Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Nitrogen balance dynamics during 2000-2010 in the Yangtze River Basin croplands, with special reference to the relative contributions of cropland area and synthetic fertilizer N application rate changes

  • Lijuan Wang,

    Roles Formal analysis, Methodology, Software, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliations State Key Laboratory of Urban and regional Ecology, Research Center for Eco-Environmental Sciences, University of Chinese Academy of Sciences, Beijing, China, University of Chinese Academy of Sciences, Beijing, China

  • Hua Zheng ,

    Roles Conceptualization, Data curation, Funding acquisition, Investigation, Resources, Supervision, Validation

    zhenghua@rcees.ac.cn

    Affiliations State Key Laboratory of Urban and regional Ecology, Research Center for Eco-Environmental Sciences, University of Chinese Academy of Sciences, Beijing, China, University of Chinese Academy of Sciences, Beijing, China

  • He Zhao,

    Roles Validation

    Affiliations State Key Laboratory of Urban and regional Ecology, Research Center for Eco-Environmental Sciences, University of Chinese Academy of Sciences, Beijing, China, University of Chinese Academy of Sciences, Beijing, China

  • Brian E. Robinson

    Roles Investigation, Validation

    Affiliation Department of Geography, McGill University, Montreal, Québec, Canada

Nitrogen balance dynamics during 2000-2010 in the Yangtze River Basin croplands, with special reference to the relative contributions of cropland area and synthetic fertilizer N application rate changes

  • Lijuan Wang, 
  • Hua Zheng, 
  • He Zhao, 
  • Brian E. Robinson
PLOS
x

Abstract

With the increases of cropland area and fertilizer nitrogen (N) application rate, general N balance characteristics in regional agroecosystems have been widely documented. However, few studies have quantitatively analyzed the drivers of spatial changes in the N budget. We constructed a mass balance model of the N budget at the soil surface using a database of county-level agricultural statistics to analyze N input, output, and proportional contribution of various factors to the overall N input changes in croplands during 2000–2010 in the Yangtze River Basin, the largest basin and the main agricultural production region in China. Over the period investigated, N input increased by 9%. Of this 87% was from fertilizer N input. In the upper and middle reaches of the basin, the increased synthetic fertilizer N application rate accounted for 84% and 76% of the N input increase, respectively, mainly due to increased N input in the cropland that previously had low synthetic fertilizer N application rate. In lower reaches of the basin, mainly due to urbanization, the decrease in cropland area and synthetic fertilizer N application rate nearly equally contributed to decreases in N input. Quantifying spatial N inputs can provide critical managerial information needed to optimize synthetic fertilizer N application rate and monitor the impacts of urbanization on agricultural production, helping to decrease agricultural environment risk and maintain sustainable agricultural production in different areas.

Introduction

Nitrogen (N) is one of a number of critical plant nutrients that determine crop yield alongside, for example, phosphorus, potassium and a wide range of micro nutrients. The increase in consumption of world synthetic N fertilizers from10 Tg (1 Tg = 1012 g) N/y in the late 1950s to 100 Tg N/y in 2008 has played an important role in the rising rate of food production, as the global population and demand for food has increased [1]. Although the benefits of adding fertilizer N to agroecosystems are evident, costs arise in part because most of the N added to agroecosystems does not reach its ultimate aim—protein in the human diet. Only about 16% of N added as fertilizers is consumed by people [2,3], with some of the surplus N lost from the agroecosystem and potentially substantially altering downwind and downstream ecosystems [4]. Some transfers of reactive N in the form of a solution from terrestrial ecosystems (often agroecosystems) to streams, rivers, and ultimately the ocean, which will drive hypoxic zones and algal blooms and have other effects on the coastal ocean, can cause environmental problems (e.g., loss of biodiversity and eutrophication) [5] and alter the dynamics of coastal ocean systems [6]. Moreover, some portion of the soluble N leaches deep into groundwater, finally affecting human health [7]. Some transfers of reactive N are in the form of gas into the atmosphere; for example, the nitrogen oxides (NOx) formed from ammonia volatilization (NH3) and denitrification under the influence of thermal infrared radiation and ozone, contribute to the greenhouse effect and acid rain [8], and ultimately affect human health [7].

Given the serious environmental problems caused by adding fertilizer N and the balance between supply and demand for N, agricultural N management is vital for ensuring sustainable agricultural practices and environmental protection. The central target of agricultural N management is to enhance agricultural productivity to reduce famine, feed a growing population, and support the increasing demand for food, while simultaneously reducing the transfer of reactive N to non-target ecosystems. It is therefore important to understand the cycle of N in cropland, especially the inputs (e.g. synthetic fertilizers, biological N fixation, and recycled N and atmospheric deposition), outputs (e.g. N in harvested crops, denitrification, NH3 volatilization and N transported to water bodies through run-off and leaching), pathways of mobilized or retained reactive N [9], and N use efficiency (NUE: the ratio of N input and yield output of the system) [10].

