Carbon-nitrogen-water Fluxes and Auxiliary Parameters of China's Ecosystems Zone II Versions EN3 Vol 4 (1) 2019
Download
A spatial and temporal dataset on atmospheric inorganic nitrogen wet deposition in China (1996 – 2015)
 >>
: 2018 - 05 - 21
: 2018 - 06 - 20
: 2019 - 01 - 11
550 12 0
Abstract & Keywords
Abstract: Atmospheric nitrogen (N) deposition is a major process in the global N cycle, with important effects on the structures and functions of natural ecosystems. With the rapid development of industry and agriculture under urbanization, N deposition in China has undergone a sharp increase in recent decades. Therefore, a method of obtaining a spatial dataset on atmospheric N deposition is key to studying the ecological and environmental effects of N deposition on ecosystems. Based on site–year data on inorganic wet N deposition from published literature, we created a spatial dataset on wet N deposition during 1996 – 2000, 2001 – 2005, 2006 – 2010, and 2011 – 2015 in China using geostatistical methods. This dataset includes NH4+ -N, NO3- -N, and DIN (the sum of NH4+ -N and NO3- -N), with a spatial resolution of 1 km × 1 km, and is in the .tiff format. As the first openly accessible spatial dataset on wet N deposition in China, this can be utilized to analyze the spatial and temporal patterns of wet N deposition and the ecological effects in China, with a resolution of over 1 km × 1 km, as well as providing policy support for N management at the national level.
Keywords: China; nitrogen deposition; spatial pattern; interpolation technique; acid deposition
Dataset Profile
English titleA spatial and temporal dataset on atmospheric inorganic nitrogen wet deposition in China (1996 – 2015)
Corresponding authorYu Guirui (yugr@igsnrr.ac.cn)
Data authorsJia Yanlong, Wang Qiufeng, Zhu Jianxing, Chen Zhi, He Nianpeng, Yu Guirui
Time range1996 – 2015
Geographical scopeChinese mainland
Spatial resolution1000 mData volume915.6 MB
Data format*.tiff
Data service system<http://www.cnern.org.cn/data/meta?id=40575>;
<http://www.sciencedb.cn/dataSet/handle/607>
Sources of fundingNational Natural Science Foundation of China (31700377), Strategic Priority Research Program of the Chinese Academy of Sciences (XDA19020302), National Key Research and Development Program of China (2016YFA0600104), Science and Technology Service Network Initiative of the Chinese Academy of Sciences (KFJ-SW-STS-169)
Dataset composition1. The dataset includes 12 files indicating the spatial patterns of NH4+ -N, NO3- -N, and DIN deposition. Each data document is recorded as XXX_YYYY_YYYY.tiff, where XXX and YYYY_YYYY denote the type of nitrogen deposition and the study period, respectively. For example, NH4+ _1996_2000.tiff records the spatial pattern of NH4+ -N deposition in China during 1996–2000.
1.   Introduction
Atmospheric nitrogen (N) deposition is the process in which atmospheric gaseous and particulate nitrogen-containing substances are deposited on Earth’s surface through precipitation (cloud droplets, fog droplets, rain, and snow) or under the action of gravity and turbulence. Owing to the rapid development of industry and agriculture under urbanization, anthropogenic active N emissions in China have been increasing rapidly, leading to a continuous increase in atmospheric N deposition in the ecosystem.1 Atmospheric N deposition plays an important role for natural ecosystems, artificial ecosystems, and human beings. For example, such N can promote productivity in forests and grasslands,2-3 and affect the greenhouse gas balance.4 Meanwhile, excessive N deposition can also lead to soil and water acidification,5 reduce the soil buffer capacity,6 and reduce biodiversity.7 Therefore, obtaining a set of scientific, systematic, and dynamic spatial pattern data on atmospheric N deposition is key to studying its ecological and environmental effects.
At the global scale, shared data on atmospheric N deposition in a spatial pattern can mainly be obtained from the simulation results of atmospheric chemical transport models. For example, Dentener8 shared global atmospheric N deposition maps obtained by the TM3 model in 1860, 1993, and 2050 (spatial resolution 5° × 3.75°, http://daac.ornl.gov/). In China, previous studies have researched the spatial pattern of atmospheric N deposition based on modeling and spatial interpolation methods,9-10 and our previous work also explored the temporal and spatial pattern of atmospheric N deposition in China.11-12 However, so far there has not been a set of shared data for the spatial pattern of N deposition with a high spatial resolution and long-term dynamic series. To some extent, this limits the research work of Chinese scientists in related fields.
