基于地理加权回归模型评估土地利用对地表水质的影响

针对传统线性回归模型大多忽视空间数据局部变化特征这一缺陷,引入地理加权回归模型(GWR)用于评估土地利用对地表水质的影响,分析了不同子流域内两者关系出现空间变化的规律并阐释了原因.同时,对比了GWR模型与普通最小二乘模型(OLS)的校正R2、Akaike信息准则(AICc)及残差的空间自相关指数(Moran's I),验证了GWR模型在预测精度和处理空间自相关过程中是否优于OLS模型.结果表明,同一土地利用类型对水质的影响随空间位置的改变而发生方向或大小的变化.以温瑞塘河流域总氮(TN)与农用地的关系为例,从GWR模型局部回归系数的方向分析,两者关系表现为农村正、城区负的现象,从大小分析,旧城区TN与农用地回归系数的绝对值高于其它区域;在溶解氧(DO)与人口密度所构建的GWR模型中,两者关系在整个研究区域内均表现为负值,与OLS结果吻合,从回归系数的大小分析,人口密度对DO的作用在郊区及农村更为显著.针对此类关系出现空间变化的原因分析表明,相邻子流域土地利用百分比的改变及水体主要污染源的不同,是导致土地利用对水质作用发生变化的根本因素.最后,对比所构建的80个GWR与OLS模型校正R2、AICc指标,验证了GWR作为一种局部统计模型,其预测精度优于OLS等传统全局模型且更能反映实际空间特征.     

不同形态的砷水生生物基准探讨及在辽河流域的初步应用

有毒元素砷在我国各流域分布广泛,主要包括三价和五价两种价态,其毒性差别很大,目前我国缺乏本土的砷水质基准研究,而国外的砷水质基准多以总砷表示,科学性有待商榷.因此,本文分别搜集了三价砷、五价砷和总砷对我国本土水生生物的毒性数据,共获得三价砷12属12种,五价砷18种23属,以及总砷25属29种的合格数据.利用美国环保局推荐的物种敏感度排序法,通过数据分析得出,三价砷的急性和慢性基准阈值分别为84.26μg·L-1和56.55μg·L-1,五价砷的急性和慢性基准阈值分别为1171μg·L-1和633.3μg·L-1,总砷的急性和慢性基准阈值分别为264.4μg·L-1和150.7μg·L-1;总砷的基准阈值与美国现行砷基准值相当,而五价砷基准阈值与三价砷基准阈值体现出数量级的差异,因此,需要分价态考虑砷基准的制定.基于推算的砷基准阈值,用商值法评价了辽河流域枯水期总砷暴露的生态风险.检测结果表明,辽河的总砷暴露浓度范围在2.6~82.4μg·L-1之间,全部位点均无风险,但西辽河段上游支流两点位的砷浓度分别超出三价砷慢性基准阈值的1.20和1.45倍,表明可能存在潜在的砷暴露风险.     

基于模糊物元模型的西苕溪流域生态系统健康评价

本文借助RS/GIS技术,结合遥感影像数据、环境监测数据和社会经济数据,运用模糊物元模型,从水域、水陆交错带和陆域子系统选取21项评价指标,建立了流域生态系统健康综合评价指标体系,对西苕溪流域生态系统健康状况的空间分布规律和限制性因子进行了探讨.研究结果表明,西苕溪流域生态系统健康整体呈现出优良状态,其中,生态系统健康优秀地区面积占68.2%,良好地区面积占31.8%,能够正常发挥流域服务功能,维持生态系统可持续发展.西苕溪流域生态系统健康在空间分布上呈现出上游好、下游差的局面.在各子系统健康评价中,水域生态系统优于陆域生态系统,水陆交错带生态系统最差.底栖动物完整性遭受破坏、湿地退化、人为干扰活动、水源涵养功能减弱、点源和面源污染负荷较重是限制西苕溪流域部分评价单元生态系统健康的重要因素.     

