decrease until no differences were observed from the control in two weeks. Beyond two weeks, until wheat harvest, the average rate of CK, NK, NP, NPK and NPKM in wheat growing season was about 1.2, 1.3, 1.1, 1.3 and 1.2 kg N ha 1 d 1, respectively, with no significant differences among these values. The rate during the maize growing season (Fig. 3-B) showed a similar pattern as that during the wheat growing season (Fig. 3-A). The highest rates (up to 11.4 kg N ha 1 d 1 ) were observed during the first two days following fertilizer applications and then declined continuously. Among the treatments, the NPKM had the highest rates, but declined at a faster rate with time, compared with the rates during the wheat growing season. For all the mineral fertilizer treatments (NK, NP, NPK), the rates were almost the same as that during the wheat growing season, in terms of both value and decreasing rate. The rate for NPKM treatment, 10 days after application, was still substantially higher than the mineral fertilizer treatments, but reduced to 1.8 kg N ha 1 d 1 15 days later, which was not significantly different from other treatments. The data from both growing seasons indicated significant loss during a short period of time, immediately following fertilizer applications. At or after two weeks until wheat harvest, during A Ammonia flux (kg N ha 1 d 1 ) Ammonia flux (kg N ha 1 d 1 ) B CK NP NK NPK NPKM Days after fertilization (d) CK NP NK NPK NPKM Days after fertilization (d) Fig. 3 Ammonia rate from different treatments following fertilizer application in wheat growing seasons (A) and maize growing seasons (B). CK, control; NP, inorganic nitrogen with phosphorus; NK, inorganic nitrogen with potassium; NPK, inorganic nitrogen, phosphorus and potassium combination; NPKM; NPK with manure. Error bars denote the lower and upper values at the 95% confidence level generated by a PROC MIXED program. the maize growing season, the average rate ranged from 1.2 to 1.3 kg N ha 1 d 1 with no significant differences among the five treatments. Cumulative losses from the CK, NK, NP, NPK, NPKM treatments during the wheat and maize growing seasons were 1.0, 2.6, 1.1, 6.0, 18.2 kg N ha 1 and 0.7, 1.0, 6.9, kg N ha 1, respectively. The losses during maize growing season were generally higher than those during wheat growing season for all the fertilizer treatments. The application of manure with mineral fertilizer resulted in the highest or significantly higher than those from all other mineral fertilizer treatments (Table 1). There were no significant differences in loss between the mineral fertilizer treatments (NK, NP and NPK) and the non-fertilized control The fate of N fertilizer applied and estimates of N loss Using the mass balance approach, the fate of N fertilizer applied including total N storage change in soil, N uptake by

7 8 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): crops and N loss to the environment after 19 years of different fertilizer treatments is shown in Table 2. With the same amount of total N input, there were significant differences in plant uptake, due to the effects of fertilizer treatments on yield (Duan et al. 2011). Soil N storage change ( %) was also significantly different among the treatments. The total N loss from mineral fertilizer treatments (NK, NP, and NPK) was about twice as that from NPKM, which had the lowest total N loss. The total N loss to the environment was broken down to gaseous loss and leaching loss (Table 2). Among the gaseous losses, the measured was the highest in mass, accounting for 65 90% of total loss. Ammonia loss was generally higher than the N 2 O emissions with the highest loss from NPKM. The N 2 O emissions were also the highest from NPKM. Both and N 2 O emissions from NPKM were significantly higher than other treatments (P<0.05). Based on Stehfest and Bouwman (2006) and Schlesinger (2009), the other minor gas emissions (N 2, NO, and NO x ) were estimated to be about % of total N input. Total gaseous N loss accounted for 1.7, 3.3, 5.1, and 21.9% for NK, NP, NPK, and NPKM, respectively. The potential N leaching loss was estimated Table 1 Treatment effect comparisons (P-value) for total loss during winter wheat and maize growing season Treatment 1) CK NK NP NPK NPKM CK NK NP NPK NPKM ) CK, control; NP, inorganic nitrogen with phosphorus; NK, inorganic nitrogen with potassium; NPK, inorganic nitrogen, phosphorus and potassium combination; NPKM; NPK with manure. P<0.05 means significant difference between treatments. Results indicate that from NPKM treatment was significantly higher than all other mineral N fertilization treatments. to be the lowest (2.1%) from NPKM compared to 35% or higher loss from all other mineral fertilizer treatments. These data show the relative differences among the treatments. However, the absolute percentage loss might be underestimated due to potential overestimation of from the method used (See Discussions). 