论文部分内容阅读
为掌握辽东湾北部海域营养盐污染状况及富营养化水平,本文基于2014——2016年夏季现场调查资料,分析了该海域无机氮、活性磷酸盐、化学需氧量和叶绿素a的时空分布特征,并建立误差逆向传播算法(Back Propagation)的人工神经网络模型对该海域富营养化水平进行评价.结果表明:受201 5年夏季辽宁省干旱和2016年夏季东北地区大规模降雨的影响,辽东湾北部海域营养盐、化学需氧量和叶绿素a浓度均呈现2014年较高,2015年降低,2016年又上升的趋势.调查海域营养盐结构在2014年和2015年主要表现为磷限制,2016年调查海域北部为磷限制,南部为氯限制.BP人工神经网络的评价结果表明,调查海域夏季富营养化水平呈现2014年较为严重,2015年降低,2016年又升高的“U”型特征.富营养化程度严重的区域主要出现在大辽河口、辽河口及其近岸海域.大辽河口富营养化程度在3年夏季均保持较高水平,其中2014年和2016年与辽河口富营养化水平相当,而2015年远高于辽东湾北部其他海域.BP人工神经网络在进行富营养化评价过程中,能够综合考虑各评价指标对海水富营养化的贡献率,避免对营养盐的过度依赖,同时减少评价过程中的主观误差,是一种更加客观合理的富营养化评价方法.“,”In order to control the nutrients pollution status and eutrophication level in the northern part of Liaodong Bay,the characteristics of temporal and spatial distributions of inorganic nitrogen,active phosphate,chemical oxygen demand and chlorophyll-a were analyzedbased on the field survey data in the summer of 2014-2016.The artificial neural network with the error back propagation algorithm was established for the eutrophication level assessment of this sea area.The results showed that the concentration of nutrients,chemical oxygen demand and chlorophyll-a in the northern part of Liaodong Bay were high in the year of 2014 and 2016 but low in 2015affected by the drought in the summer of 2015 and the large scale rainfall in the summer of 2016.Phosphate limitation was the main feature of the surveyed sea area in the year of 2014 and 2015,while in 2016 the northern part of the surveyed area was phosphate limitation and the southern part was nitrogen limitation.The assessment results using BP artificial neutral network showed that the eutrophication level in the surveyed area was high in the summer of 2014 and 2016,and low in the summer of 2015,which performed a “U” type features.The sea area with serious eutrophication mainly occurred in the Liaohe Estuary,the Daliaohe Estuary and the coastal area nearby.In the summer of 2014-2016,the eutrophication level in the Daliaohe Estuary maintained a high level,which was nearly the same as that in the Liaohe Estuary in 2014 and 2016,but much higher than that in the other sea area in the Northern part of Liaodong Bay in 2015.When using BP artificial neural network for eutrophication assessment,the contribution rate of each evaluation index can be comprehensively considered,the over reliance on nutrients can be avoid and the subjective errors can he reduced.Thus the BP artificial neural network could be a more objective and reasonable method for eutrophication evaluation.