| In this study,we took the Irtysh River in Xinjiang Uygur Autonomous Region as a model river and measured the physicochemical parameters of water bodies at nine typical sampling sites in May,July,September 2015 and May 2016.Meanwhile,we collected two key biological groups(periphyton and macrozoobenthos).The water quality was systematically evaluated in the Irtysh River using three comprehensive indices of water quality assessment(water quality index,periphyton and macrozoobenthos indices of biotic integrity)that consist of physicochemical and biological methods.In addition,the key indicators were selected to build high-efficiency models of water quality assessment and the comparative analysis was conducted based on the results of the three methods.The primary results were shown as follows:1.Analysis of physicochemical parameters showed that,except for TP and NO2--N,other physicochemical factors elucidated significant differences between six types of land use.The mean values of Cond and TDS were the highest in the dam area(136.45 ?S/cm and 94.44 mg/L for Cond and TDS,respectively).The mean value of DO was the highest in the pasture and mining area(10.98 mg/L).The average value of CODMn,Ca2+,Mg2+and TSS were the highest in the mining area(4.12 mg/L,23.35 mg/L,3.86 mg/L and71.88 mg/L for CODMn,Ca2+,Mg2+ and TSS,respectively).In addition,the areas with higher nutrient concentrations were near the city,estuary and forest,and the highest NH4+-N value was 0.11 mg/L in the city and estuary.The highest average NO2--N and TN concentrations were 1.06 mg/L and 1.35 mg/L in the city and forest,respectively.Although there was no significant difference between the geographical distribution of TP and NO2--N,higher concentrations were found near the city and estuary.2.Fourteen physicochemical parameters were selected to calculate the WQI.The mean value of WQI was “84.74”,and the water quality was classified as “good”.According to the multiple linear regression analysis,five groups of physicochemical parameters were selected from all parameters to build the candidate WQImin models.The optimal WQImin model was determined by the use of box plots,which consists of 5 water quality parameters: TN,TSS,NH4+-N,Ca2+ and DO.After the correlation coefficientcomparison and percentage error analysis,the WQImin-b3 model with considering weight was available to represent the WQI to the great extent,and considering weight improved the performance of other WQImin models as well.The mean value of WQImin-b3 was“81.98”,the water quality was classified as “good” status,which was consistent with the result of WQI.Through the linear correlation analysis and 95% confidence comparison between WQI and WQImin-b3 model based on test data,A highly significant positive correlation was found between WQImin-b3 and WQI(p<0.001),suggesting that the performance of WQImin-b3 model was excellent to replace WQI for water quality assessment.3.A total of 190 species of periphyton were observed and identified in this study and diatoms were the main dominant species.A total of 28 biological indices were calculated based on the density and biomass of algae,and four optimal indices(Cymbella/(Cymbella+Navicula),biomass,handle type and shannon index)were finally determined to build the P-IBI by comparing the R2 and Co I values of training data.The R2 and Co I values calculated by the testing data was used to testify the performance of the four biological indices.The mean value of P-IBI was “69.44” and the water quality was classified as “moderate”.In addition,a total of 124 species of macrozoobenthos were collected and identified in this study.In particular,aquatic insects were the main dominant species.A total of 23 biological indices were calculated based on the density and biomass of macrozoobenthos and the same way was used as the construction of P-IBI.Four optimal indices(relative aundance of filter feeder,relative abundance of sensitive groups,dominant taxa and biomass)was finally determined.The average B-IBI value was“75.31” and the water quality was classifed as “good”.In addition,the MMI was built for the periphyton and macrozoobenthos,respectively.The mean value of P-MMI was“69.86”,and the water quality was considered as “moderate”,which was consistent with the result of P-IBI.A highly significant positive correlation was showed between P-MMI and P-IBI(p<0.001).The mean value of B-MMI was “74.91”,and the water quality was considered as “good”,which was consistent with the result of B-IBI.A highly significant positive correlation was shown between B-MMI and B-IBI as well(p<0.001).Both the P-IBI and B-IBI developed in this study exhibited a good performance in water qualityassessment.4.WQI,P-IBI and B-IBI were significantly correlated(p<0.05).The scores of WQI ranged between 76.79-91.79,and those of P-IBI and B-IBI were 50.50-80.50 and60.50-82.00,respectively.The highest value of WQI was observed in the Chonghuer Bridge during the flooding period,and the lowest value was in Bieliezeke Copper ore during the pre-flooding period.The highest P-IBI value was in the 635 Dam during the pre-flooding period,whereas the lowest was in Bieliezeke copper ore during the flooding period.The highest B-IBI value was in Keketuohai during the pre-flooding period,whereas the lowest was in Bieliezeke copper ore during the flooding period.Based on hydrological distribution,the average WQI during the pre-flooding,flooding and post-flooding periods were 83.21,84.40 and 86.59,respectively,and the water quality was classified as “good” status.The average P-IBI during the pre-flooding,flooding and post-flooding periods were 74.06,62.28 and 72.00,respectively.And the water quality was “good”,“moderate” and “good”,respectively.The mean values of B-IBI during the pre-flooding,flooding and post-flooding periods were 78.17,71.94 and 75.83,and the water quality was considered as “good”.Based on geographical distribution,the highest mean value of WQI was observed in grazing areas while the lowest in mining areas.The maximum and minimum values were 86.07 and 83.33,respectively,and the water quality was considered as “good”.The highest mean value of P-IBI was 70.67 in the grazing area,whereas the lowest was 67.42 in the mining area,of which water quality grades were“good” and “moderate”,respectively.The highest average B-IBI value was 76.17 in the forest area and the lowest was 73.75 in the mining area,and their grades of water quality were classified as “good”.The scores of P-IBI,B-IBI and WQI based on hydrology,geography and land type showed that the values of WQI were higher than those of B-IBI,and the values of B-IBI were higher than those of P-IBI.In general,the water quality grades of B-IBI and WQI were basically consistent.The P-IBI and B-IBI were sensitive to the water quality variations,whereas the WQI was not sensitive to the water quality variations.5.P-IBI and B-IBI were significantly correlated with DO,TSS,TN,NH4+-N,Ca2+,Mg2+,TP,TSS and CODMn.DO showed significant positive correlation with P-IBI andB-IBI,with correlation coefficients of 0.331 and 0.317,respectively(p<0.01).Other parameters were negatively correlated with P-IBI and B-IBI(p<0.05).When the water quality status changed from “good” to “moderate”,the mean value of DO declined,whereas other parameters increased,and the tendency of the change of physicochemical parameters was consistent between P-IBI and B-IBI results which indicates both of the induces of biotic integrity(IBIs)are applicable to the water quality assessment of the Irtysh River.However,B-IBI performs better in this process due to the water quality assessment grades of B-IBI were closer to the current water quality condition in the Irtysh River. |
PDF Full Text Request