In many countries, N budgets have been estimated, for different periods at various scales, including by county, watershed, region, and worldwide[1113]. Similarly, many studies have addressed different aspects and scales of the N budget issue in China, such as the Yangtze River, Haihe, and Taihu Basin [14,15]. Several studies have addressed different periods of the N budget associated with the Yangtze River Basin. Studies have estimated N biogeochemical cycling in the Yangtze River Drainage Basin[16], estimated N budgets of agricultural fields in the Yangtze River Basin from 1980–1990 at the county level [17] and focused on the N budget and its changes in 1980, 1990 and 2000 using a mass balance model over the whole basin [18]. How socio-economic factors influence N consumption and surplus was also explored [3,19,20]. However, few studies have quantified the relative contributions of cropland area and fertilizer N application rate changes for the fertilizer N input at the watershed scale, and further explained the factors responsible for these spatial changes in upper, middle and lower reaches. Quantifying the fertilizer N input change due to the changes of cropland area and fertilizer N application rate will help provide critical managerial information for decreasing watershed environmental risk and maintaining sustainable agricultural production in different areas.

We examined the N inputs, outputs and their changes during 2000–2010, based on a mass balance model using updated agricultural statistics of 856 counties in the Yangtze River Basin. Using available literature from the region, we determined the N surplus intensity (divided the N balance value of the basin by its area of cropland) and spatio-temporal changes of NUE using a geographical information system(GIS) model, which allowed us to manage various data in a common spatial framework. Then we assessed the influence and drivers of the spatio-temporal changes in the N balance. Our aim was to determine how changes in N balance and fertilizer N might be influenced by land-use changes and fertilizer N application rate and then provide explicit managerial implications of agricultural N fertilizer for different parts of the Yangtze River Basin.

Materials and methods

Study areas

The Yangtze River Basin is located in the middle of southern China (Fig 1). The Yangtze River is the largest river in China and the third largest in the world. Its mainstream length is 6380 km, total area is more than 1.81 × 106 km2 and average annual discharge into the East China Sea is 9.79 × 1011 m3 [18]. The tributaries extend to eight provinces (autonomous regions). In China, the Yangtze has long been considered in three parts(upper, middle and lower reaches), based on differences in geological conditions and hydrological systems [18]. The dividing point of the upper reaches of the river is Yichang in Hubei Province,. the middle reaches extend from Yichang to Hukou and the lower reaches are below Hukou (Fig 1).

The Yangtze River Basin is one of the most densely populated and agriculturally productive areas in China. It accounts for 19% of China's land area [16]. In 2010, 450 million people lived in the Yangtze River Basin, representing 34% of China's population. Its grain yield accounts for about 37% of the total national crop production and 70%-75% of the national rice harvest [18]. With the rapid increment of economy and population, some areas of the basin are experiencing fast urbanization and industrialization [21,22]. Additionally, the basin also faces environmental degradation to the aquatic environment, risk of flood disaster and eutrophication [23]. Most of the eutrophic lakes occur in the middle and lower reaches of the Yangtze River [24].

Calculating N balance

Because major sources or sinks are difficult to measure independently, a mass balance model was used to estimate N input and output in the croplands [25] of the Yangtze River Basin, which were the focus of this study. The N inputs include synthetic fertilizer N, biological N fixation, recycled N, and atmospheric deposition N [18,26]. The N outputs include N in harvested crops, denitrification, NH3 volatilization, and riverine N loss by leaching and run-off [17,18]. The N balance value was estimated from the difference between all inputs N and crop harvest N. We divided the N balance value of a county by its area of cropland to calculate the N surplus intensity (NSI) [15]. We divided the proportion of all N inputs by N in the harvested crops to obtain the NUE [13]. The N budget was calculated for the years 2000 and 2010 for each county of the Yangtze River Basin. The aim of our research was to calculate the inputs and outputs of N, and changes in the N balance characteristics, as well as determine the spatial distribution of NUE and NSI and defining the drivers of N inputs. This would allow us to determine how changes in N balance and fertilizer N might be influenced by land-use changes and fertilizer N application rate and so provide managerial implications of agricultural N fertilizer in the watershed.

Calculating N input

The N inputs include synthetic fertilizer N(Nchemin), biological N fixation (Nfin), recycled N (Nrecycn) and atmospheric deposition N (Natmin). Recycled N includes animal excrement N (Nlivestin), rural human excrement N (Nhuman) and crop residues used as N fertilizer (Nstrawin).

