基于GA-SVM的渤海湾富营养化模型

为了更好地模拟和认知渤海湾富营养化的复杂行为,通过研究遗传算法(GA)和支持向量机(SVM)的结合形式,即参数寻优和特征选择,以渤海湾水质实测资料为依据,叶绿素a的质量浓度为输出,建立了GA-SVM的富营养化模型.无特征选择时,用遗传算法对支持向量机的参数(惩罚参数和核参数)进行自适应地优选,预测模型的均方误差可达到1.831μg/L,具有较好的认知、泛化能力.再利用遗传算法二进制编码及启发式寻优的优点,对所建模型的输入空间进行特征选择,提取出代表性的特征变量:DO%、pH值、水温、COD、盐度以及氨氮.特征提取后预测模型的均方误差可达到1.363μg/L,模型性能有了很大提高.分析表明,COD、盐度及氨氮可作为人为控制的首要指标.     

深圳湾海域氮磷营养盐变化及富营养化特征

根据 1986?2007 年深圳湾海水水质监测资料, 阐述了氮磷营养盐的变化趋势及其相互关系, 并探讨了深圳湾海域富营养化程度。结果表明, 深圳湾海域水体溶解态无机氮含量年际变化幅度较大, 但总体呈上升趋势。氨氮是溶解态无机氮的主要组成形态, 表明深圳湾无机氮处于热力学不平衡状态。深圳湾海域水体中活性磷酸盐值含量变化幅度比较小, 总体呈下降趋势。深圳湾水体 DIN/ P 值在 1998 年以前呈不断上升趋势, 1998?2002 年之间呈现波动变化, 从 2002 年开始降低, 营养盐含量和组成的改变导致浮游植物群落结构的改变, 深圳湾海域发生的赤潮生物种类有从硅藻向甲藻过渡, 然后从甲藻 向硅藻过渡的趋势;相关分析表明, 溶解态无机氮以及氮磷营养盐之间的相关关系显著( r = 0. 769,p < 0. 01) , 说明氮磷营养盐间有一定程度相似的生物地球化学过程;水体营养状态质量指数显示深圳湾海水富营养化程度很高, 高度富营养化的水体使得深圳湾海域的赤潮日趋频繁, 控制入湾水体污染刻不容缓。     

自来水中添加N,P后的浮游藻类生长试验

为了解水体营养物质低限浓度对浮游植物生长的影响,于2006年4月研究了浮游植物在自来水及添加不同浓度氮和磷后的自来水中生长情况.结果表明,在自来水中,浮游藻类仍然能够大量生长繁殖,培养28 d后个体密度可达8.00×104 inds·L-1.在添加总氮(TN)浓度分别为0.8,0.4,0.2,0.1,0.05和0.025 mg·L-1的自来水样中,浮游藻类浮游藻类密度分别达到1.69×105,2.81×105,1.19×105,1.03×105,5.88×104和8.12×104 inds·L-1.在添加总磷(TP)浓度分别为0.16,0.08,0.04,0.02,0.01和0.005 mg·L-1 的自来水样中,密度分别达到1.71×106,7.23×105,5.13×105,5.00×105,3.85×105和3.75×105 inds·L-1.藻类密度与添加的TN浓度之间关系为y=-88 650x2 895 456x 26 326,与TP浓度之间关系为y=8×106x 262 582.添加P对藻类生长的影响比N更为显著.控制水体氮磷浓度难以达到控制浮游植物大量繁殖的目的.     

基于模糊神经网络的太湖富营养化评价应用

根据湖泊营养化程度影响因素多,评价因素与富营养化等级间的关系复杂且是非线性的.而神经网络和模糊系统各有优点,研究者将二者结合建立模糊神经网络模型用于太湖营养化评价,当固定学习速率η大于0—1内某一值,将导致网络算法不收敛,因此文中采用自适应调整学习速率,实验表明,该模型具有较快的训练速度和较高的精度.     

