Font Size: a A A

Remote Sensing Analysis Of Spartina Alterniflora In The Coastal Areas Of China During 1990 To 2015

Posted on:2019-08-19Degree:DoctorType:Dissertation
Country:ChinaCandidate:M Y LiuFull Text:PDF
GTID:1360330569980931Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
Biological invasions have become important components of global change,challenging ecosystem health and safety.In particular,plant invasions in the coastal zone are widespread and have posed significant threats to coastal wetlands.Spartina alterniflora(S.alterniflora)has been one of the most widespread invasive plants in China and has caused serious ecological consequences in some places.Improved understanding of the distribution and landscape pattern of S.alterniflora is critical to effective ecosystem management and respond to further invasion.In this paper,based on Landsat imagery,spatial distribution and landscape pattern of S.alterniflora in the coastal areas of China were studied from 1990 to 2015.Object-based image analysis(OBIA)combined with the support vector machine(SVM)was applied to extract S.alterniflora,so that the classification accuracy and efficiency would be improved.Based on this results,we analyzed the dynamics and landscape structure changes of S.alterniflora at national and provincial scales,which contribute to better understanding the expansion process of S.alterniflora and making sustainable management strategies.The conversion between S.alterniflora and other land cover types were further achieved to confirm the driving factors of S.alterniflora.Quantifying the impacts of environmental factors on the distribution of S.alterniflora is vital for prediction the potential areas.In addition,the spatio-temporal patterns of S.alterniflora in the national nature reserves were also analyzed to mornitoring the invasion status and process.These results would provide scientific bases for effective control and reasonable use of S.alterniflora.The results show that:(1)Object-based image analysis(OBIA)combined with the support vector machine(SVM)is an effective method for extracting S.alterniflora by using spectral,textural and shape features.Fuzzy-based segmentation parameter(FbSP)optimizer can improve the accuracy of segmentation results and increase the efficiency of determine the optimal parameters.(2)The total areas of S.alterniflora in 1990,2000,2010,and 2015 are 4375.5 hm~2,25648.0 hm~2,43061.1 hm~2,and 54579.7 hm~2.The distribution continued to extend along the mainland coastline of China but the annual change rates gradually decreased.To 2015,S.alterniflora can be observed in extensive regions from Hebei to Guangxi Provinces,nearly 92%of this species occurred in Jiangsu,Shanghai,Zhejiang and Fujian Provinces.(3)The total area of S.alterniflora in each province overall has an increased trend from 1990 to 2015.But there were obvious differences among these provinces.For the whole study period,Jiangsu Province gained the largest areas of S.alterniflora expansion followed by Zhejiang,Shanghai and Fujian Provinces,the increased areas were more than 9000 hm~2.The annual change rate of S.alterniflora expansion generally decreased in the southern provinces of study area,such as Fujian,Jiangsu,Shanghai and Zhejiang Provinces;the annual change rate increased first and then decreased in Shangdong and Hebei Provinces;Whereas Guangdong Province and Tianjin holds the characteristic of fluctuation.(4)The amount of national nature reserves(NNRs)invaded by S.alterniflora gradually increased,by 2015,there were seven NNRs were observed S.alterniflora invasion.The total invasion area in these NNRs gradually increased from 315.4 hm~2 to17436.4 hm~2 during 1990~2000.The invasion areas at different protection levels increased as well,and the experimental zone exhibited the largest areas of S.alterniflora in all periods.Yancheng wetland national nature reserve had the largest areas of S.alterniflora among these invades NNRs in all these study years.(5)According to the conversion between S.alterniflora and other land cover types in China from 1990~2015,the driving factors of S.alterniflora varied among different provinces.The expansion of S.alterniflora were mainly driven by intentional human introductions and its biological characteristics,whilst the shrink of S.alterniflora was mainly caused by aquaculture ponds expansion,land reclamations and S.alterniflora managements.During the study period,the annual expansion rates and annual shriking rates all increased,and the increased areas of S.alterniflora mainly transferred from intertidal mudflats.(6)Temperture and precipitation play an important role in the distribution of S.alterniflora so that this species mainly concertrated in the central coast,while small areal extent distributed in the northern and southern provinces.Water depth and river density impact the local distributions of S.alterniflora,with the water depth goes on,the distribution probability of S.alterniflora decreased,water depth of-2~0 meter is the most suitable condition.Whereas with the river density increased,the distribution probability increased first and then decreased.The river estuaries and bays are more suitable than other places in the coastal zones.S.alterniflora grown in the central coast exhibited higher plant height,lower density and more aboveground biomass,while the opposite charactoristics were shown in the northern and southern coastal provinces.
Keywords/Search Tags:Spartina alterniflora, invasive plant, Landsat, national nature reserve, Object-based image analysis, support vector machine
PDF Full Text Request
Related items