  1. Natmin
    The Natmin is the result of both natural and anthropogenic reactive N emissions, which includes wet and dry deposition in the form of inorganic and organic N [15]. However, only the bulk deposition of ammonium (NH4+) N and (NO3-) N are available because this is more easily measured [27]. We calculated bulk NH4+-N and NO3N deposition by multiplying the agricultural land area (A) and the unit area amount of bulk NH4+-N and NO3N deposition (P), using the unit area amount of bulk NH4+-N and NO3N deposition rate from a recent study [28]. The data were based on 671 data points concerning the bulk deposition of NH4+-N and NO3N in terrestrial ecosystems. We use different bulk depositions of N in the upper, middle and lower reaches of the basin in different years. In Qinghai and Xizang Province, we used 3.65 kg N/ha/y in 2000 and 5.31 kg N/ha/y in 2010 as the deposition rate. In other upper and middle reaches, we used 17.97 kg N/ha/y and 23.28 kg N/ha/y in 2000 and 2010 respectively, and corresponding in lower reaches, we used 20.95 and 26.55 kg N/ha/y.
  2. Nchemin
    In our study, Nchemin consists of pure N fertilizer (Nnitrogen) and N in compound fertilizer (Ncompin). According to the data of the China Agriculture Yearbook for 2009[29], the ratio of N, phosphorus (as P2O5) and potassium (as K2O) of compound fertilizer differed for different areas. In the Yangtze River Basin, the N: P: K ratio was 1:1:0.8. Therefore, we used 35% as the average N content of compound fertilizer [30,31].
  3. Nfin
    The growing of legumes was the main source of Nfin. The N inputs from Nfin include contributions from both symbiotic N-fixation crops (peanut and soybean) and non-symbiotic N-fixation crops (wheat and rice) [32]. The N inputs from biological fixation were equal to the product of the average N-fixation rates of each crop (Fi) and the sown area (Ai) [17,33,34]. According to previous studies of N-fixation rates in China, we used average N-fixation rates of 80 kg N/ha/y for peanut [35] and soybean, 30 kg N/ha/y for rice and 15 kg N/ha/y for wheat [30,34].
  4. Nrecycn
    Harvested crops are consumed by animals (e.g. cattle, pigs and sheep) and humans, and crop residues (straw) are used as fertilizer and for burning. In China, human and animal excrement was traditionally used as organic manure for cropland. Therefore, the annual recycled N input was the sum of the animal excrement N (Nlivestin), rural human excrement N (Nhuman), crop residue N used as fertilizer (Nstrawin). our parameters were based on the study in the Yangtze River basin [18].
    1. Animal excrement and rural human N
      The annual Nlivestin was equal to the product of the animal population quantity (Pi) and the unit N emission coefficient per animal per year (Ei).The unit N emission coefficients for pigs, cattle and sheep were 8, 40 and 5 kg N/ha/y, respectively [16,17,33,34,36].
      The annual Nhuman was equal to the product of the rural population (P) and the unit N emission coefficient per person per year (E). We calculated the rural total rural population from the rural adult population using a conversion coefficient of 0.85 and the unit N emission coefficient for a human was 5 kg N/ha/y [30]. The input from people and livestock excrement N applied to farmland as manure accounted for 40% of the total N input [17]. The remaining of livestock and human excrement N could be discharged into waterbodies, emitted by volatilization and denitrification, and left unused [37,38]
    2. Crop residue N used as fertilizer
      Many kinds of crops residues are used as fertilizer for soil in the Yangtze River Basin, however, data on crop residues are not available. Therefore, we estimated the crop residue N used as fertilizer (Nstrawin) using the straw/seed ratio (Ri), the percentage of residues returned to the soil (Si), the annual yield of crop (Yi) and the N concentration in crop residues(Pi) [17,39]. Our parameters were based on a study in the Yangtze River Basin [18].

Calculating N output

The Noutput includes N in harvested crops(Nharvout) and other loss pathways to the environment crop residues (straw) are burned (Nburn), denitrification, and NH3 volatilization and N transported to water bodies and stored in cropland).

  1. Nharvout
    The Nharvout was calculated by conversion from protein. We calculated the amount of N in the harvested crops from the yields of major crops and data on N concentrations in crops [30], using parameters based on a previous study [18].
  2. Denitrification and NH3 volatilization
    Many factors influence the emissions from denitrification and volatilization, such as tillage practices and type and rate of fertilizer. Previous experiments in the basin calculated that denitrification and volatilization originated mainly from fertilizers and human and animal wastes. Denitrification includes losses from paddy fields 36%, uplands 25%, and organic manure15% [18].
    The N losses in the agriculture from fertilizer and human and animal waste are mainly by NH3 volatilization. We omitted the N2O and NO emission because they represent a very small proportion of all inputs [38]. The NH3 volatilization includes losses from paddy fields 16%, uplands 11%, and organic manure 23%[18,40].
  3. Riverine N loss by leaching and run-off
    It is complicated to measure riverine N loss by leaching and run-off because of the numerous influence factors, such as land use type, soil type, drainage network and reservoirs. We use the average loss rate from the literatures to calculate surface run-off and leaching N. The average rates of run-off loss of chemical N applied to croplands for paddy fields and uplands were 5.2 and 11%, respectively [38]. We used averaged loss rates of 1.2 and 3.2% as the leaching loss from chemical fertilizer applied to paddy fields and uplands, respectively. All organic N discharged to water bodies as run-off and leaching were 5 and 4%, respectively [38].
  4. Crop residue combustion N
    We calculated the N from crop residue combustion based Ri, Yi and the percentage of residues burned for fuel (Bi) and the conversion coefficient for estimating the NOx, N2O,N2 and NH3 formed from burning crop residues (Ci). The conversion coefficient is equal to 3.83g/kg [17,39,41].

