Characteristics Of Greenhouse Gases Emission From Urban Rivers At Different Time Scales

Urban rivers have been identified as hotspots of the potent greenhouse gases(GHGs),including carbon dioxide(CO₂),methane(CH₄)and nitrous oxide(N₂O)due to the influence of high intensity human activities.Urban rivers can not only accept the direct input of high concentrations of GHGs from sewage,but also indirectly affect various biogeochemical cycles through the altered biotic and abiotic structures in the system,and lead to changes in riverine GHGs fluxes.Although many previous studies have made great contributions to the accurate estimation of regional/global GHGs emissions by improving the spatial resolution,the temporal variability of these fluxes,regardless of long-term scales(eg.interannual and seasonal variability)or short-term scales(eg.diurnal variation),are still with many uncertainties,which will undoubtedly cause biases in GHGs fluxes estimation.In addition,the proportion of GHGs reaching to the atmosphere mainly depends on the pathways of gas emissions from sediments and their spatiotemporal characteristics.Although studies have shown that bubbles(ebullition)of CH₄ makes up a large proportion of total CH₄,almost all studies quantifying CH₄ fluxes have focused on the diffusive transport,which may lead to the total CH₄ emissions being greatly underestimated,and ultimately affects the understanding of the role of inland waters in ecosystem carbon cycle.In addition,ebullition is most likely to be underestimated because it is episodic and not representatively captured by the usual short-term measurements.In order to more accurately estimate the GHGs emission flux of urban rivers and correctly understand the role of urban rivers in the carbon and nitrogen cycle of aquatic ecosystems,this study have carried out long-term and high-frequency field monitoring and laboratory analysis on different urban rivers in Shanghai,a typical megacity in China.This study based on the project of National Science Foundation of China"Nitrous oxide emissions coefficient of rivers in urbanized areas and the modeling of watershed scale”.This study used several monitoring methods to explore the contribution of ebullition to CH₄ emissions across different urbanization areas as well as in time,and to evaluate the deviations of different monitoring methods for GHGs emissions estimation.By analyzing the temporal and spatial variation characteristics of GHGs concentrations,saturations and fluxes in urban river waters at different time scales(including interannual,seasonal and diurnal variations),the impact of global climate change and human activities on GHGs emissions from river waters is explored.On the basis of a amounts of measured data,this study supposed:(1)to comparatively investigate the highly spatiotemporal characteristics of ebullition events and to explore how the environmental factors affect the production and emissions of CH₄ from urban waters by using three different methods,including boundary layer model,the floating chamber and the gas trap;(2)to explore the diurnal variation characteristics of GHGs fluxes in river waters and analyze their control factors through high-frequency and long-term field monitoring;(3)to evaluate the errors of different monitoring methods in estimating GHGs flux in water bodies,and to propose corresponding field monitoring plans based on the differences in the temporal and spatial variation of fluxes;(4)to establish a long-term river waters GHGs fluxes database and analyze the temporal differences between interannual and seasonal scales,and explore the characteristics of the spatiotemporal distribution of GHGs fluxes in areas with different levels of urbanization.The main results and conclusions of this study are as follows:(1)This study demonstrated that eutrophic rivers have strong ebullition activities,which is characterized by instantaneous randomness and high spatiotemporal heterogeneity.Ebullition rates showed large differences in short time and small scales,which also emphasizes the importance of high spatiotemporal resolution monitoring for estimating ebullitive flux.In addition,the concentrations of the three GHGs in the bubbles barely change within the seasons,with CH₄ having the highest concentration(66.3%)among the three gases,followed by CO₂(0.67%)and N₂O(0.78 ppm).If only the contributions of diffusion and ebullition transport are considered,the results showed that bubble transport has negligible contributions to N₂O and CO₂ fluxes,but plays an important dominant role in CH₄ fluxes.The contribution of bubble activity to CH₄ flux varies greatly among different rivers.The average contribution of ebullition is the highest in urban rivers(86.05%)and the lowest in suburban rivers(33.66%),which is consistent with the degree of eutrophication of rivers.The results of the study clarified the importance of bubble transport to CH₄ emissions in urban waters.In the future,the bubble CH₄ flux should be considered in the total CH₄ flux in river waters to improve the accuracy of GHGs flux estimation.(2)In this study,the boundary layer model and the floating chamber method were used to monitor the diurnal flux in a typical eutrophic river for 13 sampling seasons.The results showed that the three types of GHGs fluxes were generally higher in the daytime than that in the nighttime.Among them,the CO₂ flux measured by floating chamber might be affected by the enclosure time and is generally smaller than the CO₂flux measured by the boundary layer model,but the floating chamber is easier to monitor the short-term and the diurnal CO₂ flux changes.Combined with the diurnal variation of CO₂,this study shows that the CO₂ flux monitored around at 9 a.m.each day using the boundary layer model can best represent the daily average flux.The main factors controlling the diurnal variation of the river CO₂ flux in this study were dissolved oxygen(DO)concentration and water temperature.With the increase of light and temperature during the day,photosynthesis in the water promoted CO₂ absorption and oxygen production.In this study,the primary productivity of the water body had little effect on the diurnal CO₂ flux.