| With the rapid development of urbanization, the ratio of impervious surface in urban areas increased gradually, leading to the frequent waterlogging by rainstorm and the degradation of water quality of urban storm runoff, and urban storm runoff becomes an important non-point pollution source that can not be ignored. Urban storm runoff from different underlying hard-surface types can transport high pollutant loads to receiving water, including solids, nutrients, heavy metals, oils, hydrocarbons, organic pollutants and bacteria, and so on. Of all pollutants, nutrients such as nitrogen (N) and phosphorus (P), have been paid the most attention due to their role in eutrophication. Moreover, the rapid development of urbanization also causes conspicuous contraction of area, matrix hardening and degeneration of ecological function to urban riparian zone; meanwhile riparian zone is the final hurdle for urban storm runoff to flow into receiving water. Hence, selecting the reasonable technology to carry out the ecological reconstruction for urban riparian zone, increasing the infiltration capacity for urban storm runoff and restoring its original ecological removal ability for pollutants have become one of the hot spots of studies about urban environment in the world today.With the support of the National Key Water Special Project of China, the National Natural Science Foundation of China and the project of Excellent Academic Leaders of Shanghai, Wenzhou, a typical coastal city in southern China, was chosen as the main research area in this study. By means of field investigation, field monitoring, indoor simulation and analysis, environmental behavior characteristics of total phosphorus (TP), dissolved phosphorus (DP) and particulate phosphorus (PP) in different types of urban storm runoff were studied; in addition, and the effects of characteristics of rainfall (Antecedent Dry Weather Period, ADWP; precipitation; rainfall intensity, etc.), regional environment, underlying surface type and other characteristics of storm runoff on phosphorus pollution were also investigated. Based on the data of field observation, pollution loads in one single rainfall event and annual emission fluxes of TP, DP and PP from different urban runoff sources were estimated, and their potential influence factors were discussed at the same time. On the other hand, based on the characteristics of urban storm runoff pollution in Wenzhou, a new type of multilayer filter system was developed, and then practice researches about its application prospect in ecological reconstruction of urban hard-type riparian zone were carried out. The main conclusions can be summed up as follows.(1) The average pH value of rainfall during the study period in Wenzhou was5.83. Although higher than the threshold of acid rain, it was quite close. TP concentrations in rainfall samples were generally low and below the threshold of Class Ⅱ suggested by environmental quality standard for surface water, which implied that summer rainfall in Wenzhou was not a pollution source for urban receiving water.(2) TP, DP and PP concentrations in summer urban storm runoff of Wenzhou ranged from0.01to4.32mg·L-1, ND to0.88mg·L-1and ND to4.31mg·L-1, respectively. Meanwhile, concentrations of different kinds of phosphorus in runoff all showed a downward trend with runoff duration, especially for the concentrations of TP and PP, while the downward trend of DP was not so significant due to its relative low concentration. In the initial period of storm runoff, PP was generally the dominant component of TP in all kinds of runoff, while the major form of P varied over time, especially in both roof runoff and outlet runoff in Jiushanwai River study area, where TDP made up the largest portion in the latter stages of runoff events.(3) There were many kinds of influence factors that could affect the concentrations of P in Wenzhou urban storm runoff. Though the relationships between ADWP and concentrations of TP and PP in runoff were not significant, DP concentrations were positively correlated to AWDP at the0.05level (2-tailed). Instantaneous rainfall intensity also had an influence on P concentrations during the runoff duration, P concentrations would decline sharply just after the rainfall intensity reached its peak, and then rose again slowly until achieving their dynamic balance. Underlying surface type was another important influence factor on P concentrations in storm runoff. Both urban community road and main road runoff were noticeable for their high concentrations of TP and PP, and DP concentration in outlet runoff was significantly higher than others, while P concentrations in roof runoff were almost the lowest among the various runoff sources, which implied that urban roof runoff was not a significant source of P in the present study areas. Both TP and PP concentrations were positively correlated with pH, SS, BOD5, CODcr, TIC and TOC concentrations (p<0.01), while TDP was just positively correlated with BOD5, CODCr, TIC and TOC (p<0.01). In addition, regional environmental conditions and human activity intensity could also affect P concentrations in urban storm runoff.