4. Discussion 4.1. High NO 3 mobility and its accumulation in soil from mineral fertilizer applications Nitrate concentration data, determined after 18 years of fertilization, showed that the NPKM had the lowest soil NO 3 N content in the cm soil profile among all treatments and significantly lower concentrations (P<0.05) in the depths below 60 cm than those above (Fig. 2). There are significant positive correlations among the population sizes of ammonia-oxidizing bacteria (AOB) and ammonia-oxidizing archaeon (AOA), soil ph, and potential nitrification rates in NO 3 production (He et al. 2007). The presumably more abundant AOB and AOA in NPKM did not appear to result in higher NO 3 than the mineral fertilizer treatments. The NO 3 distribution in soil profile appeared to be more affected by the N source and plant uptake. Fig. 2 suggests that the combined applications of mineral fertilizers with manure, decreased soil NO 3 N accumulation, which was partially due to higher N uptake and higher accumulation of organic N, especially in the surface soil. Higher yield from the NPKM treatment was attributed to the soil improvement on ph (Duan et al. 2011). The mineral fertilizer treatments, especially NK and NP treatments, showed strong acidification due to nitrification, with ph between , compared to 6.3 from the NPKM. This severely retarded plant growth, reduced crop yields and N uptake, increased NO 3 N accumulation in the rooting zone and enhanced mobility to deeper soil layer, which was shown by the higher concentration in lower depths in the profile. The NPKM had Table 2 Fertilizer nitrogen balance from 19 years of different fertilization treatments Treatment 1) Fertilizer N Environment N Plant Total N storage N NH uptake change (01 m) 3 2 O Other gas Potential N emissions emissions leaching N input (kg N ha 1 yr 1 ) 2) N output (%) 3) CK NK b 0.2 c NP b 0.3 b NPK b 0.4 a NPKM a 0.9 a ) CK, control; NK, inorganic nitrogen with potassium; NP, inorganic nitrogen with phosphorus; NPK, inorganic nitrogen, phosphorus and potassium combination; NPKM; NPK with manure. 2) Nitrogen input from the environment was estimated by the plant N uptake in the control (CK) without N fertilizer application. 3) Plant uptake data were from Duan et al. (2016); the N 2 O emission data were from Zhai et al. (2011); other gas emissions (N 2, NO, NO x ) were estimated as 1.4 times N 2 O emission based on Stehfest and Bouwman (2006), Schlesinger (2009).

8 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): the highest total N in surface soil, but most of them were in organic form (Duan et al. 2016), which indicated increased soil N storage or improved soil fertility. The repeated use of mineral fertilizer resulted in much higher NO 3 N in the soil rooting zone. In the red soil subtropical region, seasonal and annual nitrate leaching was found to differ with fertilizer application and precipitation. The monthly precipitation and nitrate leaching losses were significantly correlated (Long et al. 2012). Other studies have also supported that the combined application of fertilizers with manure, increase grain yield and reduce soil NO 3 N accumulation in soil, compared with the application of chemical fertilizers alone (Dauden and Quilez 2004; Yang et al. 2004) High loss from manure application Gaseous N loss was dominated by, accounting for % of the total gas loss. The highest (near 20% of total N input) was determined from the NPKM (Fig. 3 and Table 2) among all fertilizer treatments. During each wheat or maize growing season, the rate peaked on the first or second day (Fig. 3) and then decreased gradually with no differences among the various treatments after about two weeks. Similar observations