Font Size: a A A

Analyzing The Dynamic Change Of Coastal Wetland And Simulating Future Change Scenariosin Shenzhen Based On Satellite Remote Sensing And Cellular Automata Model

Posted on:2018-12-09Degree:MasterType:Thesis
Country:ChinaCandidate:X L ChenFull Text:PDF
GTID:2310330518956971Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
In this paper,the coastalwetland in Shenzhen city is the research object,basing on the 5 periods Landsat TM remote sensing images in 1990,1995,2000,2005,2010 and 2015 years,using the decision tree method extract the forest land,urban construction land,farmland,mangrove,lakes,rivers,farms and coastal water areas.The change of wetland and non wetland types in Shenzhen city in 1990-2015 and the dynamic transformation of land types were studied.The driving forces of land use change in Shenzhen were analyzed based on meteorological data and economic development data.Referencing the principle of population gravitycenter,this paper analyzed the area gravity center offset of each land use type in Shenzhen.Logistic-CA and PSO-CA models were used to simulate the changes of wetland and non wetland in Shenzhen in 2015 and in 2030 years.(1)During the period from 1990 to 2015 years,in addition to the urban construction landarea and the lake area has increased,the other types of land use area have different degrees of reduction.The area of forest land in Shenzhen was reduced by 43719.8ha,and the area of urban construction land increased by 60697.12 ha.Farmland area decreased by 2783.16 ha,compared with 1990 years,the farmland area reduced about 32%.Mangroves decreased from 842.26 ha to 720.12 ha,which had a decrease of 14.5%.Lake wetland area increased from 4525.01 ha to 5321.45 ha.The area of river wetland is generally decreasing,which is reduced from 429.76 ha to350.29 ha.Farming area reached 9434.55 ha,inthe of year 2015,the area of the farms is less than 1400.9ha,compared with 1990,a decrease of 85.15%.The area of coastal waters is decreasing year by year.From 1990 to 2015,the area of the coastal waters of Shenzhen decreased from 94592.60 ha to 87837.80 ha,a decrease of 7.14%.(2)During the period from 1990 to 2015 years,the forest land,urban construction land and farmland in Shenzhen were mutual transformationmore than other types.Besides this,the forest land can convert to lake.Urban construction land and farmland are mainly transformed amongthis three types,and the main source of farmland supplement is the conversion of forest land.The mangrove was mainly convert tocity construction land and forest land,but mainly transformed to the city construction land,the increased areaof mangrove was mainly from farms and coastal water areas.Compared with other wetland types,it was mutual transformation less than other types.farming land and inshore tidal flats are more likely to become the mangrove.The lake is mainly transformed into forest land and urban construction land.The river is relatively stable,it was mutual transformation less than other types.While the increased area of river source were mainly from farms,the urban construction land,forest land and mangroves.Farms were mainly convert to fores tland,farmland and mangroves,and the complement of farms was mainly depended on forest land and the coastal water areas.Coastal water areas were mainly convert to the urban construction land,farming,forest land and mangroves.(3)During the period from 1990 to 2015 years,the driven analysis results of the dynamic change of wetland shows that only the areas of forest,mangrove and lakes had higher linear correlation with precipitation,R2 reached 0.8724,0.9079 and 0.5825,and they had no significant linear correlation with sunshine duration and average temperature.According to the analysis results of linear relationship between the land use types and economic indexes,it showed thatthe city construction land,farmland,forest,farm,coastal water areashave a high linear correlation with the end of the resident population,the second industry,the third industry,per capita GDP,the real estate development project investment and total volume of foreign trade.And they had no significant linear correlation withthe first industry gross domestic product.The linear correlation between these indexes and lake or river are not significant.(4)Using Logistic-CA and PSO-CA model simulate the land use situation of2015 and 2030 years in Shenzhen,the simulation results in 2015 show that the simulation error are less than 50% with two methods,overall,the Logistic_CA simulated results had better effect than the PSO_CA simulated results.Compared with the simulation result in 2015,The simulated results in 2030 showed that only farms area has increased in all wetland types.The farms area had increased by 26.56%,But the area of mangroves,lakes,rivers and coastal waters have different degrees of reduction,the range of reduction were respectively 44.33%,7.06%,66.62% and1.26%.In the non wetland type,only the urban construction land area was increasing,the increase range is 13.8%,the area of forest land and farmland are reduced,and the reduction rate is about 11.54% and 27.01%.Study on the dynamic change of land use types show that the non wetland type convert is quite obvious,there are certain transformation among wetland types,but mainly convert to the non wetland type,especially they transferedto city construction land.The percentages that convert to urban construction land for mangrove,lakes,rivers,farms and coastal water areas wererespectively 61.03%,11.87%,70.85%,18.84% and 1.03%.
Keywords/Search Tags:Shenzhen, remote sensing image, Logistic-CA and PSO-CA model, simulation
PDF Full Text Request
Related items