(4) The accumulative characteristics of TP, DP and PP in urban storm runoff were judged using the method of M (V) curve suggested by Geiger. The results showed that the first flush effect of TP and PP in the various runoff were frequent while that of DP was not. Among different rainfall events, the first flush effect of TP and PP were also frequent, and the strength of the first flush was medium or negligible; meanwhile the first flush effect of DP was infrequent, and medium or negligible dilution of DP was dominant. Additionally, the strength of the first flush of TP and PP in main road runoff was higher than those in others; on the contrary, the strength of the first flush of DP in roof and outlet runoff was the highest. The effects of underlying surface types and most rainfall characteristics on the strength of the first flush of P in runoff were not very significant, only the strength of the first flush of DP was positively correlated with rainfall duration at the level of0.05.(5) The EMCs of TP and PP in urban runoff samples frequently exceeded the threshold of Class V suggested by environmental quality standard for surface water, and excessive degree was quite high, the max of them were as high as six times of the threshold; whereas low EMCs for TDP was observed in all runoff samples, there was only one exception that DP EMC in outlet runoff was1.2times of the threshold. Regional environmental conditions affected EMCs of TP and DP significantly, while underlying surface type affected EMCs of TP and PP significantly. The EMCs of TP and PP were negatively correlated with maximum rainfall intensity (p<0.05), while the EMCs of TDP positively correlated with AWDP (p<0.05). On the other hand, there were no significant differences of P loads among various rainfall events, runoff types or study areas; while TP and PP loads positively correlated with precipitation (p<0.05), and DP load was positively correlated with both precipitation and mean rainfall intensity (p<0.01).(6) The annual TP emission fluxes from Jiushanwai River study areas were367.33kg, with about7%of DP and93%of PP; meanwhile, the annual TP emission fluxes from Shanxia River study areas were237.85kg, and DP made up17%while PP made up83%. Underlying surface type determined the TP and PP loadings in storm runoff, but regional environmental conditions affected the export of TDP more significantly.(7) The results of micro experiments about multilayer filtration system showed that, in considering the removal efficiencies of TP and TN, the system constructed from zeolite and medical stone with the combination of3:7was the best choice, the combination of5:5was the second, while the system with the combination of7:3could not meet the need of TP removal. The removal efficiencies of both TP and NH3-N of medium experimental models were more than90%; though removal efficiencies of TN and CODCr were not so high, their max still could be about50%; the concentrations of NO3--N and NO2--N in infiltrated water increased greatly comparing with their initial concentrations due to dissolution from materials and nitration reaction. In addition, although removal efficiencies of pollutants in stagnant water showed a rising trend with duration during the interval between water distribution events, the rate was negligible. On the other hand, sponge iron did not show a significant acceleration effect on removal efficiencies of pollutants; moreover it would harden and form a blockage that slowed the infiltration if it was buried shallower.(8) The layered filtration systems showed high removal efficiencies of pollutants with an exception of NO3--N, especially in the infiltration flow. The concentrations of pollutants except NO3--N in infiltrated water were all below the threshold of Class Ⅴ even Class Ⅲ suggested by environmental quality standard for surface water, which were also better than the quality of urban water in Wenzhou.(9) The comparison between two layered filtration systems (LFS) with different permeable pavement layers suggested that permeability of turf layer was much better than that of permeable brick, causing an approximate60min gap of runoff-yielding time between two systems and more infiltration under the same constraints. The difference of permeability also affected the comprehensive removal rates (CRRs) for pollutants of LFS significantly. The max CRRs for TP, NH3-N, NO2--N, TN and CODCr of LFS-A with turf layer were73.2,479.2,25.2,171.2and2372.4mg-m-2·h-1, respectively; while those of LFS-B with permeable brick layer were54.0,173.8,9.2,156.4and2525.2mg-m’2·h-1, respectively. Hence, turf is a better option than permeable brick as a choice of permeable surface for a layered filtration system. Additionally, some engineering measures, such as the utilization of replanted turf, may reduce the CRRs for pollutants for long periods of time, especially for a LFS with turf layer. The CRR of NO2--N of LFS-A was positively correlated to its initial concentrations at the level of0.05, while no other significant relationships between the CRRs of pollutants and their initial concentrations were observed due to their slight change.(10) The most important renovation measure for the riparian zone of Jiushanwai River is to replace the matrixes of the existing greening area, which will increase its infiltration rate significantly; meanwhile, enriching the vegetation types is also very important for the increasing of its ability to removal pollutants from urban storm runoff. On the other hand, in Shanxia River study area, besides replacing the matrixes of the existing greening area, its harden surface also should be substituted for vegetation pavement such as grass or dwarf shrub. However, because there was less quantity of runoff flow into the riparian zone of Shanxia River, the depth of replacement could be shallower comparing with that in Jiushanwai River study area. |
PDF Full Text Request