基于Stacking的地面PM2.5浓度估算

为了解决地面PM_2.5监测网络在空间和时间覆盖受限的问题,提出了基于宽时空覆盖的卫星气溶胶光学厚度AOD,利用Stacking方法建立地面PM_2.5浓度估算模型,将AOD、PM_2.5和各气象参数以及与PM_2.5排放有关的数据进行训练,使用改进网格搜索对模型超参数进行优化,通过对多重共线性分析,建立基于Stacking的最优PM_2.5浓度估算模型。选取2016-01-01-2016-12-31的数据作为实验对象,结果表明:相比于随机森林、GBRT和XGBoost 3种模型,使用岭回归作为元学习器的Stacking模型性能更优,可见Stacking适用于大范围地理区域的大气污染监测。     

排水管道沉积物微生物群落及环境因子分析

选取北京市某区的排水管道沉积物进行取样,采用高通量测序手段分析,结果表明变形菌门、广古菌门、拟杆菌门、厚壁菌门是排水管道沉积物微生物中的优势门类;在纲水平上,δ-变形菌纲、甲烷微菌纲、梭菌纲、拟杆菌纲占相对优势;在属水平上,功能性微生物硫酸盐还原菌（SRB）和产甲烷古菌（MA）普遍存在于各处管道.在所选的6段管段中,管段S3、S4处MA的相对丰度分别为20.6%、40.8%,高于其他管段,厌氧产甲烷的潜能较大,有发生可燃气积累的风险;管段S5、S6处的SRB相对丰度分别为9.14%、8.19%,高于其他管段,硫酸盐还原为硫化物的潜能较大,存在管道腐蚀的风险.RDA分析表明污水的DO、水温、硫酸根、TN与管道沉积物中微生物群落存在相关性.     

建筑垃圾堆放点样本集构建与优化研究

为更好地解决手工制作的建筑垃圾堆放点样本集效率低、数据量少,难以支撑基于深度学习的遥感图像目标检测算法训练需求的问题,采用基于像素的遥感分类方法构建建筑垃圾堆放点样本集,在此基础上结合直方图均衡化,CS-LBP算子约束以及迁移学习的方法对Wasserstein生成对抗模型（WGAN）进行优化,实现了样本集扩充。研究结果表明:相对于纯手工制作的样本集,基于像素的遥感分类方法可以显著提升样本集制作的效率;同时,经过WGAN优化后,生成样本模拟了原始数据的颜色与纹理特征分布规律,增加了原始数据的多样性,满足了扩充样本集的需求。     

Regulated emission characteristics of in-use LNG and diesel semi-trailer towing vehicles under real driving conditions using PEMS

On-road driving emissions of six liquefied natural gas(LNG) and diesel semi-trailer towing vehicles(STTVs) which met China Emission Standard IV and V were tested using Portable Emission Measurement System(PEMS) in northern China.Emission characteristics of these vehicles under real driving conditions were analyzed and proved that on-road emissions of heavy-duty vehicles(HDVs) were underestimated in the past.There were large differences among LNG and diesel vehicles, which also existed between China V vehicles and China IV vehicles.Emission factors showed the highest level under real driving conditions, which probably be caused by frequent acceleration, deceleration, and start-stop.NOx emission factors ranged from 2.855 to 20.939 g/km based on distance-traveled and 6.719–90.557 g/kg based on fuel consumption during whole tests, which were much higher than previous researches on chassis dynamometer.It was inferred from tests that the fuel consumption rate of the test vehicles had a strong correlation with NOx emission, and the exhaust temperature also affected the efficiency of Selected Catalytic Reduction(SCR) aftertreatment system, thus changing the NOx emission greatly.THC emission factors of LNG vehicles were 2.012–10.636 g/km, which were much higher than that of diesel vehicles(0.029–0.185 g/km).Unburned CH_4 may be an important reason for this phenomenon.Further on-road emission tests, especially CH_4 emission test should be carried out in subsequent research.In addition, the Particulate Number(PN) emission factors of diesel vehicles were at a very high level during whole tests, and Diesel Particulate Filter(DPF)should be installed to reduce PN emission.     

