Biomass Analysis And Model Prediction Of Typical Shrub Communities In Desert Steppe Of Wulate Middle Banner

Desert steppe is the most arid type in grassland and the buffer zone between desert and typical steppe.In this paper,several typical shrub communities in the desert steppe of Urad Middle Banner,Inner Mongolia,were studied.Based on the investigation of community types,species composition and habitat conditions,plant height,plant number,crown width and coverage of typical shrubs were determined.Sample harvesting method was used to determine typical shrubs.The above-ground biomass of shrubs was analyzed,and the relationship between above-ground biomass of shrubs and soil factors was analyzed.Eight models for estimating above-ground biomass of typical shrubs were established based on canopy area and volume.The main results are as follows:(1)The community structure of desert steppe in Wulatezhong Banner is simple,species composition is relatively single,mainly xerophytic and strongly xerophytic shrubs?semi-shrubs?small half shrubs and tufted herbs.The dominant species and constructive species of shrub community are mainly as follows:Cciragaria stenophylla?Caragana tibetica?Salsola passerina?Caragana opulens?Ceratoides latens and Anabasis brevifolia is the mainly.(2)The aboveground biomass of shrub communities ranged from 21.31 g/m2 to 257.74 g/m2,with an average of 71.32 g/m2.Among them,the aboveground biomass of S.xanthoxylon was the largest,while that of the C.latens community was the smallest.The above-ground dry-fresh ratio of shrubs ranged from 30.83%to 66.03%,with an average of 49.90%.The average value of A.brevifolia was the smallest,while that of C.tibetica was the largest.(3)There was a significant positive correlation between aboveground biomass of shrub vegetation and soil water content in the desert steppe of Wulatezhong Banner(R2=0.9176,P<0.01),the aboveground biomass of shrub vegetation was positively correlated with total soil porosity(R2=0.8964,P<0.01),the relationship between aboveground biomass and soil bulk density was not significant(R2=0.0664,P>0.05).(4)The best variables for predicting aboveground biomass of C.stenophylla,C.latens and R.songcirica were crown area(P<0.01,correlation coefficient greater than 0.80);the best variables for predicting aboveground biomass of S.xanthoxylon,A.brevifolia,C.opulens,S.passerina and N.sibirica were crown volume(P<0.01),and the correlation coefficients were above 0.80 except S.xanthoxylon and A.brevifolia.(5)The power function model is ideal for forecasting aboveground biomass of R.songcirica,S.passerina and N.sibirica.Quadratic function regression model is an ideal model for forecasting aboveground biomass of S.xanthoxylon,A.brevifolia,C.stenophylla,C.opulens and C.latens.Through the test of measured values,it is proved that it has good prediction accuracy.