In this study, based on site–year data retrieved from published literature we systematically created a spatial pattern dataset on atmospheric inorganic N wet deposition in China from 1996 to 2015 (in five-year intervals and four subsets) using geostatistical methods. The indicators include ammonium N (NH4+ -N), nitrate N (NO3- -N), and soluble inorganic N (DIN, which is the sum of NH4+ -N and NO3- -N). The open sharing of this dataset will provide a scientific basis for studying the ecological and environmental effects of N deposition in China, and also provide policy support for N management at the national level.
2.   Data collection and processing
The production process of the spatiotemporal pattern dataset on atmospheric inorganic N wet deposition in China includes two parts: the collection of site data and generation of spatial data (Figure 1).
2.1   Collection and processing of site data
In this study, the site data employed to generate the spatial pattern of atmospheric inorganic N wet deposition were obtained from studies between 1996 and 2015. The generation of site datasets includes two parts: data collection and data processing.
The main site data collection process included five parts: keyword determination, standard formulation, literature collection, data extraction, and data verification (Figure 1). To comprehensively collect studies related to N deposition, the keywords for selecting studies were "nitrogen deposition," "wet deposition," "chemical composition of precipitation," "rainwater chemistry," and "acid rain." The criteria for collecting effective data were as follows: 1) the concentration or flux of NH4+ -N, NO3- -N, and DIN in precipitation were observed; 2) the observation frequency was daily, weekly, or monthly; and 3) the observation duration was greater than one year. After preliminary screening of the collected studies, data extraction was performed for the valid data. Once the data was extracted, we checked it to ensure the authenticity and reliability.


Fig.1   Flowchart of data processing and dataset establishment
The main aspects of site data processing included sampling method comparison, coordinate conversion, unit conversion, flux conversion, and outlier elimination. Owing to the complexity and inconsistency of the data from different studies, careful processing of the extracted data is key to the formation of standardized datasets. The specific data processing here was as follows: 1) Sampling method comparison. The main sampling methods for wet deposition are the barrel method and intelligent precipitation collector. The wet-only deposition is obtained by an intelligent precipitation collector. However, the barrel method usually obtains the bulk deposition, because part of the dry deposition will enter the barrel following precipitation. To unify all the data into wet-only deposition, we analyzed these two types of data when observed at the same site simultaneously. We found that there was a highly significant correlation between the wet-only deposition and bulk deposition (Figure 2), where the proportion of wet-only deposition to the bulk deposition was approximately 70%. We employed this coefficient to transform all bulk deposition data into wet-only deposition data. 2) Coordinate transformation. The spatial location of an observation site is very important, but the coordinates in different studies are given in different units or coordinate systems. Therefore, they must be unified into the same units in the same coordinate system. 3) Unit conversion. Owing to the different units of N deposition flux in the published data, we unified the different units into kg N ha-1 yr-1. 4) Flux conversion. For the data on the N concentration in precipitation, the annual deposition flux was obtained by multiplying the concentration data by the corresponding annual precipitation. 5) Exclusion of outliers. Following the above data conversion process, outliers were eliminated by the triple standard deviation method.
Finally, after performing the above data flow and checking again, we obtained 1807 site–year data items on wet N deposition in China, which covered 33 provinces in China, excluding Taiwan and the South China Sea islands. This consisted of 368, 576, 554, and 309 data points in 1996–2000, 2001–2005, 2006–2010, and 2011–2015, respectively.


Fig.2   Correlations between bulk N deposition and wet N deposition
Notes: A: NH4+-N wet deposition; B: NO3--N wet deposition; and C: DIN wet deposition (Unit: kg N ha-1 yr-1)
2.2   Generation of spatial pattern data
To obtain the spatial pattern of atmospheric N wet deposition in China, this study extrapolated the observed site data to a continuous area using geostatistical methods. Here, we employed the Kriging method to obtain the spatial distribution maps of the wet deposition of NH4+ - N, NO3- -N, and DIN in China during the four periods from 1996 to 2015.