Proportional contribution of cropland area and fertilizer N application rate to fertilizer N input

The development of a simple attribution approach is similar to that taken in studies that have used the Kaya identity to obtain the regional drivers of accelerating carbon dioxide emissions [42], and forest identity [43]. Therefore, we used fertilizer N identity to determine the relative contribution of changes in cropland area and fertilizer N application rate in different parts of the Yangtze River Basin. We assessed the relative contribution of changes in cropland area and fertilizer N application rate to the change in total fertilizer N input using Eqs (1) and (2).

(1)

Because ln(M) = ln(A) + ln(D)

The rates of change (m, a, and d) of M, A, and D are:

Then, (2) where, M, A and D represent the total fertilizer N input (Tg N), cropland area (ha) and fertilizer N application rate (kg N/ha), respectively; and m, a, and d are the corresponding derivatives (or rates of change) of these attributes over time. This identity combines the values of cropland area with the fertilizer N application rate to give the change in fertilizer N input.

Regression analysis was conducted between the socio-economic factors (rural population and urban population density, grain yields (rice, wheat, peanuts and soybean), the fertilizer N application rate in 2000 and the changes of fertilizer N application rate and cropland area in the upper, middle and lower reaches of the Yangtze River Basin. This was used to reveal the factors driving fertilizer N input changes through changing cropland area and fertilizer N application rate. All analyses were conducted using the software Stata (version 14.0; Stata MP, College Station, TX, USA).

Data sources

We created a county-level database using GIS technology. First, we used county-level agricultural statistical data for 2000 and 2010 (from the Chinese Academy of Agricultural Sciences) to calculate the N budgets of the Yangtze River Basin, including fertilizer type and amount, crop type (e.g. wheat, corn and rice) and yields, cropland area, paddy land area, upland area, and livestock type (e.g. pigs, sheep and cattle) and amount. The crop area was obtained from the Chinese planting industry information network. Then, we used GIS software to analyze the spatial data (provided by the Institute of Remote Sensing and Digital Earth Chinese Academy, Beijing, China), including the county boundary data for countries and the Yangtze River Basin.

Uncertainty analyses of N budgets

Because of the uncertainty arising from the use of literature values and extrapolating these to the Yangtze River Basin area, we performed a sensitivity analysis by Monte Carlo simulations in Crystal Ball 7.0 software (Decisioneering, Inc., Denver, CO, USA) run 20000 times in 2000 to illustrate the importance of the assumptions for the results of this budgeting exercise.

The Chinese statistical data were collected using a new system during 2000 and 2010, and we believe that the county-level agricultural statistical data were an authentic source for our analyses[38]. We used different bulk depositions of N in the upper, middle and lower reaches of the Yangtze River Basin in different years, and we believe these data were also credible [28]. Therefore, the uncertainties were mainly from the emission rates and factors. The Monte Carlo simulations can estimate the contribution of the factors to the budget by characterizing their statistical distribution functions. For N content of compound fertilizer, a coefficient of variation (CV) of 5% was assumed, for N fixation, all CVs were assumed to be 35%, and for human and livestock emission factor and rate, CVs of 30 and 20% were used, respectively [31].

Results

N inputs, outputs and changes

The total N inputs of the Yangtze cropland increased by 1.2 Tg during 2000-2010. The largest contributor to the increased N input was fertilizer N, which also accounted for 87% of the total N input changes over the ten years. The N in the harvested crops increased by 0.6 Tg. The total N budget of the Yangtze cropland increased by 0.6 Tg. The N loss through NH3 volatilization, denitrification and N transported to water bodies increased slightly decreased slightly (Table 1).

thumbnail
Table 1. Estimated agricultural N inputs and outputs (Tg) in the Yangtze River Basin for 2000 and 2010.

https://doi.org/10.1371/journal.pone.0180613.t001

Spatial changes in NSI and NUE

There were spatial variations in NSI changes. The maximum increase (35kg/ha) in the NSI occurred in the middle reaches followed by the upper reaches of the basin (30 kg/ha). However, NSI of the lower reaches decreased by 60 kg/ha (Figs 2 and 3).

thumbnail
Fig 2. Changes in NSI and NUE in the upper, middle and lower reaches of the Yangtze River Basin.

https://doi.org/10.1371/journal.pone.0180613.g002

thumbnail
Fig 3. Spatial changes of NSI (a) and NUE (b) during 2000–2010 in the Yangtze River Basin.

https://doi.org/10.1371/journal.pone.0180613.g003

The NUE of the Yangtze River Basin cropland changed slightly, and with spatial variations in the changes. The NUE decreased slightly for the upper reaches, did not change for the middle reaches and increased by 0.1 in the lower reaches (Figs 2 and 3).

