【作者】 遆超普；

【导师】 潘剑君； 颜晓元；

【作者基本信息】 南京农业大学 ， 土壤学， 2011， 博士

【摘要】 氮是生命必须的元素之一,随着人口的增加和经济的发展,人为活动增加了地球陆地生态系统的氮投入。人为活化氮的投入一方面提高了作物的产量,满足了人类对食物和能源等的需求,但是另一方面化肥过量施用、畜牧养殖废污排放、化石燃料大量使用等人类活动过程,极大地干扰了氮在大气、水、生物、土壤等圈层的相互转化及运动。地球生态系统中的氮素超负荷承载,导致了地球环境自然平衡的破坏,从而引起温室效应、臭氧层破坏、酸雨、地下水硝酸盐污染、湖泊与近海水体富营养化等一系列从区域到全球尺度的生态环境问题。因此,研究氮素的输入输出过程、控制机制、影响因素和变化趋势,对于提出针对性的氮素管理措施,从根本上控制氮素流失、加强资源保护、改善生态环境、促进社会经济可持续发展具有重要意义。近年来尽管我国在氮素投入支出方面的研究取得一定的成就,但整体上还不够完善。本研究利用实验监测数据、农户调查数据、统计年鉴数据、图形数据与公开发表的文献资料等,通过建立氮素投入(生物固氮、化学氮肥、粮食和饲料进口以及大气沉降)和氮素支出(氨挥发、流入水体、反硝化和储存、粮食和饲料出口以及生物质燃烧)模型,研究我国不同区域空间尺度氮素特征,以明确氮素的来源去向、过程机制、变化规律、影响因素及其环境效应,探索氮素流失的控制环节和主要途径。本研究取得如下主要成果和结论：1、在以稻作为主的句容农业小流域,连续两年的观测数据和农业调查数据显示,2007-2009年该流域总氮投入为1272ton,单位面积通量280 kg N ha-1 yr-1。化学氮肥是主要的氮投入来源,占总量的78.7%；流域的大气干湿沉降为39 kg N ha-1 yr-1,是第二大氮源；流域生物固氮占总氮投入的13.8%,通过粮食和饲料进口的氮体现在农作物种子的买进,仅占总氮投入量的0.6%。作为典型的农业流域,该地区没有大规模的养殖场和工厂等,流域内生产的粮食和食品大部分输出到流域之外。人畜排泄物基本还田作有机肥,通过农田有机肥和化学氮肥氨挥发的氮为188ton,占总氮投入的14.8%。调查显示该流域农作物秸秆露天燃烧量很高,通过秸秆露天和作为燃料燃烧排放到大气中的氮为140 ton,占总氮投入的11%。作为流域唯一出水口的水库,氮输出仅9.3 ton,不到总氮投入的1%,这表明句容农业流域投入的人为活化氮大部分通过氨挥发和生物质燃烧的方式排放到大气中。氮素以氨挥发、生物质燃烧排放到流域外是影响该流域氮素循环的主要因素。该农业流域内每年有637 ton氮被土壤系统本身所去除或被储存于土壤中,占总氮素投入的50%。历史资料和实测数据显示,当地土壤全氮含量并没有明显增加,因此大部分盈余氮通过反硝化进入环境。当地独特的自然景观对氮素的拦截作用、反硝化作用对水体氮的去除和大量的氨挥发、生物质燃烧是水体氮输出低的主要原因。因此,从减少氮素损失和治理环境污染的角度出发,对于稻作农业流域内的氮素管理,应该更加关注氮素的气态损失。2、在人口密集,工农业、养殖业高度发达的太湖流域常熟地区,单位面积氮素总投入为23927 kg N km-2 yr-1,折合27707 ton。化学氮肥是最大的氮源,占总氮投入的56.6%；大量的粮食和饲料的进口是该地区氮投入的主要特征,总量为6186ton,单位面积5342 kg N km-2 yr-1,占总氮投入的22.3%；大气氮沉降和生物固氮分别占总氮投入的15.5%和5.6%。通过氨挥发和生物质燃烧进入大气的氮分别为3535ton和172ton,占总量的12.8%和6.2%；通过水体输出的氮为8108ton(7002 kg N km-2 yr-1),占总投入量的29.3%。差减法显示,51.7%的氮素通过反硝化进入环境或储存在系统中。太湖地区水污染严重,大量未经处理的生活污水、工业废水直接排入水体是导致水体氮污染的主要原因。本研究显示,该地区地表水输出平均氮浓度为6mg N L-1,高于我国地表水劣五类标准。通过农村生活污水、禽畜养殖排放到水体的氮是水体氮的第一大源,占总量的26.5%；城镇居民生活污水和工业废水是第二的水体氮源,为2085ton,占流入水体总氮量的25.7%；来自农田径流和淋溶的氮占总流入水体总氮量的17.9%,大气氮沉降所占的比例仅次于农田径流和淋溶,为17.0%。作为典型的河网平原地区,水产养殖也是该地区水体氮的一个重要来源之一,占总量的8.1%。因此,在控制当地城镇生活污水、工业废水等主要水体氮污染源的同时,也要采取措施控制农村生活污水、禽畜排泄物排的排放。3、1985、1990、1995、2000、2005、2007年我国大陆生态系统氮投入总量分别为3081、3778、4418、4610、5238、5426kg N km-2 yr-1,22年间投入量增加76.1%。化学氮肥是第一大氮源,生物固氮所占比例呈现逐步下降的趋势,大气氮沉降约占总氮投入的24%。1985年以后出现粮食和饲料净进口的现状,截止2007年,通过粮食和饲料进口的氮占总氮投入的3.5%。氨挥发和流入水体的氮占总氮投入比例相当,约20%。22年间生物质燃烧排放氮占总氮投入的5.3-7.7%。通过物质平衡法估算,反硝化约占总氮支出的41%,系统储存占17%，大量的氮投入通过反硝化途径进入环境。与其他国家和地区相比,我国化学氮肥的施用是导致氮素投入水平过高的主要原因,而因此导致的大气氮沉降、氨挥发等也显著高于其他国家和地区。受地理位置、人口密度、经济发展水平、土地利用类型等因素的影响,我国氮素收支在省域尺度上差别很大。人口密度与反硝化和储存、流入水体的氮、总氮投入呈极显著直线正相关关系；耕地面积占总面积百分比、人均GDP与氮总投入、氨挥发、反硝化和储存、流入水体直线显著正相关。除林地面积百分比与生物质燃烧排放的氮直线显著正相关外,草地面积百分和林地面积百分比与氮素总投入、氨挥发、反硝化和储存、流入水体直线显著负相关。因此,在人口密度较大、经济相对发达的东部和东南沿海地区以及耕地面积比例较大的河南、山东、河北等省,氮素投入总量较高,相应的氮素流失量也较大。而在广大西北、西南地区,经济发展水平相对较低、人口密度小、土地利用类型以林地和草地为主,氮素投入较低、对环境的影响相应较小。我国长江、黄河、珠江三大河流水体存在不同程度的氮污染情况,本研究对三大流域的水体氮输出估算显示,长江流域是三大流域中水体氮输出最高的流域,1985、1990、1995、2000、2005、2007年水体氮输入分别为1.90、2.43、2.59、2.76、2.61Tg,相应年份河口氮输出量为1.33、1.7、1.81、1.93、1.83Tg；黄河、珠江流域流入水体、河口输出的氮相对较低,不足1Tg。1985-2007年间,三大流域的氮输出均呈增加的趋势,以长江流域增长较低,为37.4%,黄河流域增长最大,为61.5%。三大流域河口氮输出估算值与实测值相比,除黄河流域因常年断流、农田灌溉用水等原因导致估算值偏高外,其他两个流域估算值与实测值一致。4、不同尺度下的氮素收支研究显示,人为活动严重影响了氮素的生物地球化学循环。化学氮肥的投入在不同尺度下均为最大的氮素来源。不同尺度下,大部分氮素都通过反硝化或者系统储存进入到环境中,反硝化与系统储存的氮约占总氮投入的50%左右。在以农业种植为主的农业小流域,氮素投入支出与农业活动紧密相关,投入到环境中的氮素主要通过氨挥发、生物质燃烧等气体形式排放到大气中,地表水径流输出很低。而在经济发达的中尺度河网平原地区,除化学氮肥外,粮食的进口占氮素投入的比重较高。大量未经处理的生活污水、工业废水、人畜排泄物等流入水体导致水体氮浓度过高,水体富营养化严重。国家尺度下,氮素的投入支出时空变化差异较大,氮素的循环过程主要受人均GDP、土地利用类型和人口密度的影响。投入到地表的氮素超过70%通过反硝化、生物质燃烧、氨挥发排放到大气中,产生酸雨沉降、温室气体等环境问题,而约有20%的氮素流入到水体,造成我国主要河流的水质污染。更多还原