were reported by Zhang et al (2011). The from a maize field increased quickly after fertilizations, with peak emissions during the first 1 4 days and most losses were measured in less than 7 days. In our study, the rates after the peak declined at a slower rate during the maize growing season, compared with the rates during the wheat growing season. This might be due to the fact that 70% of annual N was applied during the maize growing season and the average temperature and accumulated rainfall during the measurement period in maize (14.4 C and mm) were higher than the measurement period in wheat (8.2 C and 22.2 mm). The highest loss from manure was most likely due to its higher NH 4 + content, which is common in most fresh manure with high ph (~8.8) that favors formation and subsequent loss to the atmosphere. The from mineral fertilizer treatments (NK, NP, and NPK) was much lower and accounted for only % of total N input, which could be explained largely by lower soil ph ( ). This result is similar to the measurement in an earlier study in dryland red soil under smooth crabgrass-winter radish rotation, where accounted for % of the total N applied ( kg N ha 1 yr 1 ) (Zhou et al. 2007). Soil ph has significant effect on the abiotic (Dewes 1996). Even within the narrow ph range ( ) among the different treatments in this study, there is a significant positive correlation between loss and soil ph (y=21.0x86.8, where y is cumulative loss (kg N ha 1 ) and x is soil ph, R 2 =0.55, P<0.05). Application method may be another reason for high from NPKM. In this study, to simulate farmers practices, the mineral fertilizers with manure were mixed and buried at about 10 cm depth from the surface. Huijsmans et al. (2003) has shown that deep placement could reduce by 20% compared to surface incorporation and by 75% compared to surface spreading in 3 soil types (sand, sandy loam and clay). Thus, deeper placement of manure should be promoted to reduce losses in the red soil. Cumulative N 2 O emissions were generally lower than (Table 2). Although its loss is insignificant in mass ( % of total N input), minimizing N 2 O emissions is necessary because of its much stronger global warming potential than other greenhouse gases. The NPKM treatment also resulted in twice or higher N 2 O emissions than other treatments, which is consistent with Meng et al. (2005). In a sandy loam soil under wheat-maize rotation, the relative N 2 O emissions were 0.24, 0.21, 0.15 and 0.15% of applied N in M, NPK, NP and NK treatments, respectively (Meng et al. 2005). The high N 2 O loss from manure could be due to enhanced microbial activity as N 2 O is produced from both nitrification and denitrification processes. On the other hand, N 2 O may originate from the hot spots in soil induced by the organic C and N of the manure when denitrification and nitrification processes are simulated (Dambreville et al. 2008) Estimated N leaching loss - high from mineral fertilization and low from manure incorporation Significant differences were found in both total N loss to the environment and the estimated N leaching loss based on the field data collected from the 19 years of mineral fertilizer treatments and manure incorporation (Table 2). The total N loss (including both gaseous loss and leaching) to the environment were % of total N input for various fertilizer treatments. These values were obtained after taking soil N storage changes into the mass balance equation, thus more accurately reflecting the N loss and impact on the environment (Duan et al. 2016). The total loss was determined by both soil N storage changes and plant uptake. The manure incorporation accumulated up to 27.0% soil N and so did the NK and NP treatments ( %), but by totally different means. Manure incorporation leads to organic build-up with nutrient storage in surface soil (Duan et al. 2016) and the N accumulation in NK and NP treatment was most likely due to low NUE because of lower yield. The plant N uptake from NPKM (49%) was times higher (higher yield) than other mineral fertilizer treatments (NK, NP, and NPK) (Table 2). The NPK application had higher

