| Algal bloom is a characterization of eutrophication,algal bloom seriously affects the ecological service function of Lake.Therefore,study the bloom-forming mechanism and law is of great practical significance.The enrichment of nutrients is an important cause of algal bloom.The source of eutrophication of lake water bodies includes all kinds of sewage and the pond effluent.Researches have shown that silver carp(Hypophthalmichthys molitrix),bighead carp(Aristichthys nobilis)and aquatic plants can inhibit algae growth,and nitrogen and phosphorus nutrients can be removed from water bodies by silver carp and bighead carp manipulation of lake and aquatic plants restoration,and the input of exogenous nutrients can be reduced by reducing the discharge of pond effluent.These methods can control eutrophication of lake water bodies.However,few studies have reported the influence of nutrients on algae growth under the interference of silver carp,bighead carp and aquatic plants and the production managemental factors on the discharge of pond effluent.In this study,West Yangcheng Lake,polyculture ponds of grass carp(Ctenopharyngodon idellus)and crucian carp(Carassius auratu),and ponds of Chinese mitten crab(Eriocheir sinensis),were used as the project to analyze the effect of water quality factors on algae growth under the interference of silver carp,bighead carp and aquatic plants by Principal component analysis method,and the production managemental factors influencing the discharged amount of pond effluent were also analyzed.And to explore the feasibility of chlorophyll-a(Chl-a)concentration and the discharged amount of pond effluent prediction applying BP neural network algorithm.The aims of the research are to better understand the law of algal growth and occurrence mechanism of algal bloom and to provide a reference for the control of algal bloom.The main research contents and results are as follows:1.Water quality factors affecting Chl-a concentrationBy analyzing and sorting out the releasing data of silver carp and bighead carp and field observation records of West Yangcheng Lake since 2009,it is found that in recent years,the algal bloom in West Yangcheng Lake has been effectively curbed under the manipulation of silver carp and bighead carp.Therefore,the purpose of this study was to investigate the water quality factors affecting the Chl-a concentration under the interference of silver carp and bighead carp.Water temperature,water depth,transparency(SD),pH,dissolved oxygen(DO),total suspended solids(TSS),total nitrogen(TN),total ammonia nitrogen(TAN),nitrite nitrogen(NO2--N),total phosphorus(TP),and active phosphorus(PO43--P),potassium permanganate index(CODMn)and Chl-a concentration were monitored monthly when water temperature was over 20℃ in five sites of West Yangcheng Lake from the year of 2017 to 2018,whereas the data monitored by the same site since 2009 were collected.Total 65 complete water quality records were obtained.The data showed that the average Chl-a concentration of West Yangcheng Lake was 21.44±20.93 μg/L,the comprehensive trophic level index(TLI(∑))was 56.88,which suggested mild eutrophication state.The correlation analysis showed that the Chl-a concentration of West Yangcheng Lake has a positive correlation with pH,TSS concentration,NO2--N concentration,water depth and TAN concentration(P<0.05),while no monitoring index was significantly negatively correlated with Chl-a concentration,and the correlation between water temperature and Chl-a concentration was not significant(P=0.822).Principal component analysis showed that there were 4 principal factors(F)with eigenvalues>1,F1 associated with easily assimilated form of nitrogen,phosphorus concentration and TP concentration.F2 associated with pH and DO,F3 have high load but opposite direction on water temperature and TN concentration,F4 mainly associated with SD.The variance contribution rate of each principal component ranged from 9.246%to 28.810%.These results indicated that the Chl-a concentration increase with the rise of nutrient concentration even if it was inhibited by silver carp and bighead carp,which indicate that control of algal bloom should proceed from several water quality factors simultaneously.Results also indicated that with the increase in water temperature,the growth of the grazing intensity on algae by silver carp and bighead carp in West Yangcheng Lake is faster than the growth of the algae biomass.Different fish ponds have different water quality and different stocking density of silver carp and bighead carp,which provides the possibility to study the water quality factors affecting Chl-a concentration at the same time scale,and is also helpful to analyze the effect of silver carp and bighead carp on Chl-a concentration under high nutrition level.Therefore,a lot of fish ponds were investigated in this experiment,and carries on the data statistical analysis.The water quality of 34 polyculture ponds of grass carp and crucian carp was monitored in July 2018 in farms located in Suzhou and Yancheng City,Jiangsu Province.Water was sampled at 4:00-6:00 p.m.on clear days.Monitoring parameters were the same as above.The yields of silver carp and bighead carp were investigated in harvest season.The results showed that the average Chl-a concentration of these fish ponds was 375.86 ± 221.25 μg/L,TLI(∑)was 82.70,suggesting in a severe eutrophication state.Chl-a concentration has significant positive correlations with TN and CODMnconcentration(P<0.05),water temperature(P=0.066)and TP concentration(P=0.098)also have higher correlations with Chl-a concentration.Principal component analysis showed that there were 4 principal components that have eigenvalues>1,which can be defined as water temperature factor,phosphorus source and easily assimilated form of nitrogen,nitrogen source and SD factor,respectively.The variance contribution rate of each principal component was 12.721%～22.040%,indicated that each principal component has similar influence intensity on Chl-a concentration.The coverage rate of submerged plants in Chinese mitten crab culture ponds is up to 60%,which have the function of purifying the water quality and inhibiting algae.Water quality factors affecting Chl-a