Ammonia Volatilization from Farmland Measured by Laser Absorption Spectroscopy
QUE Hua-li1, 2, YANG Wen-liang1, XIN Xiu-li1, MA Dong-hao1, ZHANG Xian-feng1, ZHU An-ning1*
1. Fengqiu Agro-ecological Experimental Station, State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences, Nanjing 210008, China
2. University of Chinese Academy of Sciences, Beijing 100049, China
摘要： 氨挥发是农田氮素向环境输出的重要途径，也是我国空气中PM2.5形成的主要因素，给环境和农业生产带来了诸多不利影响。传统的农田氨挥发测定大多依赖酸吸收法，但因采样时间长、劳动量大，难以获取氨挥发日内动态变化规律。基于开放光程可调谐二极管激光吸收光谱技术进行田间痕量氨气测定时，测量精度高、选择性好、系统响应速度快，不需要复杂的采样操作，就可以实现激光发射器与反射镜之间数十至数百米的高时间分辨率的氨气浓度原位快速监测。其与微气象反向拉格朗日随机扩散模型相结合（TDLAS-BLS法）是目前农业源氨挥发监测技术领域的研究热点。通过田间试验，分析比较TDLAS-BLS法与微气象水平通量积分法（IHF法）测定的氨挥发速率及氨挥发损失结果，实现对TDLAS-BLS法测定大面积农田氨挥发的可靠性验证。利用监测获取的高时间分辨率数据研究冬小麦追肥期氨挥发日内变化规律及影响因素。结果表明：TDLAS-BLS法和IHF法测定农田氨挥发速率基本一致（斜率为0.97，R2=0.97，n=14），TDLAS-BLS法测定氨挥发速率仅比IHF法低3%，总氨挥发损失仅低6%，证明TDLAS-BLS法可用于冬小麦追肥期大面积农田氨挥发监测中。冬小麦追肥期白天氨浓度明显高于夜间，且受风速波动影响，氨浓度瞬时波动较大。氨挥发速率在追肥后缓慢升高，施肥后第6天出现氨挥发速率峰值8.9 kg N·ha-1·d-1，随后逐渐降低，至第15天与背景接近。氨挥发损失主要集中在施肥后的第5～8 d（79～175 h），该时段氨挥发损失占总氨挥发损失的69%。整个监测期间TDLAS-BLS法测定总氨挥发损失为8.8 kg N·ha-1（占施氮量6.3%），较低的损失量与沟施覆土的施肥方式及低温、低光照强度有关。TDLAS-BLS法实现了在线监测大面积农田氨挥发日内变化规律，高时间分辨率数据可更准确地评估气象因素对氨挥发的影响。冬小麦追肥期氨挥发日内波动较大，存在明显的昼高夜低变化规律，与温度、风速、光照有很高的相符性。相关分析表明风速、光照、土壤温度、降水都与氨挥发有显著相关性，异常天气下主导气象因素（如降水）是氨挥发主要控制因素。
关键词：激光吸收光谱技术；反向拉格朗日随机扩散模型；微气象水平通量积分法；冬小麦；氨挥发；影响因素
Abstract：Ammonia volatilization is an important path of nitrogen loss from farmland into the environment, and is also the main factor of PM2.5 formation, which has serial disadvantageous effect on the ecological environment and agricultural production. Previously, most of the traditional methods for ammonia emission measurement collected atmospheric ammonia by an acid absorbent. However, these techniques were labor intensive, making it difficult to determine diurnal variation of ammonia volatilization. The open-path tunable diode laser absorption spectroscopy is a reliable tool with high precision, high selectivity, and fast response time for continuous and nonintrusive monitoring of ammonia concentrations over distances from tens to hundreds of meters under field conditions. Currently, the combination of tunable diode laser absorption spectroscopy and micrometeorological backward Lagrangian stochastic diffusion model (TDLAS-BLS) has become a popular technique in the measurement of ammonia volatilization in the field. The objectives of the study are first, to compare TDLAS-BLS technique with the micrometeorological integrated horizontal flux method for its accuracy and applicability for quantitatively measuring ammonia emission from large area of farmland through field experiments. Second, determine the dynamics of ammonia volatilization via high-temporal resolution data and identify the factors that govern ammonia volatilization from urea applied to winter wheat. The results indicated that the estimates made by TDLAS-BLS method were statistically equivalent to those made by the IHF method (regression gradient=0.97, R2=0.97, n=14). The ammonia emission rates and total ammonia loss estimated by the TDLAS-BLS technique were only 3% and 6% lower than those from the IHF method, respectively. This implied that TDLAS-BLS technique can be used to quantitatively estimate ammonia emission from large area of farmland during topdressing period of winter wheat with acceptable accuracy. Ammonia concentration was higher in daytime than in the night at topdressing stage of winter wheat, due to wind speed fluctuation causing it to fluctuate greatly. Ammonia volatilization rate increased slowly after fertilization, and reached a maximum value at the sixth day after fertilization and then decreased gradually after 15 days. This was mainly concentrated in the 5～8 days after fertilization, and accounted for 69% of the total during the overall monitoring period. During this period, the total loss determined by TDLAS-BLS method was 8.8 kg N·ha-1 (6.3% of the total applied N). The lower loss was related due to furrow application of urea and low temperature. This demonstrated the ability of the TDLAS-BLS method to characterize the diurnal patterns of ammonia emission and the environmental influences on ammonia emission from cropland via high-temporal resolution data. Ammonia volatilization showed large diurnal variability during the daytime, which was coincident with temperature, wind speed and solar radiation. Wind speed, solar radiation, soil temperature and precipitation are significantly correlated with ammonia volatilization. Meteorological factor (such as precipitation) are the main factors influencing ammonia volatilization in abnormal weather.
[1] Xu P, Zhang Y S, Gong W W, et al. Atmospheric Environment, 2015, 115: 141.
[2] Hu E Z, Babcock E L, Bialkowski S E, et al. Critical Reviews in Analytical Chemistry, 2014, 44(3): 200.
[3] Ungermann J, Blank J, Dick M, et al. Atmospheric Measurement Techniques, 2015, 8(6): 2473.
[4] Stutz J, Werner B, Spolaor M, et al. Atmospheric Measurement Techniques, 2017, 10(3): 1017.
[5] TANG Qi-xing, ZHANG Yu-jun, CHEN Dong, et al(唐七星，张玉钧，陈 东，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析)，2018，38(11): 3328.
[6] Grant R H, Boehm M T, Heber A J, et al. Agricultural and Forest Meteorology, 2016, 228-229: 73.
[7] Wang X, Xu S J, Wu S H, et al. Journal of Environmental Sciences, 2018, 66: 199.
[8] Zhang T, Liu H B, Luo J F, et al. Science of the Total Environment, 2018, 633: 230.
[9] Li M N, Wang Y L, Adeli A, et al. Field Crops Research, 2018, 218: 115.
[10] Bai M, Sun J, Denmead O T, et al. Environmental Pollution, 2017, 230: 584.
[11] Yang W L, Zhu A N, Zhang J B, et al. Agricultural and Forest Meteorology, 2017, 234-235: 196.
[12] Yang W L, Zhu A N, Chen X M, et al. Nutrient Cycling in Agroecosystems, 2014, 99(1-3): 107.