9 10 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): yield or N uptake, but similar total N loss as compared to the NK and NP treatments because of lower accumulation in the rooting zone. The data indicate that multiple factors were affecting plant uptake that resulted in the different N loss. The accumulation of N in red soil was significantly high, indicating lower rate or split applications of N fertilizer during the growing seasons is better than all of the fertilizer N as a basal dressing once. Organic N in manure can be incorporated into the soil organic matter pool, and can be converted to NH 4 + through mineralization or absorbed by soil microorganisms for their growth. The manure N is likely more efficient during crop growing season because it only becomes available to plants through slow mineralization to NH 4 + and NO 3. Although there is a competition by microbes, a combination of manure with mineral fertilizer would prevent this potential problem. In China, soil organic matter has been found to correlate with cereal crop productivity and yield stability across several provinces (Pan et al. 2009) and combined application of organic fertilizer can promote immobilization of fertilizer N to reduce N loss (Liu et al. 2009). Our study was able to quantify the total N loss from the wheat-maize production system into gaseous (mainly as ) based on field measurements and leaching loss. However, it should be noted that the losses measured from the field may only provide relative differences among the treatments, i.e., do not necessarily represent absolute losses because of the method utilized. We used the continuous airflow chamber method to measure the, but the air flow rate in the chamber might be exceeding the ambient flow rate that likely caused overestimation of. These may lead to overestimation in leaching loss. Nonetheless, the results showed a much lower leaching loss from the NPKM treatment (2.1%) than those from the mineral fertilization ( % of total N input) (Table 2). The NO 3 N data (Fig. 2) in soil profile supports this assessment by showing a much more downward movement of NO 3 in mineral N fertilizer treatments and much lower concentration and mobility in NPKM treatment. The long-term experiment in this study has clearly shown a positive impact of manure incorporation on crop yield (plant uptake) and soil total N storage as well as reduction in leaching loss. The significantly lower leaching loss from NPKM treatment than all other mineral fertilizer treatments is a proof of the multiple benefits of manure on soil productivity and plant growth. Many studies have reported that manure resulted in good establishment of crop with efficient nutrient uptake and improvement on soil physical properties (Stenberg et al. 2012). Our study has demonstrated that the integrated benefit of manure incorporation resulted in significantly reduced N leaching loss. 5. Conclusion The aim of agricultural N management is to provide sufficient N to plants to maximize crop growth and yield, as well as minimize environmental impact. Our results have shown that long-term use of mineral N fertilizer (300 kg N ha 1 y 1 ) will not only lead to reduced crop yield or low plant N uptake, but also high leaching risk (estimated about % of total N inputs). When combining mineral N fertilizers with manure to provide 70% total N (NPKM), crop yield and soil N storage were significantly improved and had the lowest leaching loss (2.1%). Although the losses (19.7%) and N 2 O emissions (0.9%) were substantially higher from the manure incorporation than mineral fertilizer treatments (NP, NK, and NPK), improving the application method such as deeper placement in soil, could reduce the gaseous loss. Overall, this research has demonstrated that incorporation of manure with chemical fertilizer (NPKM) is one of the most effective fertilization management strategies in dryland red soil for sustainable crop production. Acknowledgements We appreciate Mr. Bruce Mackey (USDA, Agricultural Research Service, Albany, CA) for his advice and help in data analyses. We also appreciate Mr. Tom Pflaum from USDA Agricultural Research Service, San Joaquin Valley Agricultural Sciences Center for his editorial help. This work was supported by the National Key Research and Development Program of China (2016YFD ), the open fund of Key Laboratory of Non-point Source Pollution Control, Ministry of Agriculture, China ( ) and the National Key Technologies R&D Program of China (2012BAD14B04). References Akiyama H, Mctaggart I P, Ball B, Scott A N 2 O, NO and emissions from soil, after the application of organic fertilizers, urea and water. Water Air and Soil Pollution, 156, Bao S D Analysis of Soil Agrochemistry. 