The Kriging method is designed for the unbiased optimal estimation of regionalized variables in a limited area, based on the theory of variograms and structural analysis.13 The variables for applying the Kriging method must satisfy three basic conditions: spatial continuity, the existence of spatial autocorrelation, and the normal distribution.14 Atmospheric N deposition is a spatially continuous variable, for which spatial autocorrelation can be tested by the nugget coefficient. When the nugget coefficient is below 25%, the spatial autocorrelation of variables is strong. In this study, the nugget coefficients of variables (NH4+ -N and NO3- -N deposition) were less than 25%. Therefore, the Kriging method can be applied in this study if the data satisfy the normal distribution. The Kriging method includes multiple interpolated methods, such as ordinary Kriging, simple Kriging, co-Kriging, and universal Kriging, which each have conditions for their application.15 The atmospheric N deposition satisfied the conditions for the ordinary Kriging method, which was employed in this study as the interpolation method. The specific processes for spatial interpolation included an exploratory data analysis, data normal distribution test and transformation, determination of the optimal variogram, and so on. (Figure 1B). A detailed description is provided by Jia et al.11
To test the accuracy of the interpolation results, we validated the results using independent data. Finally, based on the collected site data on wet deposition in different periods, the spatial distribution maps of NH4+ -N and NO3- -N were obtained by the Kriging method for 1996–2000, 2001–2005, 2006–2010, and 2011–2015, and then the DIN distribution maps were obtained by summing the distribution maps of NH4+ -N and NO3- -N in each time period. The spatial resolutions of the spatial distribution maps are 1 km × 1 km, in the .tiff format.
3.   Sample description
3.1   Name forms
Each data record of the spatial patterns of atmospheric N deposition in China is recorded as XXX_YYYY_YYYY.tiff, where XXX and YYYY_YYYY denote the type of N deposition and study period, respectively. For example, NH4+ _1996_2000.tiff records the spatial pattern of NH4+ -N deposition in China during 1996–2000.
3.2   Data sample
The dataset includes the spatial distribution maps of the wet deposition of NH4+ -N, NO3- -N, and DIN in the four periods of 1996–2000, 2001–2005, 2006–2010, 2011–2015, with a total of 12 documents. Figure 3 illustrates the data on the spatial distribution maps of the wet deposition of NH4+ -N, NO3- -N, and DIN in 2011–2015. The units of wet deposition are kg N ha-1 yr-1. The colors from green to red represent the gradual increase in the wet deposition flux, while the white area contains no data. In particular, because this study obtained the spatial pattern of wet N deposition through an interpolation method, the national average N deposition flux in this study is lower than previous estimation results obtained by simple arithmetic averaging methods. Unlike in Europe or North America, economic development is unbalanced across regions in China, with higher N deposition and more monitoring stations in the developed areas of Southeast China and lower N deposition and fewer monitoring stations in the vast underdeveloped areas of Northwest China. Therefore, the N deposition estimation results using a simple arithmetic average method are approximately 35%–45% higher than those obtained by spatial interpolation or modeling methods in China. Detailed discussions are provided by He et al.16


Fig.3   The spatial patterns of wet N deposition in 2011–2015 in China (data excluding Taiwan and South China Sea Islands)
Notes: A:NH4+-N wet deposition; B:NO3--N wet deposition; and C:DIN wet deposition (Unit: kg N ha-1 yr-1); Drawing-censoring No. GS (2018) 4935.
4.   Quality control and assessment
The dataset is derived from observation site data of N deposition retrieved from studies and spatial pattern data obtained using geostatistical methods. From the collection of site data to the generation of spatial pattern data, a complete data quality control and evaluation system was adopted to ensure the accuracy and reliability of the results.
The N deposition site data is derived from published literature, and the key is how to use these data from different studies to form a set of comparative standardized data. In previous studies on N deposition, most of the chemical methods for determining NH4+ -N and NO3- -N in precipitation studies were standard methods (ion chromatography and a spectrophotometer, national standard GB 11894-89), which provides the basis of the current study to integrate these data. Moreover, the differences between wet-only and bulk deposition resulting from different methods for precipitation sampling have also been effectively unified in this study (Figure 2). In addition, regarding data retrieval from the literature, the determination of search keywords, determination of key parameters, production of extraction tables, and formulation of data post-processing methods have all been discussed by experts to ensure the accuracy, completeness, and standardization of data extraction. After data collection, the data were checked again and processed through data unit conversion, same point-value elimination, outlier screening, and so on. Finally, a standardized site dataset on wet N deposition was generated.
The spatial pattern of wet N deposition was generated by a spatial interpolation method based on standardized site datasets. To ensure the reliability of the interpolation results, the determination of interpolation methods, exploratory data analysis, determination of the best variance function, and method of operating the statistical module in ArcGIS have all been discussed by experts. In addition, the interpolation results were independently validated, and the validation data were extracted from the observation network of N deposition established by China Agricultural University (43 stations).17 The evaluation indexes included R2, the root-mean-square error (RMSE), the regression coefficient, and the P value (Figure 4). Among these, the R2 values for NH4+ -N, NO3- -N, and DIN were 0.61, 0.40, and 0.61, respectively, with respective RMSEs of 3.21, 3.22, and 4.93 and regression coefficients of 1.03, 0.67, and 0.89. Furthermore, the P values were all less than 0.001. The results show that the interpolation results can better express the spatial pattern of atmospheric N wet deposition in China.


