Support vector machine―an alternative to artificial neuron network for water quality forecasting in an agricultural nonpoint source polluted river?

Water quality forecasting in agricultural drainage river basins is difficult because of the complicated nonpoint source (NPS) pollution transport processes and river self-purification processes involved in highly nonlinear problems. Artificial neural network (ANN) and support vector model (SVM) were developed to predict total nitrogen (TN) and total phosphorus (TP) concentrations for any location of the river polluted by agricultural NPS pollution in eastern China. River flow, water temperature, flow travel time, rainfall, dissolved oxygen, and upstream TN or TP concentrations were selected as initial inputs of the two models. Monthly, bimonthly, and trimonthly datasets were selected to train the two models, respectively, and the same monthly dataset which had not been used for training was chosen to test the models in order to compare their generalization performance. Trial and error analysis and genetic algorisms (GA) were employed to optimize the parameters of ANN and SVM models, respectively. The results indicated that the proposed SVM models performed better generalization ability due to avoiding the occurrence of overtraining and optimizing fewer parameters based on structural risk minimization (SRM) principle. Furthermore, both TN and TP SVM models trained by trimonthly datasets achieved greater forecasting accuracy than corresponding ANN models. Thus, SVM models will be a powerful alternative method because it is an efficient and economic tool to accurately predict water quality with low risk. The sensitivity analyses of two models indicated that decreasing upstream input concentrations during the dry season and NPS emission along the reach during average or flood season should be an effective way to improve Changle River water quality. If the necessary water quality and hydrology data and even trimonthly data are available, the SVM methodology developed here can easily be applied to other NPS-polluted rivers.     

Laccase production by Phomopsis liquidambari B3 cultured with food waste and wheat straw as the main nitrogen and carbon sources

Food waste is the most difficult waste to manage because of the high treatment costs and the risk of environmental contamination. Wheat straw burning has become an increasingly discussed topic in recent years because of the air pollution it causes. However, food waste and wheat straw contain high amounts of nutrient elements; efficient utilization of these wastes can improve the environment by lessening airborne pollutant emission and the amount of waste that must be landfilled. It is well known that use of low-cost and abundant waste materials in microbial fermentations can reduce product costs. We aimed to use these resources while improving laccase production by the endophytic fungus Phomopsis liquidambari B3. We cultured P. liquidambari B3 in medium containing food waste and wheat straw as the main nitrogen and carbon sources, respectively. We optimized the fermentation conditions by response surface methodology (RSM), using a Box–Behnken design for RSM I and a central composite design for RSM II. Optimization resulted in an 11.07-fold (1.98-fold RSM I; 5.59-fold RSM II) increase in laccase yield compared with that before optimization. The model was validated by mathematical evaluations and by comparisons between predicted and experimental values. Under optimized conditions, 53.76% of lignin in wheat straw was degraded. By optimizing fermentation conditions and using multiple bioresources, laccase production by this fungus was increased. These results provide the foundation for future research and for scaled-up laccase production.

Implications:Food waste and wheat straw are waste products, but they could be bioresources if they were managed properly. In this work, we innovatively used a mixture of food waste and wheat straw as a substrate for the novel strain Phomopsis liquidambari B3 to produce laccase. This process, which provided the foundation for the subsequent research and for scaled-up laccase production, offers solutions for both rural and urban pollution problems: that is, it reduces the amount of waste material that needs to be disposed of by burning or dumping, and it also produces a valuable product.     

Cadmium accumulation characteristics of the winter farmland weeds Cardamine hirsuta Linn. and Gnaphalium affine D. Don

In a preliminary study, we found that the cadmium (Cd) concentrations in shoots of the winter farmland weeds Cardamine hirsuta Linn. and Gnaphalium affine D. Don exceeded the critical value of a Cd-hyperaccumulator (100 mg kg^?1), indicating that these two farmland weeds might be Cd-hyperaccumulators. In this study, we grew these species in soil containing various concentrations of Cd to further evaluate their Cd accumulation characteristics. The biomasses of C. hirsuta and G. affine decreased with increasing Cd concentrations in the soil, while the root/shoot ratio and the Cd concentrations in shoot tissues increased. The Cd concentrations in shoots of C. hirsuta and G. affine reached 121.96 and 143.91 mg kg^?1, respectively, at the soil Cd concentration of 50 mg kg^?1. Both of these concentrations exceeded the critical value of a Cd-hyperaccumulator (100 mg kg^?1). The shoot bioconcentration factors of C. hirsuta and G. affine were greater than 1. The translocation factor of C. hirsuta was less than 1 and that of G. affine was greater than 1. These findings indicated that C. hirsuta is a Cd-accumulator and G. affine is Cd-hyperaccumulator. Both plants are distributed widely in the field, and they could be used to remediate Cd-contaminated farmland soil in winter.     

