Research On Spatio-temporal Characteristics And Information Extraction Of Artemisia Scoparia In Desert Steppe Based On GF-1/WFV Time Series

Grassland ecosystem,as one of the most widely distributed ecosystem types in terrestrial ecosystems,has a succession of vegetation types under the influence of natural and human factors.Artemisia as a dominant species in the process of desertification to restore vegetation succession,a large number of scholars have carried out lots of researches around the ecological characteristics,it has important ecological significance,and the use of remote sensing method to extract Artemisia scoparia could further verify the ecological significance of spatiotemporal distribution.In recent years,remote sensing time data can accurately reflect the regularity of vegetation growth season,and provide strong support for grassland vegetation information identification and monitoring.Based on the data of WFV(GF-1 WFV),which is one of the high-grade satellites in China,and the ground observation and investigation data were used.On the basis of investigation on the community structure of different vegetation types in the study area,the temporal spectrum of Artemisia scoparia was analyzed by using the time series spectrum.The temporal distribution spectrum of Artemisia scoparia was established,and the spatial distribution of Artemisia scoparia was studied.The spatial distribution of Artemisia scoparia.was estimated by remote sensing.Biomass and cover height,and then analyze the changes of growth of Artemisia scoparia,so as to provide a new idea for the extraction of grassland communities.The results are as follows:[1]Through the long-term fixed monitoring of Artemisia scoparia in the study area,the ecological characteristics of the typical vegetation larvae were studied,and the important value of Artemisia scoparia was determined by the important value.The important value of Artemisia scoparia was the highest proportion in the study area,and it had an obvious competitive relationship with Stipa,and a companion relationship with Sophora alopecuroides.Yearly precipitation pattern can be divided from Statistics about precipitation data in research area from 1950 to 2016 combining continuous monitoring the relationship between rainfall and growth of Artemisia scoparia community from2012 to 2016,conclusion as the growth of Artemisia scoparia in this area has continuity and precipitation in spring and summer has significant correlation with the height and coverage of Artemisia scoparia in spring,summer and autumn,especially,the growth condition of Artemisia scoparia is closely related to precipitation from April to August.Meanwhile,different precipitation events in the study area had some influence on the germination and growth of Artemisia scoparia.[2]The canopy spectral characteristics of Artemisia scoparia,Stipa and Sophora alopecuroides were analyzed by observing the canopy spectra of different grassland types in the study area,and resampling the spectral according to the GF-1 WFV spectral response function.The results showed that the spectral curves could express the ecological characteristics of different types of vegetation at different growth stages and different growth conditions.The spectral curves of Stipa and Sophora alopecuroides were different from those of Artemisia scoparia.And also the spectral characteristics of Artemisia scoparia were different in different periods.In the different stages of the growth period of Artemisia scoparia,the reflectance of the spectral resampling results in the red and near infrared wavelengths was different with the increase of the coverage,and the vigorous period was more sensitive to the spectral reflection.The time series curve of NDVI can accurately express the growth curve of Sophora alopecuroides,Stipa and Artemisia scoparia,and the difference of the time series curves of different types of vegetation was significant.Therefore,the three vegetation in the time series of remote sensing images were separable.[3]Based on the analysis of the NDVI time series curve,the characteristics of the sequence of the GF-1 WFV NDVI were determined by the fixed monitoring sample.The NDVI sequence characteristic curve of the Artemisia scoparia was determined,and the remote sensing information extraction method based on the integrated time series spectrum was used to extract the distribution of Artemisia scoparia,The results showed that the distribution range of Artemisia scoparia was wide in the study area,and the different coverage of Artemisia scoparia was distributed in the same area.Due to the severe salinization in the northern part of the study area,the area of Artemisia scoparia was poor and the distribution was more intensive there.The distribution of Artemisia scoparia was relatively uniform in the middle and southern areas of the study area.The overall accuracy of extracting Artemisia scoparia by comprehensive time series was 96.7%,which indicated that the method was an effective method for extracting grassland information.And because the spectral similarity scale can measure the size and shape of the spectrum at the same time,it was helpful to reduce the influence of matter spectrum,and further improve the accuracy of remote sensing information extraction[4]Using the regression analysis method,a linear regression model(correlation coefficient R2 is 0.575,0.645)based on ground measured NDVI and grassland ground biomass and cover height product was established,and he aboveground biomass and cover height of Artemisia scoparia were estimated and extracted through the model of GF-1 WFV NDVI.At last,the temporal and spatial patterns of aboveground biomass and cover height of Artemisia scoparia were studied,and their distribution and dynamic changes in time and space were summarized.