Fig.4   Independent validations of interpolation results
Notes: . A:NH4+-N wet deposition; B:NO3--N wet deposition; and C:DIN wet deposition (unit: kg N ha-1 yr-1)
The observed site data of wet N deposition utilized in this study were taken from published papers. Although the results are highly reliable after strict screening, quality control, and result verification in the processes of data collection, data collation, and spatial interpolation, there remain some uncertainties that may affect the results. These uncertainties mainly arise from the following three aspects: 1) Sample preservation and determination methods. The samples of wet N deposition mainly originate from the precipitation collected by rain barrels. Therefore, the differences in the methods of sample preservation and determination by different studies may lead to some errors in the wet N deposition determination results. 2) The unevenness of the distribution of observation sites. This study systematically collected literature data on wet N deposition in China, and obtained a large amount of data. However, the uneven spatial distribution of N deposition observation sites, exhibiting a trend of dense distribution in the East and scarce distribution in the West, would affect the accuracy of the spatial interpolation results. 3) Ratio coefficient of wet-only to bulk deposition. Because there were few studies in this area, we utilized the proportion coefficient 70% at the national scale in this study. Moreover, the validated dataset for the spatial pattern results in this study (the data from the observation network of N deposition established by China Agricultural University) also consisted of bulk deposition. The conversion of proportion coefficient has also been applied to these data, which may also increase the uncertainty in the wet N deposition assessment and verification in China to a certain extent. Therefore, to reduce the uncertainty in the spatial assessment, verification, and studies of the ecological effects of N deposition, long-term, networking, and standard observations of N deposition monitoring in China should be strengthened, especially in the western region of China.
5.   Usage notes
The dataset can be downloaded through the Science Data Back website (http://www.sciencedb.cn/dataSet/handle/607). All the data were generated using ArcGIS in the .tiff format, which can be viewed and employed in the visualized geographic information system software supporting the above format. This shared dataset mainly provides the spatial pattern data of atmospheric inorganic N wet deposition. The results of estimating N deposition in China are further discussed in our previous papers.11-12 In addition, this data is suitable for the assessment of the spatial distribution of N deposition and its ecological and environmental effects, but there may be some errors in site assessments of N deposition and its impact. The site observation data of N deposition and acid deposition are discussed in the published papers of our research group.12, 18-21
Acknowledgments
The authors are thankful for support from the Chinese Ecosystem Research Network (CERN). The authors are grateful for help in collecting data from literature from Ma Anna, Jiao Cuicui, Zheng Han, Xu Li, Zhao Hang, Yu Haili, Wen Ding, Wang Chunyan, Tian Miao, Liu Yuan, Liu Congcong, Li Ying, Song Guangyan, and Zhu Guili.
1.
Liu XJ, Zhang Y, Han WX et al. Enhanced nitrogen deposition over China. Nature 494 (2013): 459 – 462.
2.
Fleischer K, Rebel KT, van der Molen MK et al. The contribution of nitrogen deposition to the photosynthetic capacity of forests. Global Biogeochemical Cycles 27 (2013): 187 – 199.
3.
Thomas RQ, Canham CD, Weathers KC et al. Increased tree carbon storage in response to nitrogen deposition in the US. Nature Geoscience 3 (2010): 13 – 17.
4.
Templer PH, Pinder RW & Goodale CL. Effects of nitrogen deposition on greenhouse-gas fluxes for forests and grasslands of North America. Frontiers in Ecology and the Environment 10 (2012): 547 – 553.
5.
Vitousek PM, Aber JD, Howarth RW et al. Human alteration of the global nitrogen cycles: sources and consequences. Ecological Applications 7 (1997): 737 – 750.
6.
Bowman WD, Cleveland CC, Halada Ĺ et al. Negative impact of nitrogen deposition on soil buffering capacity. Nature Geoscience 1 (2008): 767 – 770.
7.
Stevens CJ, Dise NB, Mountford JO et al. Impact of nitrogen deposition on the species richness of grasslands. Science 303 (2004): 1876 – 1879.
8.
Dentener FJ. Global maps of atmospheric nitrogen deposition, 1860, 1993, and 2050. Data set. Available on-line [http://daac.ornl.gov/] from Oak Ridge National Laboratory Distributed Active Archive Center, Oak Ridge, Tennessee, U. S. A. (2006). DOI: 10.3334/ORNLDAAC/830.
9.
Lü CQ & Tian HQ. Spatial and temporal patterns of nitrogen deposition in China: Synthesis of observational data. Journal of Geophysical Research 112 (2007). DOI: 10.1029/2006JD007990.