Synergistic effects of Cu species and acidity of Cu-ZSM-5 on catalytic performance for selective catalytic oxidation of n-butylamine

In this work, a series of Cu-ZSM-5 catalysts with different SiO₂/Al₂O₃ ratios (25, 50, 100 and 200) were synthesized and investigated in n-butylamine catalytic degradation. The n-butylamine can be completely catalytic degradation at 350°C over all Cu-ZSM-5 catalysts. Moreover, Cu-ZSM-5 (25) exhibited the highest selectivity to N₂, exceeding 90% at 350°C. These samples were investigated in detail by several characterizations to illuminate the dependence of the catalytic performance on redox properties, Cu species, and acidity. The characterization results proved that the redox properties and chemisorption oxygen primarily affect n-butylamine conversion. N₂ selectivity was impacted by the Brønsted acidity and the isolated Cu²⁺ species. Meanwhile, the surface acid sites over Cu-ZSM-5 catalysts could influence the formation of Cu species. Furthermore, in situ diffuse reflectance infrared Fourier transform spectra was adopted to explore the reaction mechanism. The Cu-ZSM-5 catalysts are the most prospective catalysts for nitrogen-containing volatile organic compounds removal, and the results in this study could provide new insights into catalysts design for VOC catalytic oxidation.     

基于边介数的省域生态廊道构建方法优化

我国城市建设和生态保护工作均对土地资源有大量需求,二者之间的矛盾在经济优先发展区表现尤为明显.为了有效地改善生态环境,管控土地利用并引导其变化发展,需要建设具备不可替代特征的省域生态廊道.最小累积阻力模型（minimum cumulative resistance,MCR）是识别生态廊道最常用、有效的模型,但在应用于省域尺度时,MCR模型识别的潜在廊道路由存在冗余的问题.因此,通过引入网络科学中的边介数指数（edge-betweenness）对MCR模型进行优化,计算潜在廊道路由的边介数指数值,选取出其中最为重要和简明的结构来连通生态源地,即提取潜在路由中的骨干路由（backbone route）和关键战略点（key strategic point）作为不可替代的结构来指导省域生态廊道建设.将优化后的MCR模型应用于广东省,构建了全长5 493 km的省域生态廊道,其中包含生态源地20处,关键战略点11个,骨干生态廊道29条.骨干路由与关键战略点构成的不可替代省域生态廊道（irreplaceable provincial corridor）能够实现"廊道数量和占地面积最少、连通性基本不变"的目标.研究显示,边介数能够对潜在路由进行优化筛选,识别出维护省域生态安全的关键结构;不可替代生态廊道能够指导省域生态规划和土地空间的发展利用,并为更高水平的生态安全环境提供了演进的基础;同时也为土地资源紧张的地区提供了建设生态廊道的参考与依据.      

Chemical characteristics of road dust PM2.5 fraction in oasis cities at the margin of Tarim Basin

In order to understand the compositions characteristics of particulate matter with aerodynamic diameter less than 2.5 μm (PM_2.5) fraction in road dust (RD_2.5) of oasis cities on the edge of Tarim Basin, 30 road dust (RD) samples were collected in Kashi, Cele, and Yutian in the spring, 2018, and RD_2.5 was collected using the resuspension approach. Eight water-soluble ions, 39 trace elements and 8 fractions of carbon-containing species in PM_2.5 were analyzed. Ca²⁺ and Ca were the most abundant ions and elements in RD_2.5 (7.1% and 9.5%). Cl^- in RD_2.5 was affected not only by attributed to saline-alkali soils in oasis cities of the Tarim Basin and dust from Taklimakan Desert but also by human activities. Moreover, the organic carbon/elemental carbon (OC/EC) ratio indicated that carbon components in RD_2.5 in Cele town mainly come from fossil fuel combustion, while those in Yutian and Kashi mainly come from biomass combustion. It is noteworthy that high Ca in RD_2.5 was seriously affected by anthropogenic emissions, and high Na and K contents in RD_2.5 could be derived from soil and desert dust. It was estimated that Cd, Tl, Sn and Cr were emitted from anthropogenic emissions using the enrichment factor. The coefficients of divergence (COD) result indicated that the influence of local emission on road dust emission is greater than that of long-distance transmission. This study is the first time to comprehensively analyze the chemical characteristics of road dust in oasis cities, and the results provides the sources of road dust at the margin of Tarim Basin.     

