Estimation Of Riverine Organic Carbon Flux Based On Remote Sensing And In-situ Data

River connects the two largest carbon reservoirs on the earth,the terrestrial and marine ecosystems.On the global,rivers transport about 219.50±54.36 Tg C/yr(1 Tg =1×1012 g)of dissolved organic carbon(DOC)and 204.00±21.73 Tg C/yr of particulate organic carbon(POC)into marginal seas annually.The lateral transport of riverine organic carbon not only impacts global carbon balance,but also reflects natural processes and human activities in river watershed.Moreover,riverine organic carbon can have impacts on ecological environment characteristics of estuary and marginal sea.Under the influences of precipitation,air temperature,land use change,human activities,etc.in watershed,transports of riverine organic carbon are tempo-spatially varied.Under the influences of natural processes and human activities,transport of riverine organic carbon is featured by highly tempo-spatial variations,which make it difficult to estimate riverine organic carbon flux.So far,most of the reported studies about riverine organic carbon transport focused on non-monsoon rivers in the Europe and America,for which there are long-time sequenced in-situ DOC and POC data.For Asian rivers under the influence of East Asian Monsoon,however,there are few reports about organic carbon transport,especially for spatial variations of different rivers and temporal variations of a specific river.Based on these,this study focused on spatial variations of organic carbon flux from mainland Chinese rivers.Moreover,taking the Changjiang River as an important study area,we synchronously measured optical and chemical in-situ data at Datong hydro metric station section(monthly)of the Changjiang River and its estuary(seasonally),estimated long-time sequenced DOC and POC flux,and discussed the impacts of basin characteristics on tempo-spatial variations of riverine organic carbon transport.This study laid a foundation for estimation of organic carbon flux from global oceanic rivers.Main contents and conclusions of this study are shown as below.(1)Spatial variations of organic carbon transport in mainland Chinese rivers.Based on data of riverine organic carbon(DOC and POC)and basin characteristics,we built estimation models of annual mean DOC and POC concentrations in mainland Chinese rivers respectively.By combining annual water discharge,we further calculated annual DOC and POC fluxes from mainland Chinese rivers in 2010.In 2010,mainland Chinese rivers transported about 5.05 Tg C of DOC and 7.00 Tg C of POC.Among them,about 34.65%of DOC and 42.43%of POC were transported by the Changjiang River.In the spatial distribution,precipitation decreased riverine DOC concentration and POC(%TSM)(POC content in total suspended matter(TSM)),but DOC and POC exports from mainland Chinese river watersheds are spatially consistent with precipitation,both with higher values in the southeast.Moreover,air temperature,watershed cropland area,watershed slope,etc.can also impact the spatial distributions of riverine organic carbon transport in mainland Chinese rivers.(2)Estimation of DOC flux in the Changjiang River(Datong section).Taking river discharge,cropland area ratio and mean gross primary productivity in watershed as independent variables,we built a multilayer back-propagation neural network model(MBPNN)to estimate monthly mean DOC concentration at Datong section.By combining monthly water discharge at Datong,we calculated monthly DOC fluxes from 2000 to 2013.Monthly DOC flux ranged from 3.88×104 t C to 42.78×104 t C,with mean of 13.49±6.73×104 t C/month.We found that DOC flux at Datong was determined by water discharge.There was significantly linear relationship between monthly DOC flux and water discharge(R=0.8914,P<0.0001).(3)Estimation of POC flux in the Changjiang River(Datong hydrometric station section).Based on simulated data from a bio-optical model and photo-chemically synchronous in-situ data,we built a POC remote sensing algorithm which can be applied to Landsat satellite data(with spatial resolution of 30 m).At Datong section,POC concentration was linearly related to TSM concentration,and with obviously spatial variations.By combining monthly water discharge at Datong,using data of thematic mapper onboard Landsat-5 satellite(TM/Landsat-5)and enhanced thematic mapper onboard Landsat-7 satellite(ETM+/Landsat-7)we calculated monthly POC fluxes at Datong from 2000 to 2016.Monthly POC flux ranged from 1.4×104 t C to 52.04×104 t C,with mean of 13.04±7.17×104 t C/month.We found that there was significantly linear relationship between monthly DOC flux and water discharge(R=0.7143,P<0.05).At Datong hydrometric station of the Changjiang River,monthly POC flux was almost equal to monthly DOC flux.The Changjiang River was not featured by high POC transport.(4)Estimation of POC flux in the Changjiang River Estuary(Xuliujing hydrometric station section).Based on photo-chemically synchronous in-situ data,we built POC remote sensing algorithms which can be applied to satellite data of Geostationary Ocean Color Imager(GOCI)(with spatial resolution of 500 m).In the Changjiang River Estuary,POC concentration was linearly related to TSM concentration.At Xuliujing section,current velocity linearly decreased with the increasing of water depth;however,POC concentration exponential increased with the increasing of water depth.By combining monthly water discharge at Datong,using GOCI data we calculated monthly POC fluxes at Xuliujing.From May 2015 to April 2016,mean monthly POC flux was 7.14×104t C/month at Xuliujing section,which was 44.58%higher in average than in-situ values at Datong section.We found that tide,wind,etc.in the Changjiang River Estuary could increase POC flux from the Changjiang River to marginal sea.(5)Temporal variations of riverine organic carbon transport in the Changjiang River(Datong section)and impacts of basin characteristics.Based on long-time sequenced monthly DOC and POC fluxes,we discussed temporal variations of organic carbon transport in the Changjiang River(Datong section).Due to dilution effects of precipitation water,DOC concentration was usually lower in the wet season(from May to October)than that in the dry season.Mean DOC concentration in the dry season is 1.14 times of that in the wet season.Due to flushing effects of precipitation water,however,POC concentration was usually higher in the wet season than that in the dry season.Under the influence of the three gorges dam(TGD),temporal variations of POC concentration was gradually weakened.From 2000 to 2013,both DOC concentration and DOC flux at Datong did not changed significantly.However,both POC concentration and POC flux at Datong decreased exponentially significantly from 2000 to 2016.To summary,this study estimated total organic carbon fluxes from mainland Chinese rivers and long-time sequenced DOC and POC fluxes from the Changjiang River,and discussed tempo-spatial variations of organic carbon transport in mainland Chinese rivers.However,transformation,decomposition,and deposition,etc.may happen to riverine organic carbon in the estuary,so it is necessary to further study effective transport of riverine organic carbon flux into marginal oceans in the future.