3th ed. Chinese Agriculture Press, Beijing, China. (in Chinese) Cai G X, Chen D L, Ding H, Pacholski A, Fan X H, Zhu Z L Nitrogen losses from fertilizers applied to maize, wheat and rice in the North China Plain. Nutrient Cycling in Agroecosystems, 63, Cai Z, Gao S, Hendratna A, Duan Y, Xu M, HansonB D Key factors, soil N processes, and nitrite accumulation affecting nitrous oxide emissions. Soil Science Society of American Journal, 80, Cameron K C, Di H J, Moir J L Nitrogen losses from the soil/plant system: A review. Annals of Applied Biology, 162,

10 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): Cicerone R J Changes in stratospheric ozone. Science, 237, Constantin J, Mary B, Laurent F, Aubrion G, Fontaine A, Kerveillant P, Beaudoin N Effects of catch crops, no till and reduced nitrogen fertilization on nitrogen leaching and balance in three long-term experiments. Agriculture, Ecosystems and Environment, 135, Cui F, Yan G X, Zhou Z X, Zheng X H, Deng J Annual emissions of nitrous oxide and nitric oxide from a wheat maize cropping system on a silt loam calcareous soil in the North China Plain. Soil Biology and Biochemistry, 48, Dambreville C, Morvan T, Germon J C N 2 O emission in maize-crops fertilized with pig slurry, matured pig manure or ammonium nitrate in Brittany. Agriculture, Ecosystems and Environment, 123, Dauden A, Quilez D Pig slurry versus mineral fertilization on corn yield and nitrate leaching in a Mediterranean irrigated environment. European Journal of Agronomy, 1, Dewes T Effect of ph, temperature, amount of litter and storage density on ammonia emissions from stable manure. Journal of Agricultural Science, 127, Duan Y H, Xu M G, Gao S, Liu H, Huang S M, Wang B R Long-term incorporation of manure with chemical fertilizers reduced total nitrogen loss in rain-fed cropping systems. Scientific Reports, 6, Duan Y H, Xu M G, Wang B R, Yang X Y, Huang S M, Gao S D Long-term evaluation of manure application on maize yield and nitrogen use efficiency in China. Soil Science Society of America Journal, 75, Erisman J W, Bleeker A, Galloway J, Sutton M S Reduced nitrogen in ecology and the environment. Environmental Pollution, 1, FAO (Food and Agriculture Organization) FAOSTAT Database Collections. FAO, Rome. Ferm M Atmospheric ammonia and ammonium transport in Europe and critical loads: A review. Nutrient Cycling in Agroecosystems, 1, He J Z, Shen J P, Zhang L M, Zhu Y G, Zheng Y M, Xu M G, Di H J Quantitative analyses of the abundance and composition of ammonia-oxidizing bacteria and ammoniaoxidizing archaea of a Chinese upland red soil under longterm fertilization practices. Environmental Microbiology, 9, Huijsmans J F, Hol M G, Vermeulen G D Effect of application method, manure characteristics, weather and field conditions on ammonia from manure applied to arable land. Atmospheric Environment, 26, Ju X T, Kou C L, Zhang F S, Christie P Nitrogen balance and groundwater nitrate contamination: Comparison among three intensive cropping systems on the North China Plain. Environmental Pollution, 143, Liu M Q, Hu F, Chen X Y, Jiao J Organic amendments with reduced chemical fertilizer promote soil microbial development and nutrient availability in a subtropical paddy field: The influence of quantity, type and appli cation time of organic amendments. Applied Soil Ecology, 42, Long G Q, Sun B Nitrogen leaching under corn cultivation stabilized after four years application of pig manure to red soil in subtropical China. Agriculture, Ecosystems and Environment, 1, Meng L, Ding W X, Cai Z C Long-term application of organic manure and nitrogen fertilizer on N 2 O emissions, soil quality and crop production in a sandy loam soil. Soil Biology and Biochemistry, 11, Milliken, G A, Johnson D E Analysis of Messy Data, Volume II, Nonreplicated Experiments. Van Nostrand Reinhold, New York. Mosier A R, Chapman S L, Freney J R Determination of dinitrogen emission and retention in floodwater and porewater of a lowland rice field fertilized with 15 N-urea. Fertilizer Research, 19, Pan G, Smith P, Pan W The role of soil organic matter in maintaining the productivity