深水网箱养殖对海区水质影响研究

通过对箬跳、青山2个深水网箱养殖区的水质状况的调查数据分析发现,养殖海区的无机氮和氨氮含量明显升高,但是深水网箱养殖海区的总体污染状况要比海区小。在养殖产品起捕后,由于海流作用,深水网箱养殖海区的水质很快就达到与未开展养殖海区一致。     

北京大学未名湖水质与底泥菌群结构的时空变化

通过测定不同季节北京大学未名湖不同区域的水质指标,发现在无显著人为外源影响的条件下,水体氮相关指标较好,四季全部采样点的氨氮都优于Ⅱ类水指标,总氮都优于Ⅳ类水指标,但COD,TP和DO等指标在有的采样点仅达到、甚至劣于Ⅴ类。未名湖的平均综合营养状态指数(TLI)为56.1,属于轻度富营养化。利用PCR-DGGE分析未名湖的底泥微生物群落,结果表明其多样性随季节变化较明显,春、秋季多样性指数都较高,夏、秋季多样性指数与地理位置及水体营养盐含量呈显著正相关关系。线性回归结果提示未名湖底泥微生物群落可能受藻类固氮作用的影响。     

Establishing Submersed Macrophytes via Sinking and Colonization of Shoot Fragments Clipped off Manually

In this paper, sinking and growth of apexes and mid-stems of Myriophyllum spicatum L., Hydrilla verticillata （L.f.） Royle and Ceratophyllum demersum L. in concrete ponds containing eutrophic water and sediment were investigated. Sinking rates of apexes and mid-stems reached 34.8% and 4.4% at the 6 th day and 91.1% and 66.7% at the 22 nd day for M. spicatum, 57.8% and 55.6% at the 6 th day and 100% and 97.8% at the 22 nd day for H. vertieillata, 18.9% and 86.7% at the 6 th day and 95.6% and 100% at the 22 nd day for C. demersum, respectively. Most sunken fragments established themselves successfully with significant growth. Total shoot length ofplantlets developed from apexes and mid-stems increased by 399% and 61% for M. spicatum, 593% and 256% for H. vertieillata and 114% and 104% for C. demersum, respectively. The results showed that it was feasible to establish submersed macrophytes via sinking and colonization of shoot fragments clipped off manually.     

富营养水体中3种沉水植物的生长竞争及其净化效果

研究苦草、黑藻和马来眼子菜3种沉水植物在太湖富营养化水体中的生长竞争状况及其对水体的净化效果．结果表明：3种沉水植物短期内可以在混合种群中共存,但竞争优势差别明显．黑藻具有在水中形成上位冠层、根系发达的特性,与苦草、马来眼子菜相比在占领水体上层空间、获取光照和吸收营养方面优势明显,表现出较强的竞争优势;马来眼子菜与苦草之间的竞争不明显,其占据水面面积大,获取光照多,苦草则主要占据水体下层空间,地下部分发达,混种时两者生物量相当．它们对水体中总氮、总磷和藻类叶绿素a的降低作用明显,单种时以黑藻对总氮、总磷的去除率最高,苦草对藻类叶绿素a的降低效果最好;混种时以苦草＋黑藻组对总氮去除率最高,以苦草十黑藻＋马来眼子菜组对总磷去除率最高．     

Late Devonian red tide and mass extinction

Molecular stratigraphical, carbonate carbon isotopic, stratigraphical and paleontological data show that algal booming, eutrophication, anoxia, hypersalinity, positive δ¹³C excursion and biomass decreasing occurred in the offshore carbonate environments of the Frasnian-Famennian (F-F) transition, which hints that red tide might frequently take place in the F-F transition of Guangxi, South China. We suggest that the mass extinction of the reef ecosystems and the shallow-water marine organisms in the F-F transition of the lower-middle latitudes may be related to the frequent occurrence of red tide in that time.     

Mitigation of lake eutrophication: Loosen nitrogen control and focus on phosphorus abatement

Traditionally, nitrogen control is generally considered an important component of reducing lake eutrophication and cyanobacteria blooms. However, this viewpoint is refuted recently by researchers in China and North America. In the present paper, the traditional viewpoint of nitrogen control is pointed out to lack a scientific basis: the N/P hypothesis is just a subjective assumption; bottle bioassay experiments fail to simulate the natural process of nitrogen fixation. Our multi-year comparative research in more than 40 Yangtze lakes indicates that phosphorus is the key factor determining phytoplankton growth regardless of nitrogen concentrations and that total phytoplankton biomass is determined by total phosphorus and not by total nitrogen concentrations. These results imply that, in the field, nitrogen control will not decrease phytoplankton biomass. This finding is supported by a long-term whole-lake experiment from North America. These outcomes can be generalized in terms that a reduction in nitrogen loading may not decrease the biomass of total phytoplankton as it can stimulate blooms of nitrogen-fixing cyanobacteria. To mitigate eutrophication, it is not nitrogen but phosphorus that should be reduced, unless nitrogen concentrations are too high to induce direct toxic impacts on human beings or other organisms. Finally, details are provided on how to reduce controls on nitrogen and how to mitigate eutrophication.