【Abstract】 Nitrogen (N) is a fundamental component of living organisms, and it has been strongly influenced by human activity. From the pre-industrial era to 1990, reactive N input to the global terrestrial system increased twofold. The massive N input has enabled humankind to greatly increase food production. However, excessive N can induce a series of economic and environmental problems such as the greenhouse effect, destruction of the ozone layer, acid rain, nitrate pollution in groundwater, eutrophication of lakes and offshore water, and biodiversity reduction locally, nationally and globally. The findings of manifold consequences of human alteration of the N cycle have led to a much improved understanding of the scope of the anthropogenic N problem and possible strategies for managing it.China is the third largest country in the world and has diverse climatic conditions ranging from tropical in the south to cold temperate in the north, and from humid in the east to arid in the northwest. Rapid economic development and expansion of the human population in the past three decades has resulted in a large increase in chemical fertilizer and fossil fuel consumption, and thus greatly altered the N cycle. However, the changes in the input and fate of reactive N are not well understood. This study compile a N budget for China’s mainland with spatial and temporal distribution by using measurements, household interview and more up-to-date activity data and flux parameters to analyze the fluxes of N inputs and output and their impacts on the environment. In the N budget model, N input include chemical fertilizer, N fixation, atmospheric deposition and net food/feed import, N output include ammonia volatilization, biomass burning, net food and feed export and N export to surface water. The difference between N input and output was assumed as denitrification and storage. The followings are the productions and conclusions of this dissertation:1.We conducted N budget calculations for a rice paddy dominated agricultural watershed(Jurong Reservoir Watershed, JRW) in eastern China for 2007-2009, based on intensive monitoring of stream N dynamics, atmospheric deposition, ammonia (NH3) volatilization and household interviews about N related agricultural activities. The results showed that total N input in JRW was 1272ton yr-1. Chemical fertilizer was by far the dominant N input, totaling 1001 ton yr-1, accounts for 78.7% of total N input, atmospheric N deposition was the second most N input in this watershed,39 kg N ha-1 yr-1. N fixation and seeds import accounts for 13.8%,0.6%, respectively.Although total N input to the watershed was up to 280 kg N ha-1 yr-1, riverine discharge was only 4.2 kg N ha-1 yr-1, accounting for 1.5% of the total N input, and was further reduced to 2.0 kg N ha-1 yr-1 after reservoir storage and/or denitrification removal. The watershed actually purified the N in rainwater, as N concentrations in river discharge were much lower than those in rain water. The low riverine N output was because of the characteristics of paddy-dominated watersheds greatly influence the hydrologic flow path, increase the water residence time and the transport of N and associated elements. In addition, the low proportion of riverine N export in this watershed is also due to the fact that N is subject to other losses and export. Major N outputs included food/feed export, NH3 volatilization from chemical fertilizer and manure, and emissions from crop residue burning. Net reactive gaseous emissions (emissions minus deposition) accounted for 5.5% of the total N input, much higher than riverine discharge. Most of the N inputs were on croplands through N fertilization and fixation, and this part of N inputs is susceptible to NH3 volatilization and biomass burning, as well as food export. On average,10% of the chemical N fertilizer was volatilized as NH3 in the watershed. Therefore, the agricultural N cycle in such paddy-dominated watersheds impacts the environment mainly through gas exchange rather than water discharge.2. The Taihu Lake region in China is highly developed, but surface water pollution has become a serious environmental problem in recent years, with nitrogen (N) a major pollutant. A N-budget