Spatial Distribution Research And Vegetation Biomass Of Wetland In Yellow River Delta Natural Conservation Area

Wetland is a kind of important ecosystems on the earth, can not only provide rich resources,as well as regulating climate, water conservation, biodiversity conservation, etc.As the primaryproducers in the wetland,vegetation plays an irreplaceable role in maintaining the balance of thewhole wetland. As an important basic data in the wetland ecological system, vegetation biomasscan reflect the growth and distribution of vegetation, evaluat the health of the wetlandecosystem and has an important significance to the study of wetland ecological environment.The Yellow River delta nature reserve was selected as the research area. Data from the fieldmeasured wetland vegetation aboveground biomass data, Landsat-8image and the soildata(Organic matter, nitrogen, phosphorus and potassium, water soluble salt, pH, moisturecontent).The field survey sampling points is43and each sample’s size is1×1m.By analysing thecorrelation between the various remote sensing factors raster calculator tool and the measuredvegetation biomass(NDVI,DVI,RVI,Band1,Band2,Band3,Band4,Band5,Band6),we can selectsome remote sensing factors which has larger correlation with vegetation biomass factor.Establish linear and nonlinear regression model between these factors and the vegetationaboveground biomass in marsh vegetation and meadow. By comparison and analysis theR2,select a regression model which has the largest discriminant coefficient.This good model isused to calculate the aboveground biomass of marsh vegetation and meadow vegetation andgenerat the corresponding spatial distribution of vegetation biomass by raster calculator tool intoArcgis10.1and analyze the spatial distribution of vegetation biomass.The main conclusions are as follows:1, Analysing the relationship between the wet weight of aboveground biomass and extractedremote sensing factors(NDVI,DVI,RVI,Band1,Band2,Band3,Band4,Band5,Band6) and the rela-tionship between the dry weight of total aboveground biomass and extracted remote sensingfactors in swamp and meadow vegetation,it shows the latter has higher precision than former.There are NDVI,DVI,RVI,Band4,Band5which have the larger correlation with the vegetationbiomass. 2, Using the factors extracted from the Landsat-8image as the independent variables and the dryweight of the biomass in swamp and meadow vegetation as the dependent variable,we can findthe coefficient of determination(R2=0.546) is maximum when DVI is independent variables in ayuan linear regression model and the coefficient of determination(R2=0.566) is maximum whenDVI is independent variables in a yuan nonlinear regression model and the coefficient ofdetermination(R2=0.691) is maximum when eight factors (NDVI, DVI, RVI, Band1, Band2,Band3, Band4and Band5)are the independent variables in multiple linear regression models.When the fitting value is compared with the measured values in the model which has8factors,the coefficient of average residual error is14.47%, less than15%, smally, and the fittingprecision is higher.3, Through the calculation in the Aicgis, low value area (0-502g/m2) covers624.45km2,accounting for79.81%of the whole area of swamp and Median area (502-1064g/m2) covers anarea of152.55km2, accounting for19.5%of the whole area of swamp and meadow vegetation,and high value area(1064-2698g/m2) area covers5.48km2, accounting for0.7%of the whole areaof swamp and meadow vegetation.The areas of high value are more on the edge of someartificial farmland,woodland,grassland,influenced by human activities and accumulation ofvegetation organic matter is rich,which is conducive to the growth of vegetation.But The areas oflow value are the poor moisture condition,high soil salinity tidal flats, wasteland, etc.4, By the dry weight of biomass spatial distribution maps,it shows that dry weight of vegetationbiomass is reductive from land to sea and from the Yellow River way to the both sides.Throughthe analysis of the advantage cultivation such as reed and the alkaline,water depth has the mostimpact to dry weight of the reed biomass,the correlation coefficient R=0.782,and soil moisturecontent has the greatest effect to the dry weight of alkaline biomass in the various environmentalfactors. the correlation coefficient R=-0.314.In addition, vegetation species diversity have impactto the dry weight of biomass and the aromatic index R value is maximum0.449.There are twocauses of the trend of distribution.The frist one,when soil salt increase from the land to the sea,biodiversity decrease and the individual plant becomes short.The second one, it is the change ofgroundwater in the Yellow River on both sides of the conditions, the farther offshore, recharge ofgroundwater by the Yellow River is less,which leads to the reduction of the vegetation biomassfrom the river to both sides. The water and salt are the dominant factors to the current spacedistribution of dry weight of advantage vegetation biomass.