Relative contributions of cropland area and fertilizer N application rate to fertilizer N input

The increase in the N fertilizer input was mainly affected by the area of cropland and the fertilizer N application rate. The area of cropland area increased by 0.2%, fertilizer N application rate increased by 10%, and the fertilizer N input increased by 1.1 Tg from 2000 to 2010. The fertilizer N input in the upper and middle reaches of the Yangtze River Basin increased, but it decreased in the lower reaches (Fig 4).

thumbnail
Fig 4. Relative contributions of cropland area and synthetic fertilizer N application rate to fertilizer N input during 2000–2010.

https://doi.org/10.1371/journal.pone.0180613.g004

The extent of the contributions of the area of cropland and the synthetic fertilizer N application rate to the change in fertilizer N input differed. The area of cultivated land and the contributions of the synthetic fertilizer N application rate in the whole of the Yangtze River Basin were 15% and 85%, respectively. Similarly, in the upper and middle reaches, the increase in the synthetic fertilizer N application rate was the main factor responsible for the increased N fertilizer input (representing 84% and 76% of the total N input, respectively). However, the decreased N fertilizer input in the lower reaches was due to both the area of cropland and the decrease of synthetic fertilizer N application rate, with proportional contributions of 48 and 52%, respectively (Fig 4).

In the upper and middle reaches, increased synthetic fertilizer N application rate mainly resulted from croplands with low fertilizer N application rate in 2000. For the increase of fertilizer N application rate in upper and middle reaches, coefficients for fertilizer N application rate in 2000 were negative and significant (P < 0.01 and P < 0.05, respectively) (Table 2). In the lower reaches, urbanization contributed significantly to the decrease in fertilizer N application rate and cropland area. For fertilizer N application rate and cropland area, coefficients for proportion of construction land area in 2000 were negative and significant (P < 0.05 and P < 0.001, respectively), however, coefficients for change rate of construction land area were positive (P < 0.01 and P < 0.001, respectively) (Table 2).

thumbnail
Table 2. Factors associated with the change of synthetic fertilizer N application rate and cropland area in upper, middle and lower reaches of the Yangtze River Basin.

https://doi.org/10.1371/journal.pone.0180613.t002

Uncertainty analyses of N budgets

After running 20000 Monte Carlo simulations, the average N budget of 2000 was 8.1 Tg with a 90% uncertainty range of 6.3–11.4Tg. The largest variance stemmed from livestock and human excrement N emission rates, which accounted for 52% of the variability of the N budget. Cattle and pig N emission factor accounted for 14 and 12%, respectively (Fig 5).

thumbnail
Fig 5. Contributions of various parameters to variability of N budget of 2000.

https://doi.org/10.1371/journal.pone.0180613.g005

Discussion

There were clear spatial and temporal variations in the Yangtze River Basin. The changes of NSI and fertilizer N input during 2000–2010 differed but their spatial changes were similar. The NSI and fertilizer N input changes during 2000–2010 in the upper and middle reaches of the basin increased, but it decreased in the lower reaches. Fertilizer N is usually the largest source of total N input [20,4446] and account for 87% of total N input in this study. The increase of NSI was usually related to the increase in fertilizer N input. The results were similar to those of previous studies with about 71% of the N input coming from fertilizer N in croplands and causing many environmental problems by entering waterbodies or the atmosphere, destroying the balance of ecosystems and harming human health [2,45].

We further identified the relative contributions of changes in cropland area and fertilizer N application rate to fertilizer N input changes in the different parts of the Yangtze River Basin. Fertilizer N application rate was the main factor resulting in increased fertilizer N input in the upper and middle reaches, fertilizer input decreased in the lower reaches due to the decrease of fertilizer N application rate and cropland area (Fig 4). These results were similar to a study of N inputs for the Yangtze River Basin and suggested that the intensity of agricultural activities increased gradually from the lower to the middle and upper reaches mainly through increases in synthetic fertilizer N application rate, and fertilizer N was the major input for the middle and lower reaches [26].