Analytical methods and environmental processes of nanoplastics

The degradation of plastic debris may result in the generation of nanoplastics (NPs). Their high specific surface area for the sorption of organic pollutions and toxic heavy metals and possible transfer between organisms at different nutrient levels make the study of NPs an urgent priority. However, there is very limited understanding on the occurrence, distribution, abundant, and fate of NPs in the environment, partially due to the lack of suitable techniques for the separation and identification of NPs from complex environmental matrices. In this review, we first overviewed the state-of-the-art methods for the extraction, separation, identification and quantification of NPs in the environment. Some of them have been successfully applied for the field determination of NPs, while some are borrowed from the detection of microplastics or engineered nanomaterials. Then the possible fate and transport of NPs in the environment are thoroughly described. Although great efforts have been made during the recent years, large knowledge gaps still exist, such as the relatively high detection limit of existing method failing to detect ultralow masses of NPs in the environment, and spherical polystyrene NP models failing to represent the various compositions of NPs with different irregular shapes, which needs further investigation.     

外加氮源影响下铁铝氧化物在土壤氮素转化中的作用

土壤中铁铝氧化物在团聚体稳定性和有机碳吸附方面具有重要作用,而氮添加对土壤氮循环影响的变化也可能与其有关,但是目前尚缺乏在氮循环方面的研究.为了探究铁铝氧化物在土壤氮素转化中的作用,选择福建省建瓯罗浮栲森林土壤为研究对象,采用选择性溶提技术准备不同的土壤——未经处理（T1）的土壤和去除游离态铁铝氧化物（T2）土壤、去除非晶质铁铝氧化物（T3）土壤、去除络合态铁铝氧化物（T4）土壤,在这些土壤中添加不同形态氮（40 mg/kg）——丙氨酸（氨基酸态氮,AA）、硫酸铵（铵态氮,AN）、硝酸钠（硝态氮,NAN）和亚硝酸钠（亚硝态氮,NIN）,进行室内培养试验,分析氮含量变化和氮素转化情况.结果表明:①与CK处理相比,AA和AN处理均增加了T1土壤中w（NH₄+-N）,NAN处理增加了w（NO₃--N）,但低于添加量,表明添加氨基酸和铵态氮均会促进氮矿化,添加硝态氮会增加NO₃--N的固定且抑制其硝化.②在CK处理下,与T1土壤相比,T2和T4土壤中w（NH₄+-N）、w（NO₃--N）和w（氨基酸）均降低,但T3土壤中w（NH₄+-N）和w（氨基酸）增加、w（NO₃--N）降低,表明土壤中游离态氧化铁铝和络合态氧化铁铝的存在有助于氮素矿化,非晶质氧化铁铝有助于硝化.③在不同氮处理下,各土壤的氮含量及其转化速率与CK处理规律相似.与CK处理相比,各氮处理均未显著增加T2和T4土壤中w（NH₄+-N）,且AA和AN处理均未影响T2、T3和T4土壤中w（NO₃--N）和w（氨基酸）.结果显示,氮添加并没有改变铁铝氧化物的作用,其中,矿化和氨化作用均表现为游离氧化铁铝>络合氧化铁铝>非晶质氧化铁铝,硝化作用表现为非晶质氧化铁铝>游离氧化铁铝>络合态氧化铁铝.因此,土壤铁铝氧化物的不同存在状态应该是调节氮素转化的重要土壤条件.      

基于野外原位抽注试验和同位素技术的含水层反硝化速率研究

反硝化作用是地下水硝酸盐污染去除最重要的过程.由于水文地质条件和水文地球化学环境的复杂性和不确定性,精准测定含水层反硝化速率是反硝化过程的研究难点.选取潮白河冲洪积扇中部中国环境科学研究院地下水创新野外基地作为研究区,基于野外原位试验和15N同位素示踪法提出一种含水层反硝化速率的测定方法.该方法综合体现了研究区实际水文地质条件和水文地球化学环境对反硝化作用的影响,并充分考虑了硝酸盐在含水层中稀释弥散作用对计算结果的影响.结果表明:①潮白河冲洪积扇中部某地地下26~28 m处于还原环境,含水介质以粉细砂为主,ρ（NO₃-N）平均值为2.77 mg/L.②地下26~28 m反硝化速率在349.52~562.99 μg/（kg·d）（以N计,下同）之间,平均值为450.31 μg/（kg·d）.通过与研究区含水介质、采样深度和硝酸盐背景值相似的国内外案例对比研究,初步评估结果处于合理区间.③测试结果具有一定不确定性,主要来自忽略中间产物NO₂-和NO的计算方法、扰动采样方法、N₂O的操作规范程度及采样频率等方面.研究方法为测定含水层硝酸盐速率研究提供了新的思路,研究结果可为地下水中硝酸盐转化过程机理研究、地下水硝酸盐污染修复及风险管控提供关键的理论支撑数据.