(3)Different methods have a great influence on the estimation of CH₄ flux in river water.In particular,the boundary layer model can only monitor the diffusive CH₄ flux,resulting in a large underestimation of the total CH₄ flux.Comparing the bubble CH₄flux measured by the floating chamber and the gas trap shows that the total CH₄ flux monitored by the floating chamber will be underestimated by?60%.In addition,the bubble CH₄ flux showed obvious temporal heterogeneity on both the diurnal and seasonal scales,but the fluctuation of the diffusive CH₄ flux on the long/short time scale was very small.Water temperature and DO concentration are the key factors to control the CH₄ flux in water bubbles.High temperature and hypoxic environment greatly promote methanogenesis and CH₄ production,and the synergistic effect of these two factors especially promote bubble CH₄ emission.Considering the high variability of bubble activity on the diurnal scale,it is necessary to increase the time of continuous monitoring to obtain a more accurate estimate of total CH₄.(4)There is a good correlation between the N₂O fluxes measured by the boundary layer model and the floating chamber,but the floating chamber may still be affected by the time of the enclosure,which is slightly lower than the N₂O flux measured by the boundary layer model.Similar to the diurnal variation patterns of the first two gases,the N₂O fluxes also showed a daytime average higher than the nighttime average but showed more complex variation patterns in some sampling seasons.In order to reduce the follow-up monitoring deviation on the diurnal scale and improve the monitoring accuracy,this study believes that the N₂O fluxes monitored by both methods around 9a.m.each day can best represent the daily average flux.Even though the diurnal variation of various nitrogen concentrations in this study was small,they still played an important role in the N₂O production process.Among them,the day and night N₂O flux is inversely proportional to the nitrate(NO₃^-)concentration and is directly proportional to the nitrite(NO₂^-)and ammonium(NH₄⁺)concentrations.The results show that the N₂O might be mainly produced from the nitrification process at daytime,while might be mainly from the denitrification process at night.(5)In this study,after long time monitoring the CO₂ concentration,saturation and flux of river water in different urban areas,it is proved that urban rivers are an important source of atmospheric CO₂ emission.The annual average CO₂ flux is 141.61±104.18mmol m^-2 d^-1.The watershed with a large amount of exogenous carbon and nutrient input represents as a stronger source of CO₂ emission,while the oligotrophic river with a lack of exogenous input has the lowest CO₂ flux,and the CO₂ emission capacity of the water body depends to some extent on respiration and primary productivity.In addition,the CO₂ fluxes of different urban river water bodies showed significant seasonal and interannual differences,emphasizing the importance of high temporal and spatial resolution field monitoring to achieve higher-precision river CO₂ estimation.Because the rivers in urban areas receive many anthropogenic pollutions,the primary productivity of the water body has a weak effect on the seasonal variation of the CO₂flux in the water body.Temperature might be the main factor causing seasonal differences in CO₂fluxes in urban river waters.The high temperature in summer promotes respiration and lead to the maximum CO₂ emissions(accounting for 38%of the annual flux).In addition,the interannual variation patterns indicated that overall water quality management in the watershed had a significant impact on river CO₂fluxes.The CO₂ emissions from river water decreased from 282.45±44.69 mmol m^-2 d^-1 in2012-2013 to 123.62±30.97 mmol m^-2 d^-1 in 2020-2021,indicating that with the decrease of nutrient input and water eutrophication with the improvement of chemical conditions,the CO₂ emissions of urban rivers generally showed a downward trend.(6)After monitoring the CH₄ concentration,saturation and flux of river water in different urban areas for a long time,this study proved that urban rivers are an important source of atmospheric CH₄ emission,and the annual average diffuse CH₄ flux is0.64±0.76 mmol m^-2 d^-1.However,since most field monitoring activities only consider diffusive CH₄ fluxes,the contribution of ebullitive CH₄ fluxes is often neglected,which may lead to greatly underestimated total CH₄ emission in urban river waters.If the contribution of bubble transport to river water is considered,the estimated annual average total CH₄ flux increased by more than 3 times(2.01±2.59 mmol m^-2 d^-1).The spatial and temporal distribution characteristics of different urban rivers showed that the nutrient concentration and DO level in the water were the key factors in determining the total CH₄ flux.Anaerobic methanogenesis processes are more likely to occur in hypoxic waters with sufficient nutrient supply,resulting in more CH₄ production and emissions.In addition,temperature has a significant promoting effect on CH₄production and emission,especially on ebullitive CH₄ flux.The total CH₄ flux in the river water in this study showed an overall downward trend from the year of 2011 to2021,and this result has a great relationship with the greatly improved eutrophication in the urban waters.When the concentration of DO in the water increases and the nutrient content decreases,these conditions are not conducive to the methanogenesis process of sediments,resulting in the reduction of CH₄ emissions in the water.(7)After monitoring the N₂O concentration,saturation and flux of river waters in different urban areas for a long time,it is proved that urban rivers are an important source of atmospheric N₂O emission,and the annual average diffused N₂O flux is87.89±74.40 mmol m^-2 d^-1.The DO concentration and nitrogen loading conditions and their components in river waters are the key factors to control N₂O emissions.From the year of 2011 to 2021,the nitrogen concentration in the river water decreased,but DO concentration in the water body increased year by year,and the N₂O flux in the water showed a downward trend.In addition,the seasonal variation of N₂O flux in urban water was not significantly correlated with temperature,and the stimulating effect of high temperature on microbial processes may be masked by the promoting effect of nitrogen concentration on N₂O production.This conclusion further proves the critical influence of eutrophication conditions(especially DO concentration and nitrogen load)on N₂O flux in urban river water.