for Changshu, a representative county-level city in the Taihu Lake region, was established by using N-related human activities data from an intensive household survey conducted in 2007, measurement data on N fluxes and literature data on other necessary parameters. The total N input was 23927 kg N km-2 yr-1. Chemical fertilizer input was heavy and averaged 13553 km-2 yr-1, being the largest source of N input. Atmospheric N deposition contributed 15.5% to the total N input and food/feed import contributed another 22.3%. Average N input through biological N fixation was 1332 kg N km-2 yr-1, making it a minor contributor.Nearly half of the N input was denitrified or stored in soil, amounting to 12381 kg N km-2 yr-1. There was no N output through net food/feed export from the region. N transport to water was 7002 kg N km-2 yr-1, accounting for 29.3% of the total N input. NH3 volatilization from fertilizer and human and animal waste amounted to 3053 kg N km-2 yr-1, or 12.8% of the total N input. About 6.2% of the total N input was emitted to the atmosphere through burning of crop residue.N transport from human and animal waste, fertilizer and waste water, atmospheric N deposition directly on the water surface is an important source of N in water bodies. The largest sources of N load in the surface water were rural human and animal excreta and domestic sewage contributed 26.5%. Urban domestic water and human excreta, industrial waste, accounting for 25.7% of the total load. N leaching and runoff from farmland was the third most important source of N load, accounting for 17.9%. Runoff from other land uses also contributed 4.8%. Due to the wide cultivation of crabs and fish, aquaculture in the county also contributed 8.1% to the N load. The huge amount of N load to surface water would result in N concentrations of> 6.0 mg N L-1 even after denitrification removal in wetlands.3. The total N inputs in China mainland increased from 3081 kg km-2 in 1985 to 5426 kg km-2 in 2007. Chemical fertilizer N consumption dominated N input and accounted for 53.6% of the total N input in 2007. Atmospheric N deposition increased continuously, from 767 to 1300 kg N km-2 during 1985-2007. While the total amount of N2 fixation changed little from 1985 to 2007, its contribution to total N input decreased from 32.5 to 18.9%. Net N input through food/feed import increased steadily. Although there was net export of grains such as rice and maize in the past two decades, import of soybean with high N concentration increased greatly during 1995-2007.At a provincial scale, there was large spatial variability in total N inputs, ranging from 588 to 50582 kg N km-2 yr-1 for the Tibet Autonomous Region and the Shanghai Municipality, respectively. Total N inputs of different provinces in different years were significantly correlated with cropland areas, since chemical N fertilizer was the dominant source of N input. As a result, there was a large total N input in provinces in eastern and central China (e.g. Jiangsu, Sandong, Henan and Anhui) where land use is predominantly agriculture. Relatively large N inputs were also found in southern and southeastern provinces (e.g. Zhejing, Fujian, and Guangdong) where with high per capita Gross Domestic Product and 50% of the land area was forest, which has a higher N2 fixation rate. It is no wonder that the vast western area had very low N input as the major land use was desert. N input was relatively low in the most northeastern part of China due to the low chemical N fertilizer application rate and low crop index. Total N input steadily increased in all provinces during 1985-2007. The highest increase was in Ningxia and Tianjin, where total N input increased about tripled from 1985 to 2007. Tibet had the lowest increase in total N input, with a rate< 