Based on the relative contributions of cropland area and synthetic fertilizer N application rate changes to fertilizer input (Fig 4), the main factors influencing the changes in fertilizer N application rate and cropland area were further analyzed. The increase of fertilizer application rate in croplands with low fertilizer N application rate in 2000 contributed to the increase of fertilizer N application rate in upper and middle reaches (Table 2). However, in lower reaches, the decrease of fertilizer N application rate and cropland area due to urbanization lead to the decrease of fertilizer N input (Table 2). These results were similar to the study in the lower reaches of Yangtze River Basin, For example, rapid urbanization and industrialization of Jiangsu Province (in the lower reaches) occupied rice paddy fields and dryland [21]

Quantitative analysis of the contributions to fertilizer N input can also provide important information for spatial management of risk decrease in the agricultural environment. In the upper and middle reaches, the focus should be on monitoring and optimization of fertilizer N application rate [47,48], especially in croplands that previously had low fertilizer N application rate during agricultural development. Moreover, recent study suggested that joint study of N and phosphorus is important to understand surplus and balance and improve their use efficiency [49]. It is also necessary to clearly understand the relationship between N inputs and possible N yield [48]. In the lower reaches, the focus should be on the impacts of urbanization on agricultural production. These conclusions are in accord with a study of net anthropogenic N inputs in the upper Yangtze River Basin, in which fertilizer N input decreased with increased urbanization in the city of Chengdu [50]. The strictest farmland protection system should be implemented to prevent the decline of arable land and basic farmland area, it plays an important role in farmland protection. In addition, spatial differences in crop composition and crop management strategies may also be responsible for spatial differences in fertilizer N input, NSI and NUE [20,47,51], and these should be analyzed further.

There were some uncertainties in calculation of the N budget, mainly caused by obtaining some parameters from the literature, such as N deposition and N from animal and human excrement. During this study, we tried to use approximate parameters (e.g., N contents in products and N deposition) from local data in different areas of the Yangtze River Basin, to make our results more valid and reliable. In addition, we did not consider grassland and wetland, which may influence the N budget of croplands. The results of uncertainty analyses of N budgets showed that the largest variance stemmed from livestock and human excrement N emission rates (Fig 5). Therefore, to reduce the influence of uncertainty in future study, the anthropogenic parameters for N budgets in rural regions should be identified [52].

We revealed the spatial heterogeneity of synthetic fertilizer N input changes during 2000–2010 and quantitatively analyzed the contribution of different factors to these changes in the Yangtze River Basin. In the upper and middle reaches, the increased synthetic fertilizer N application rate, mainly due to increased N input in cropland that previously had low fertilizer N application rate, accounted for 84% and 76% of the N input increase, respectively. However, in lower reaches, mainly due to urbanization, the decreased cropland area and fertilizer N application rate nearly equally contributed to the fertilizer N input decrease. The quantitative analysis of different factors contributing to fertilizer N input can provide spatial information needed to optimize synthetic fertilizer N application rate and monitor the impacts of urbanization on agricultural production [53], helping to decrease agricultural environment risk and maintain sustainable agricultural production in different areas.

Acknowledgments

We thank the anonymous reviewers and academic editor for the invaluable suggestions.