钱塘江(杭州段)水中有机氯农药残留污染特征及健康风险评价

利用气相色谱法对采集于2013年5月、7月和11月的钱塘江(杭州段)表层水中10种有机氯农药(OCPs)残留进行了分析,初步明确其组成特征及来源,并对其健康风险进行了评价.结果表明,钱塘江(杭州段)水体中检出5种微量的OCPs,检出频率最高的是β-HCH和δ-HCH.14个采样点均有不同浓度检出,钱塘江干流∑OCPs的浓度范围在1.31~6.68 ng·L-1,属低污染水平,3条主要支流∑OCPs的浓度范围则是1.32~4.23 ng·L-1.钱塘江(杭州段)水体中OCPs污染空间分布各异,上游兰江点位及桐庐县、富阳市境内各点位总体浓度相对较高.通过特征组分比例可确认HCHs污染主要源于农田中使用有机氯类农药的降解残留及新的林丹污染输入,且非固定污染源输入.利用EPA推荐方法对通过饮水和皮肤接触途径摄入钱塘江水体中OCPs的健康风险进行评价:非致癌健康风险指数介于8.56×10-5~1.82×10-3,根据评价标准均未超标;致癌健康风险指数在3.96×10-8~1.93×10-7,均位于可忽略风险范围内.结果表明钱塘江(杭州段)作为饮用水水源,其水体中OCPs残留引起对人体产生的致癌、非致癌健康危害可忽略.     

亚热带丘陵小流域土壤有效磷空间变异与淋失风险研究

肥料过施导致的土壤磷素累积和淋失是农业面源污染的重要方面.以湖南省长沙县金井镇脱甲河小流域(52 km2)为研究区,采用高密度布点采样、Arc GIS软件和属性相似反距离加权插值法研究了亚热带丘陵小流域表层(0~20 cm)土壤有效磷(Olsen-P)含量(以P计,下同)的空间分布特征与磷素的淋失风险.结果表明,菜地、果园、稻田和茶园土壤Olsen-P平均含量为62.0、16.1、14.4和13.7 mg·kg-1,是林地(平均含量为2.36 mg·kg-1)的5.8~26.3倍.5个土地利用类型土壤Olsen-P含量均具有高等变异水平和中等程度的空间自相关性(块基比C0/(C0+C)=50%),这与区内地形地貌、土壤母质、人工施肥等具有密切关系.根据土壤0.01 mol·L-1Ca Cl2浸提态P和Olsen-P的非线性关系可确定区内红壤和水稻土P的淋失风险临界值分别为69.97和98.40 mg·kg-1,并据此对脱甲河小流域土壤磷素淋失的风险进行了定量评价,结果表明旱地土壤具有明显较高的淋失风险,其中中等以上的比例占36.4%,而稻田土壤仅有0.2%,为中等以上淋失风险.因此,控制旱地(尤其是菜地)磷肥的投入是降低亚热带丘陵小流域土壤P淋失风险和减轻农业面源污染的关键.     

“引江济太”过程中塑化剂类污染物的输入特征和环境健康风险评价

为了解"引江济太"过程中塑化剂类污染物对贡湖的输入特征,于2013年8月对14个采样点的7种邻苯二甲酸酯类化合物(PAEs)的浓度进行采样分析,并通过美国环保署(USEPA)推荐的方法,对其环境风险进行评价.结果表明:各采样点邻苯二甲酸二丁酯(DBP)和邻苯二甲酸二环己酯(DCHP)均有检出,除采样点10、11和12外,邻苯二甲酸二乙酯(DEP)也有检出,其余4种均未检出,长江引水经望虞河进入贡湖后,DEP、DBP和DCHP浓度均有不同程度的降低,且其最高浓度均出现在望虞河长江引水处(采样点1),说明在引水过程中长江水对望虞河有明显的PAEs输入,而贡湖锡东水厂DEP浓度高于贡湖其他采样点,南泉水厂DBP和DCHP浓度高于贡湖其他采样点,金墅湾水厂PAEs浓度总体处于较低水平.环境健康风险评价结果显示:DEP的最大非致癌风险值为3.91×10-8a~(-1),未超过国际组织规定范围,而DBP的最大非致癌风险值为1.17×10-5a~(-1),超过英国皇家协会RS规定的1×10-6a~(-1),但均对人体无明显非致癌危害.     

Investigation of carbonyl compound sources at a rural site in the Yangtze River Delta region of China

Carbonyl compounds are important intermediates in atmospheric photochemistry, but their primary sources are still not understood well. In this work, carbonyls, hydrocarbons, and alkyl nitrates were continuously measured during November 2011 at a rural site in the Yangtze River Delta region of China. Mixing ratios of carbonyls and hydrocarbons showed large fluctuations during the entire measurement. The average level for total measured volatile organic compounds during the pollution episode from 25th to 27th November, 2011 was 91.6 ppb, about 7 times the value for the clean period of 7th-8th, November, 2011. To preliminarily identify toluene sources at this site, the emission ratio of toluene to benzene (T/B) during the pollution episode was determined based on photochemical ages derived from the relationship of alkyl nitrates to their parent alkanes. The calculated T/B was 5.8 ppb/ppb, significantly higher than the values of 0.2-1.7 ppb/ppb for vehicular exhaust and other combustion sources, indicating the dominant influence of industrial emissions on ambient toluene. The contributions of industrial sources to ambient carbonyls were then calculated using a multiple linear regression fit model that used toluene and alkyl nitrates as respective tracers for industrial emission and secondary production. During the pollution episode, 18.5%, 69.0%, and 52.9% of measured formaldehyde, acetaldehyde, and acetone were considered to be attributable to industrial emissions. The emission ratios relative to toluene for formaldehyde, acetaldehyde, and acetone were determined to be 0.10, 0.20 and 0.40 ppb/ppb, respectively. More research on industrial carbonyl emission characteristics is needed to understand carbonyl sources better.     