湖泊重要入湖河流稳定水像元提取技术

利用2010年ALOS影像研究了归一化水体指数(NDWI)在两种极端情况下,即像元四边与河岸线最小夹角分别为0°和45°,与离岸距离之间的关系,提出了获得稳定水像元的最小河宽,并以太湖6条重要入湖河流(西苕溪、望虞河、太浦河、乌溪港、梁溪-东清河、武进港)为例,研究了稳定水像元的分布。结果表明,NDWI与离岸距离之间的关系可用Logistic方程进行拟合,两种极端情况的p值均小于0.01;最小夹角为0°和45°的情况下,获得稳定水像元的最小河宽分别为96 m和88 m,即当河宽大于96 m时,可在ALOS影像上获得稳定的水像元;6条入湖河流可获得稳定水像元的长度占总长的4.59%~100%,难以获取稳定水像元的最长距离为0~18.55 km。     

固相萃取-气相色谱法测定地表水和底泥中硝基苯类化合物

采用固相萃取-气相色谱法测定地表水和底泥中9种硝基苯类化合物。方法在0mg／L—4.00mg／L范围内线性良好,硝基苯和间-二硝基苯的检出限为0.05mg／L（水样）和0.05mg／kg（底泥）,其余7种硝基苯类化合物的检出限为0.01mg／L（水样）和0.01mg／kg（底泥）,标准溶液平行测定的RSD≤6.5％,地表水和底泥加标回收率分别为85.1％—103％和78.4％—106％。     

The Avoidance Responses of Daphnia magna to the Exposure of Organophosphorus Pesticides in an On-Line Biomonitoring System

In this study, avoidance behavior of the freshwater cladoceran Daphnia magna Straus was used as indicator to assess the early stress of accidental organophosphorus pesticide (OP) contamination. The movement behavior was detected by a multi-species biomonitoring system. There was obvious concentration–response relationship between the OP stress and the behavioral response even at sublethal exposure. A rising OP stress resulted in a significant decrease of response time to escape (RTE; p<0.05). In comparison of different OPs, it was found that there was a power regression between RTE and the toxic unit of OPs. Therefore, the avoidance behavior of D. magna was a sensitive indicator of sublethal OP stress, and the power relationship could be used to predict the early warning thresholds of more OPs in the on-line biomonitoring system.     

土地利用景观格局对信江水质的影响

利用GIS空间分析与统计方法,从景观尺度和类型尺度两方面分析了流域景观格局空间分异对河流高锰酸盐指数(CODMn)、氨氮、总氮(TN)、化学需氧量(CODCr)的影响。信江流域的景观组成对CODMn、氨氮、TN、CODCr浓度存在显著影响,耕地和建设用地的面积比例与各项指标浓度间存在显著正相关,林地与各指标浓度存在显著负相关。各项指标在流域上游变化不大,而在流域的下游变化显著。从景观尺度上看,流域景观以少数大斑块为主或同一类型的斑块高度连接时,河流中CODMn、氨氮、TN、CODCr浓度较低,水质较好。从流域类型尺度上看,各类型的景观结构对河流中CODMn、氨氮、TN、CODCr浓度影响不同,建设用地以及耕地的集中大面积彼此相临的连片分布会导致河流中CODMn、氨氮、TN、CODCr等浓度的升高,而林地则表现出相反的效应。     

夏季对流云降水资源分析

对2002—2003年7—9月间夏季对流云降水观测取得的36个对流云降水单体资料,利用数理方法进行了计算、处理和分析,结果表明：合并对流云降水单体液态含水量,比强对流云降水单体液态含水量偏少269．76&#215;10^6m^3,偏少48．45％;比普通对流云降水单体液态含水量偏多198．68&#215;10^6m^3,偏多2．25倍;强对流云降水单体降水效率最大,为25．74％,比普通对流云降水单体偏多15．74％,比合并对流云降水单体偏多13．21％。     

渤海海面冰情变化趋势预测

渤海地处中国北方,是我国唯一出现海冰的海域。海冰影响着航运以及海上油气的开发和生产。渤海海面冰情的发展和生消,与大尺度环流形势和海区气象水文条件密切相关。以环渤海地区的气候特征为背景,分析冷暖空气活动的特点,做出气温变化的预测,进而推测海面冰情的发展趋势。分析了1990—1999年冬季渤海周边10个气象台站的气温资料,挑出有明显降温的个例,并将当时的500hPa的高度场形势归为：远槽型、近槽型、低压型、横槽型、锋区型和槽后型共6种类型。不同的天气类型反映冷空气活动的不同路径。根据天气分型,选择500hPa和850hPa天气图中反映冷空气活动和描述冷暖气团状态的气象要素作为方程因子,如高空冷暖平流强度、冷气团强度、锋区强度、冷空气引起的降温幅度以及偏』匕风速等,建立10个气象台站的24,48和72h变温预报方程。方程平均拟合率在60％以上。将预测的变温对冰情的影响分成6个等级。验证结果表明,预报方程式能够预报出渤海海面冰情变化趋势,与海冰遥感监测结果一致。     

重庆典型岩溶区地下河中溶解态正构烷烃、脂肪酸来源、迁移及转化

为阐述青木关地下河中溶解态正构烷烃和脂肪酸的来源、迁移及转化研究,2013年7月31日、10月25日分别在青木关地下河入口、天窗和出口处进行采样,并利用气相色谱-质谱联用仪(GC-MS)对样品中溶解态正构烷烃、脂肪酸的组分进行定量分析.结果表明,7月和10月样品中溶解态正构烷烃、脂肪酸的平均含量分别为1 354、667 ng·L-1和24 203、2 526ng·L-1.溶解态正构烷烃和脂肪酸的含量随地下河运移距离的增加均呈降低的趋势;基于正构烷烃分子特征参数CPI、OEP、Paq和R(ΣC≤20含量与总量的百分比)发现7月青木关地下河中溶解态正构烷烃主要来源于细菌等微生物和藻类.10月主要来源于地表水生植物,但随着地下河运移距离的增加,藻类和细菌等微生物的贡献逐渐增大;溶解态脂肪酸C16:0比例最高,结合碳峰分布特征显示7月和10月水样中,藻类和细菌等微生物为地下河中溶解态脂肪酸的主要来源.