References

  1. 1. Robertson GP, Vitousek PM (2009) Nitrogen in agriculture: balancing the cost of an essential resource. Annual Review of Environment and Resources 34: 97–125.
  2. 2. Galloway JN, Townsend AR, Erisman JW, Bekunda M, Cai Z, Freney JR, et al. (2008) Transformation of the nitrogen cycle: recent trends, questions, and potential solutions. Science 320: 889–892. pmid:18487183
  3. 3. Bo S, Run-Ping S, Bouwman A (2008) Surface N balances in agricultural crop production systems in China for the period 1980–2015. 18: 304–315.
  4. 4. Giles J (2005) Nitrogen study fertilizes fears of pollution. Nature Publishing Group. pp. 791–791.
  5. 5. Rabalais NN, Turner RE, Wiseman WJ Jr (2002) Gulf of Mexico hypoxia, aka “The dead zone”. Annual Review of ecology and Systematics 33: 235–263.
  6. 6. Diaz RJ, Rosenberg R (2008) Spreading dead zones and consequences for marine ecosystems. science 321: 926–929. pmid:18703733
  7. 7. Powlson DS, Addiscott TM, Benjamin N, Cassman KG, de Kok TM, van Grinsven H, et al. (2008) When does nitrate become a risk for humans? Journal of environmental quality 37: 291–295. pmid:18268290
  8. 8. Haywood J, Lean J, Lowe D, Myhre G, Nganga J, Prinn R, et al. (2007) Changes in atmospheric constituents and in radiative forcing. Climate change: 129–234.
  9. 9. OECD. (2008) Environmental Performance of Agriculture in OECD Countries since 1990. Eurochoices 12(3): 30–31.
  10. 10. Cassman KG, Dobermann A, Walters DT (2002) Agroecosystems, nitrogen-use efficiency, and nitrogen management. AMBIO: A Journal of the Human Environment 31: 132–140.
  11. 11. Shindo J (2012) Changes in the nitrogen balance in agricultural land in Japan and 12 other Asian Countries based on a nitrogen-flow model. Nutrient Cycling in Agroecosystems 94: 47–61.
  12. 12. Bouwman A, Van Drecht G, Knoop J, Beusen A, Meinardi C (2005) Exploring changes in river nitrogen export to the world's oceans. Global Biogeochemical Cycles 19: 4–14.
  13. 13. Van Breemen Nv, Boyer E, Goodale C, Jaworski N, Paustian K, Seitzinger S, et al. (2002) Where did all the nitrogen go? Fate of nitrogen inputs to large watersheds in the northeastern USA. Biogeochemistry 57: 267–293.
  14. 14. Zhang h, Li h, Li X, Li Z (2014) Temporal Changes of Nitrogen Balance and Their Driving Factors in Typical Agricultural Area of Lake Tai Basin. Chinese Journal of Soil Science: 1119–1129.
  15. 15. Zheng M, Zheng H, Wu Y, Xiao Y, Du Y, Xu W, et al. (2015) Changes in nitrogen budget and potential risk to the environment over 20years (1990–2010) in the agroecosystems of the Haihe Basin, China. Journal of Environmental Sciences 28: 195–202.
  16. 16. YanW J, Zhang S, Wang J (2001) Nitrogen biogeochemical cycling in the Changjiang Drainage basin and its effect on Changjiang River dissolved inorganic nitrogen: temporal trend for the period 1968~ 1997. Acta Geographica Sinica 56: 505–514.
  17. 17. Bao X, Watanabe M, Wang Q, Hayashi S, Liu J (2006) Nitrogen budgets of agricultural fields of the Changjiang River basin from 1980 to 1990. Science of the total environment 363: 136–148. pmid:16182342
  18. 18. Liu C, Watanabe M, Wang Q (2008) Changes in nitrogen budgets and nitrogen use efficiency in the agroecosystems of the Changjiang River basin between 1980 and 2000. Nutrient Cycling in Agroecosystems 80: 19–37.
  19. 19. Ma L, Wang F, Zhang W, Ma W, Velthof G, Qin W, et al. (2013) Environmental assessment of management options for nutrient flows in the food chain in China. Environmental science & technology 47: 7260–7268.
  20. 20. Han Y, Fan Y, Yang P, Wang X, Wang Y, Tian J, et al. (2014) Net anthropogenic nitrogen inputs (NANI) index application in Mainland China. Geoderma 213: 87–94.
  21. 21. Liu Y, Wang J, Long H (2010) Analysis of arable land loss and its impact on rural sustainability in Southern Jiangsu Province of China. Journal of Environmental Management 91: 646–653. pmid:19853366
  22. 22. Wu Y, Su Y, Zhang L (2006) Economic structure transformation and land use change of the Changjiang River Basin. Chinese Geographical Science 16: 289–293.
  23. 23. Sun ZM, Zhi jie WU, Chen LJ, Liu YG (2006) Research Advances in Nitrogen Fertilization and its Environmental Effects. Chinese Journal of Soil Science 37: 782–786.
  24. 24. Jin X (1995) Lakes in China: research of their environment: China Ocean Press. 500 p.
  25. 25. Howarth RW, Billen G, Swaney D, Townsend A, Jaworski N, Lajtha K, et al. (1996) Regional nitrogen budgets and riverine N & P fluxes for the drainages to the North Atlantic Ocean: Natural and human influences. Nitrogen cycling in the North Atlantic Ocean and its watersheds: Springer. pp. 75–139.
  26. 26. Wang Q, Koshikawa H, Liu C, Otsubo K (2014) 30-year changes in the nitrogen inputs to the Yangtze River Basin. Environmental Research Letters 9: 115005.