秦皇岛洋河-戴河流域浅层地下水咸化程度评价

为了解秦皇岛洋戴河流域浅层地下水咸化程度,开展了野外调查和水样采集,现场测试EC、TDS指标,并对水样进行了室内水化学分析.应用层次分析法结合Arcgis对选取的6个指标(Cl离子、TDS、Li离子、钠吸附比、硝酸盐、潜在盐度)进行权重计算、归一化处理和栅格计算,得出地下水咸化等级分区(未咸化区、轻微咸化区、中等咸化区、严重咸化区).咸化最严重区域位于沿海平原,包括枣园村、王各庄村.中度咸化区部分位于沿海平原,包括都寨村、西陆庄村、蒋营村、樊各庄,主要是由海水入侵导致的;而樊各庄咸化的原因是海水入侵和工业废水的污染;其他中度咸化区位于低山丘陵区,包括大湾子村、兴隆寨村,则是由水岩作用、农业污水、生活污水造成的.轻微咸化区主要分布在中等咸化区周围,未咸化区集中在山前平原区.     

Long‐Term Trends of Nutrients and Sediment from the Nontidal Chesapeake Watershed: An Assessment of Progress by River and Season

To assess historical loads of nitrogen (N), phosphorus (P), and suspended sediment (SS) from the nontidal Chesapeake Bay watershed (NTCBW), we analyzed decadal seasonal trends of flow‐normalized loads at the fall‐line of nine major rivers that account for >90% of NTCBW flow. Evaluations of loads by season revealed N, P, and SS load magnitudes have been highest in January‐March and lowest in July‐September, but the temporal trends have followed similar decadal‐scale patterns in all seasons, with notable exceptions. Generally, total N (TN) load has dropped since the late 1980s, but particulate nutrients and SS have risen since the mid‐1990s. The majority of these rises were from Susquehanna River and relate to diminished net trapping at the Conowingo Reservoir. Substantial rises in SS were also observed, however, in other rivers. Moreover, the summed rise in particulate P load from other rivers is of similar magnitude as from Susquehanna. Dissolved nutrient loads have dropped in the upland (Piedmont and above) rivers, but risen in two small rivers in the Coastal Plain affected by lagged groundwater input. In addition, analysis of fractional contributions revealed consistent N trends across the upland watersheds. Finally, total N:total P ratios have declined in most rivers, suggesting the potential for changes in nutrient limitation. Overall, this integrated study of historical data highlights the value of maintaining long‐term monitoring at multiple watershed locations.     

Reducing Fertilizer‐Nitrogen Losses from Rowcrop Landscapes: Insights and Implications from a Spatially Explicit Watershed Model

We present conceptual and quantitative models that predict changes in fertilizer‐derived nitrogen delivery from rowcrop landscapes caused by agricultural conservation efforts implemented to reduce nutrient inputs and transport and increase nutrient retention in the landscape. To evaluate the relative importance of changes in the sources, transport, and sinks of fertilizer‐derived nitrogen across a region, we use the spatially explicit SPAtially Referenced Regression On Watershed attributes watershed model to map the distribution, at the small watershed scale within the Upper Mississippi‐Ohio River Basin (UMORB), of: (1) fertilizer inputs; (2) nutrient attenuation during delivery of those inputs to the UMORB outlet; and (3) nitrogen export from the UMORB outlet. Comparing these spatial distributions suggests that the amount of fertilizer input and degree of nutrient attenuation are both important in determining the extent of nitrogen export. From a management perspective, this means that agricultural conservation efforts to reduce nitrogen export would benefit by: (1) expanding their focus to include activities that restore and enhance nutrient processing in these highly altered landscapes; and (2) targeting specific types of best management practices to watersheds where they will be most valuable. Doing so successfully may result in a shift in current approaches to conservation planning, outreach, and funding.