10%, due to N2 fixation being the dominant source of N input in Tibet, which changed little during the period. For most regions, the increase rate of total N input was 0-200 kg N km-2 yr-1, with greater increase rates in eastern and central China.The N that is stored and denitrified was estimated to amount to 1499kg N km-2 in 1985 to 3140 kg N km-2 in 2007, accounting for 48.7% and 57.9% of total N input, respectively. Both denitrification and storage are difficult to quantify by assuming a C/N ratio of 250 for vegetation and a C/N ratio of 10 for soils, the N storage in terrestrial ecosystems in China could be estimated at about 7.9 Tg N yr-1, and thus the total denitrification in Chinese terrestrial ecosystem would be over 20 Tg N yr-1 during the period, indicating that about 16% of the total N input was stored and about 42% was denitrified in Chinese terrestrial ecosystems in recent years.To more precisely account for the spatial and temporal variability in total N input and various N outputs, we conducted a correlation analysis between these N fluxes and land use types, human population density and per capita Gross Domestic Product (GDP). Total N input, N export to water bodies, denitrification and storage were highly correlated with population density, implying that most of the N is of anthropogenic origin. Total N input, N export to water bodies, denitrification and storage also had significant positive correlations with per capita GDP, indicating that economic development may enhance N load. Because of chemical fertilizer N which applied to cropland accounts for about half of the total N input, the percentage of cropland of total land area showed significant positive correlations with total N input and all N outputs. In contrast, the percentage of grassland and forestland of total land area were negatively correlated with total N input and all N output. The exception is biomass burning emission, which was positively correlated with the percentage of forestland of total land area since wood fuel was a major source of biomass burning.We summarized the amount of N export to water bodies for the three river basins (Yangtze River basin, Yellow River basin and Zhujiang River basin). The results showed that, the modeled riverine N exports agree reasonably well with the measured one for the Yangtze River basin and Zhujiang River basin. For the Yellow River basin, however, the estimated riverine N exports are much larger than the observed ones. This is likely due to the interception of the river water for irrigation and other purposes that may lead to low or zero flow in the lower part of the river in certain periods of the year.4. N cycling has been strongly influenced by human activity at different scales in China mainland. Chemical fertilizer was the biggest N input from a 45.5km2 scale watersheds to the whole China mainland and more than half of the N input was denitrificated or storaged in the system. At the small scale, although total N input is high, riverine N output can be <1.5% of the total N input, agricultural activities resulted in much more atmospheric N pollution. While in development area, food and feed import was much higher than the agricultural watershed, and large amount of total N transport to surface water body which resulted in water eutrophication in this area. There were large temporal variabilities in total N input and outputs at the national sacle, and total N input, N export to water bodies, denitrification and storage could be very well explained by human population density. Nitrogen input and major outputs were also positively related to per capita Gross Domestic Product and the percentage of total land area used as cropland. Large amout of N input resulted in soil acidification and the major river water pollution at national scale.更多还原