concentration in crab ponds were analyzed in this study.In order to reduce the research error caused by the small amount of data,a large number of water quality of crab ponds were randomly investigated in Suzhou in July 2019,and 48 groups of water quality data were obtained.The sampling time was from 3 to 6 pm on clear days,and the monitoring parameters were the same as above.The results showed that the mean Chl-a concentration in crab pond was 35.17±50.58 μg/L,TLI(∑)was 54.82,suggesting in a mild eutrophication state.The results of correlation analysis showed that Chl-a concentration has a significant negative correlation with SD(P<0.05),and has significant positive correlations with TN and CODMn concentration and water temperature(P<0.05).pH(P=0.199),DO(P=0.087),TSS concentration(P=0.097),PO43--P concentration(P=0.090)also have higher correlations with Chl-a concentration.Principal component analysis showed that there are 5 principal components that have eigenvalues>1,and the variance contribution rate ranged from 13.130%to 18.999%.The principal components were difficult to define according to the main related factors.This may due to the influence of aquatic plants on water quality.Compare the relationship between nutrients’ concentrations and Chl-a concentration in three types of water bodies,results indicated that there may be such mechanism:when the exogenous input nutrients is higher than that absorbed by algae,they are positively correlated,that is,the higher the concentration of nutrients,the higher the Chl-a concentration;When the exogenous input nutrients is lower than that absorbed by algae,they are negatively correlated,that is,the faster the growth of algae,the greater the decrease of nutrient concentration.In addition,the results also support that algal growth could be inhibited by silver carp and bighead carp as well as aquatic plants.Meanwhile,results showed that even in the presence of interference with silver carp and bighead carp and aquatic plants,the Chl-a concentration is still positively correlated with the nutrient concentration,suggesting that the algae be bloomed when the nutrient concentration is high enough.The prediction models of Chl-a concentration of West Yangcheng Lake,polyculture ponds of grass carp and crucian carp and ponds of Chinese mitten crab,were constructed using BP neural network.Two models were constructed respectively for polyculture ponds of grass carp and crucian carp according to covering or not the index of silver carp and bighead carp.The results showed that the determination coefficient(R~2)of the Chl-a concentration prediction models of West Yangcheng Lake and the crab pond were 0.85021 and 0.97627,respectively,and the average relative errors were 15.34%and 17.47%,respectively.When the Chl-a concentration model did not cover the index of silver carp and bighead carp,R~2 was 0.97841 with an average relative error of 12.98%;When the index of silver carp and bighead carp were covered,R~2 was 0.92455 with an average relative error of 11.20%.These results suggested that the prediction model has good performance which was further improved when silver carp and bighead carp were considered.The above results show that BP neural network algorithm can predict the Chl-a level of different water bodies.2.Production managemental factors affecting the discharged amount of effluent from fish and crab cultural pondsThe discharged pond effluent in high temperature period is one of the causes of eutrophication and water bloom.Controlling the discharge of pond effluent is helpful to reduce the pollution of water body by exogenous nutrients.Therefore,through a large-scale field visit in Jiangsu Province in 2018,a total of 184 fish ponds and 114 Chinese mitten crab ponds were investigated,including some production managemental data,such as stocking density,feed,water source,artificial aeration,etc.Among them,the complete production managemental data of 123 groups of polyculture ponds of grass carp and crucian carp and 106 groups of ponds of Chinese mitten crab were selected,respectively.Based on the principal component analysis,the production managemental factors affecting the amount of discharged effluent in high temperature period of polyculture ponds of grass carp and crucian carp and ponds of Chinese mitten crab were analyzed,and the prediction model was constructed by BP neural network algorithm.Correlation analysis showed the amount of discharged water from fish ponds was significantly negative correlated with the intensity of artificial aeration,feed types as well as water quality regulation(P<0.05).The principal component analysis showed that there are 6 principal component that have eigenvalues>1,which represent the suitable fishing state of water quality,input materials types and the control effect on algal bloom,the stocking density of silver carp and bighead carp,the culture pattern and pond area,feeding fish species stocking density,water source,respectively.The results suggest that the amount of discharged pond effluent can be reduced by increasing artificial aeration.The amount of discharged effluent from crab ponds was significantly positive correlated with premature senescence of aquatic plants,the quantity of formula feed used,average maximum water depth and water source(P<0.05),and was significantly negative correlated with culture experience(P<0.05).The principal component analysis showed that there are 4 principal components that have eigenvalues>1.The logical relationship between the factors of each principal component was not obvious and the definition was difficult.It might due to the decisive influence of aquatic plants in crab pond on water quality.In addition,the premature senescence of aquatic plants should be prevented to reduce the discharged pond effluent amount.BP neural network algorithm was used to construct the prediction model of the amount of discharged effluent from fish ponds and crab ponds,respectively.R~2 was 0.97028 and 0.99997,respectively,and the average relative error was 14.18%and 0.18%,respectively.All the model errors are small,indicating that the BP neural network algorithm can be used to predict the amount of discharged effluent of fish and crab ponds. |
PDF Full Text Request