10.
Zhao YH, Zhang L, Chen YF et al. Atmospheric nitrogen deposition to China: A model analysis on nitrogen budget and critical load exceedance. Atmospheric Environment 153 (2017): 32 – 40.
11.
Jia YL, Yu GR, He NP et al. Spatial and decadal variations in inorganic nitrogen wet deposition in China induced by human activity. Scientific Reports 4 (2014). DOI: 10.1038/srep03763.
12.
Zhu JX, He NP, Wang QF et al. The composition, spatial patterns, and influencing factors of atmospheric nitrogen deposition in Chinese terrestrial ecosystems. Science of The Total Environment 511 (2015): 777 – 785.
13.
Holland EA, Braswell BH, Sulzman J et al. Nitrogen deposition onto the United States and Western Europe: synthesis of observations and models. Ecological Application 15 (2005): 38 – 57.
14.
Combardella CA, Moorman TB, Novak JM et a1. Field-scale variability of soil properties in central lowa soils. Soil Science Society of America Journal 58 (1994): 1501 – 1511.
15.
Liu XN, Huang F & Wang P. The Principle and Methods of Spatial Analysis in GIS. Beijing: Science Press, 2008.
16.
He NP, Zhu JX & Wang QF. Uncertainty and perspectives in studies of atmospheric nitrogen deposition in China: A response to Liu et al. (2015). Science of The Total Environment 520 (2015): 302 – 304.
17.
Xu W, Luo XS, Pan YP et al. Quantifying atmospheric nitrogen deposition through a nationwide monitoring network across China. Atmospheric Chemistry and Physics 15 (2015): 12345 – 12360.
18.
Zhu JX, Wang QF, He NP et al. Imbalanced atmospheric nitrogen and phosphorus depositions in China: Implications for nutrient limitation. Journal of Geophysical Research-Biogeoscience 121 (2016). DOI: 10.1002/2016JG003393.
19.
Zhu JX, Wang QF, Yu HL et al. Heavy metals deposition through rainfall in Chinese natural terrestrial ecosystems: Evidences from national-scale network monitoring. Chemosphere 164 (2016): 128 – 133.
20.
Yu HL, He NP, Wang QF et al. Development of atmospheric acid deposition in China from the 1990s to the 2010s. Environmental Pollution 231 (2017): 182 – 190.
21.
Yu HL, He NP, Wang QF et al. Wet acid deposition in Chinese natural and agricultural ecosystems: Evidence from national scale monitoring. Journal of Geophysical Research Atmospheres 121 (2016): 10995 – 11005.
Data citation
1. Jia Y, Wang Q, Zhu J et al. A spatial and temporal dataset on atmospheric inorganic nitrogen wet deposition in China (1996 – 2015). Science Data Bank. DOI: 10.11922/sciencedb.607 (2018).
Article and author information
How to cite this article
Jia Y, Wang Q, Zhu J et al. A spatial and temporal dataset on atmospheric inorganic nitrogen wet deposition in China (1996–2015). China Scientific Data 4(2019). DOI: 10.11922/csdata.2018.0031.zh
Jia Yanlong
His major contribution to this work includes experiment design, data processing and quality control.
PhD, Lecturer; his main research interest is in ecosystem ecology.
Wang Qiufeng
Her major contribution to this work is experiment conduct.
PhD, Associate Professor; her main research interest is in global change and C-N-H2O cycle.
Zhu Jianxing
His major contribution to this work includes data acquisition and data processing.
PhD; his research interests are in ecosystem ecology and global change ecology.
Chen Zhi
Her major contribution to this work includes data acquisition, data processing and quality control.
PhD, Assistant Professor; her main research interest is in global change and carbon cycle.
He Nianpeng
His major contribution to this work includes experiment design and experiment conduct.
PhD, Professor; his research interests are in ecosystem traits, functional ecology, and biogeography ecology.
Yu Guirui
His major contribution to this work is integral experiment design.
yugr@igsnrr.ac.cn
PhD, Professor; his research interests are in ecosystem ecology, global change and C-N-H2O cycle.
National Natural Science Foundation of China (31700377), Strategic Priority Research Program of the Chinese Academy of Sciences (XDA19020302), National Key Research and Development Program of China (2016YFA0600104), Science and Technology Service Network Initiative of the Chinese Academy of Sciences (KFJ-SW-STS-169)
Publication records
Published: Jan. 11, 2019 ( VersionsEN3
Released: June 20, 2018 ( VersionsZH2
Published: Jan. 11, 2019 ( VersionsZH3
References
中国科学数据
csdata