  27. 27. Chapin FS III, Matson PA, Vitousek P (2011) Principles of terrestrial ecosystem ecology: Springer Science & Business Media. 529 p.
  28. 28. Liu X, Zhang Y, Han W, Tang A, Shen J, Cui Z, et al. (2013) Enhanced nitrogen deposition over China. Nature 494: 459–462. pmid:23426264
  29. 29. Editorial Committee of Chinese Yearbook of Agriculture (2009) China Agriculture Yearbook: Beijing: China Agriculture Press.
  30. 30. Li S, Jin J (2011) Characteristics of nutrient input/output and nutrient balance in different regions of China. Scientia Agricultura Sinica 44: 4207–4229.
  31. 31. Ti C, Pan J, Xia Y, Yan X (2012) A nitrogen budget of mainland China with spatial and temporal variation. Biogeochemistry 108: 381–394.
  32. 32. Chen N, Hong H, Zhang L, Cao W (2008) Nitrogen sources and exports in an agricultural watershed in Southeast China. Biogeochemistry 87: 169–179.
  33. 33. Yan W, Zhang S, Sun P, Seitzinger SP (2003) How do nitrogen inputs to the Changjiang basin impact the Changjiang River nitrate: a temporal analysis for 1968–1997. Global Biogeochemical Cycles 17: 1091–1091.
  34. 34. Smil V (1999) Nitrogen in crop production: An account of global flows. Global biogeochemical cycles 13: 647–662.
  35. 35. Yan W, Yin C, Zhang S (1999) Nutrient budgets and biogeochemistry in an experimental agricultural watershed in Southeastern China. Biogeochemistry 45: 1–19.
  36. 36. Burkart MR, James DE (1999) Agricultural-nitrogen contributions to hypoxia in the Gulf of Mexico. Journal of Environmental Quality 28: 850–859.
  37. 37. Liu X, Xu J, Wang F, Zhang F, Ma W (2005) The resource and distribution of nitrogen nutrient in animal excretion in China. J Agric Univ Hebei 28: 27–32.
  38. 38. Gu B, Ju X, Chang J, Ge Y, Vitousek PM (2015) Integrated reactive nitrogen budgets and future trends in China. Proceedings of the National Academy of Sciences 112: 8792–8797.
  39. 39. GuangWen M, Bao X, Shan Y, JianYing G, HuiLing Y (2009) Budget changes and district diversity of non-point source nitrogen in agricultural fields in the Yangtze River basin. Research of Environmental Sciences 22: 132–137.
  40. 40. Change IPOC (2006) 2006 IPCC guidelines for national greenhouse gas inventories. 2013-04-28]. http://www.ipcc-nggip.iges.or.jp/public/2006gl/index.html.
  41. 41. Andreae MO, Merlet P (2001) Emission of trace gases and aerosols from biomass burning. Global Biogeochemical Cycles 15: 955–966.
  42. 42. Raupach MR, Marland G, Ciais P, Le Quéré C, Canadell JG, Klepper G, et al. (2007) Global and regional drivers of accelerating CO2 emissions. Proceedings of the National Academy of Sciences 104: 10288–10293.
  43. 43. Fang J, Guo Z, Hu H, Kato T, Muraoka H, Son Y (2014) Forest biomass carbon sinks in East Asia, with special reference to the relative contributions of forest expansion and forest growth. Global change biology 20: 2019–2030. pmid:24464906
  44. 44. Varshney RK, Ribaut J-M, Buckler ES, Tuberosa R, Rafalski JA, Langridge P (2012) Can genomics boost productivity of orphan crops? Nature biotechnology 30: 1172–1176. pmid:23222781
  45. 45. Xiong Z, Freney J, Mosier A, Zhu Z, Lee Y, Yagi K (2008) Impacts of population growth, changing food preferences and agricultural practices on the nitrogen cycle in East Asia. Nutrient Cycling in Agroecosystems 80: 189–198.
  46. 46. Chen X-P, Cui Z-L, Vitousek PM, Cassman KG, Matson PA, Bai J-S, et al. (2011) Integrated soil–crop system management for food security. Proceedings of the National Academy of Sciences 108: 6399–6404.
  47. 47. Xu X, He P, Zhang J, Pampolino MF, Johnston AM, Zhou W (2017) Spatial variation of attainable yield and fertilizer requirements for maize at the regional scale in China. Field Crops Research 203: 8–15.
  48. 48. Mueller ND, Lassaletta L, Runck BC, Billen G, Garnier J, Gerber JS (2017) Declining spatial efficiency of global cropland nitrogen allocation. Global Biogeochemical Cycles.
  49. 49. Bouwman A, Beusen A, Lassaletta L, Van Apeldoorn D, Van Grinsven H, Zhang J (2017) Lessons from temporal and spatial patterns in global use of N and P fertilizer on cropland. Scientific Reports 7: 40366. pmid:28084415
  50. 50. Wang A, Tang LH, Yang DW, Lei HM (2016) Spatio-temporal variation of net anthropogenic nitrogen inputs in the upper Yangtze River basin from 1990 to 2012. Science China Earth Sciences: 1–13.
  51. 51. Meng Q, Hou P, Wu L, Chen X, Cui Z, Zhang F (2013) Understanding production potentials and yield gaps in intensive maize production in China. Field crops research 143: 91–97.
  52. 52. Liu C, Wang QX, Lei A, Yang YH, Ouyang Z, Lin YM, et al. (2009) Identification of anthropogenic parameters for a regional nitrogen balance model via field investigation of six ecosystems in China. Biogeochemistry 94: 175–190.
  53. 53. Ewing PM, Runck BC (2015) Optimizing nitrogen rates in the midwestern United States for maximum ecosystem value. Ecology and Society 20: 18.