and yield stability of cereals in China. Agriculture, Ecosystems and Environment, 129, Ruser R, Flessa H, Russow R, Schmidt G, Buegger F Emission of N 2 O, N 2 and CO 2 from soil fertilized with nitrate: Effect of compaction, soil moisture and rewetting. Soil Biology & Biochemistry, 38, Sahrawat K L, Keeney D R Nitrous oxide emission from soils. Advances in Soil Science, 4, SAS Institute SAS 9.2. SAS Int., Cary, NC. Schlesinger W H On the fate of anthropogenic nitrogen. Proceedings of the National Academy of Sciences of the United States of America, 106, Shi Z L, Jing Q, Cai J, Jiang D, Cao W X, Dai T B The fates of N 15 fertilizer in relation to root distributions of winter wheat under different N splits. European Journal of Agronomy, 40, Steel R G, Torrie J H Principles and Procedures of Statistics, a Biometrical Approach. 2nd ed. McGraw-Hill, New York. Stehfest E, Bouwman L N 2 O and NO emission from agricultural fields and soils under natural vegetation: summarizing available measurement data and modeling of global annual emissions. Nutrient Cycling in Agroecosystems, 73, Stenberg M, Ulen B, Soderstrom M, Roland B, Delin K, Helander C A Tile drain losses of nitrogen and phosphorus from fields under integrated and organic crop rotations. A four-year study on a clay soil in southwest Sweden. Science of Total Environment, 434, Sun B, Chen D L, Li Y, Wang X X Nitrogen leaching in an upland cropping system on an acid soil in subtropical China: Lysimeter measurements and simulation. Nutrient Cycling in Agroecosystems, 3, Tian G M, Gao J L, Cai Z C, Ren L T Ammonia from winter wheat field top-dressed with urea. Pedosphere, 4,

11 12 HUANG Jing et al. Journal of Integrative Agriculture 2017, 16(0): Tong H J, Zheng X H, Wang R, Zhou Z X, Yue J, Liu C Y, Li M, Liang W G, Dong H B A preliminary study of measurement of from cropland using quasi-dynamic chamber. Climatic and Environmental Research, 4, (in Chinese) Williams E J, Hutchinson G L, Fehsenfeld F C NO x and N 2 O emissions from soil. Global Biogeomineral Cycles, 6, Xu J Z, Peng S Z, Yang S H, Wang W G Ammonia losses from a rice paddy with different irrigation and nitrogen managements. Agricultural Water Management, 104, Xu L G, Zhang Q, Huang L J Nitrogen leaching in a typical agricultural extensively cropped catchment, China: Experiments and modelling. Water and Environment Journal, 2, Xu M G, Li D C, Li J M, Qin D Z, Yasukazu Hosen, Shen H P, Cong R H, He X H Polyolefin-coated urea decreases ammonia in a double rice system of southern China. Agronomy Journal, 1, Xu R K, Zhao A Z, Li Q M, Kong X L, Ji G L Acidity regime of the red soils in a subtropical region of southern China under field conditions. Geoderma, 1, Yang S M, Li F M, Sukhdev M, Wang P, Suo D R, Wang J G Long-term fertilization effects on crop yield and nitrate nitrogen accumulation in soil in northwestern China. American Society of Agronomy, 3, Zhai L M, Liu H B, Zhang J Z, Huang J, Wang B R Long- Term Application of organic manure and mineral fertilizer on N 2 O and CO 2 emissions in a red soil from cultivated maize-wheat rotation in China. Agricultural Sciences in China, 11, Zhang J S, Zhang F P, Yang J H, Wang J P, Cai M L, Cheng F L, Cao C G Emissions of N 2 O and, and nitrogen leaching from direct seeded rice under different tillage practices in central China. Agriculture, Ecosystems and Environment, 12, Zhang Y Y, Liu J F, Mu Y J, Pei S W, Lun X X, Chai F H Emissions of nitrous oxide, nitrogen oxides and ammonia from a maize field in the North China Plain. Atmospheric Environment, 45, Zhao R F, Chen X P, Zhang F S, Zhang H L, Schroder J, Römheld V Fertilization and nitrogen balance in a wheat-maize rotation system in North China. American Society of Agronomy, 4, Zhou J, Cui J, Wang G Q, He Y Q, Ma Y H Ammonia in relation to N application rate and climate factors in upland red soil in spring and autumn. Acta Pedologica Sinica, 3, (in Chinese) Zhang W L, Wu S X, Ji H J, Kolbe The current situation and controlling manures of non-point source pollutions in China. Scientia Agricultura Sinica, 37, (in Chinese) Zhu Z L, Cai G X, Simpson J R, Zhang S L, Jackson A V, Chen D L, Freney J R Process of nitrogen loss from fertilizers applied to flooded rice fields on a calcareous soil in north-central China. Fertilizer Research, 18, (Managing editor SUN Lu-juan)