| Artificial ecosystems, which are constructed on the basis of natural or semi-natural ecosystems, are manipulated or intervened differently by different human needs. Accounting for more than70%of the terrestrial ecosystems, artificial ecosystems have important impacts on the region and the world. Most artificial ecosystems are distributed in the urban areass, making their influences on regional and global ecosystems increasingly. Approaches to monitoring the distribution of artificial ecosystems in the urban areas are mainly distributed into two categories: automated or semi-automated and manual classification of aerial or satellite images. While the accuracy of the first approach is not high, which is about60-87%, and the cost of remote sensing images used for monitoring are overpriced. With the development of high-resolution remote sensing images in Google Earth, the method of manual interpretation has become an important way for extracting fine-scale artificial ecosystems in urban areas.Based on the analysis of the existing image in Google Earth for identifying, and according to the features of artificial ecosystems in the urban areas, we combined Google Earth aerial images with ArcGIS software to characterize the spatial distribution of existing artificial ecosystems in the urban areas at fine spatial scales and to create a dataset and sampling framework. Application study is carried out in several typical ecosystems including Greenhouse Agriculture, Green Spaces in Build-up Area and Woody Plants along Highway to discuss the effectiveness of manual extraction created in this study. The main content and results are as follows:(1) Identification system for artificial ecosystems in urban areas. According to the experience of other scholars on manual extraction of aerial images in Google Earth, an identification system, which utilizes high-resolution Google earth images with seasonal variation characteristics and ArcGIS software, is constructed.(2) The accuracy of visual interpretation in urban areas is very high, especially for artificial ecosystems in urban areas. Such as, the extraction accuracy of greenhouse is almost96%, the identification accuracy of parks in build-up area could also reach96%.(3) Since plastic greenhouses were first introduced into China in the late1970s, the total land area that they cover has increased by nearly100-fold in30years, making China become the country with the largest area of greenhouse agriculture, which accounting for95%of total area of the world. The Yangtze River Delta region is chosen, and the method of visual interpretation is applied to characterize the distribution of greenhouses in seven cities-31urban areas/county-level cities/counties and to analyze factors affecting greenhouses’distribution. The results showed that:firstly, the density of greenhouse horticulture changing with the distance to the city center in urban scale or above presented obvious nearly normal distribution in Yangtze River Delta. Secondly, areas of greenhouse horticulture in different scales varied hugely; even at the same scale, the areas of greenhouse horticulture were quite different. The reasons why this phenomenon occurred were that the demographic, economic and physical factors were quite different in urban areas/county-level city/county. The results suggest that the influences of demographic factors and economic factors are comparatively large. The research on greenhouses’distribution can provide support urban policy making.(4) The construction of highway has a very large negative impact on the surrounding ecological environment, while the vegetation along the road could reduce these effects through a variety of ways. As important ecosystem services, carbon storage and carbon sink provided by trees along highway were rarely analyzed from the perspective of the entire road network. Taking highways in Zhejiang Province as an example, we use visual interpretation of Google Earth images and stratified random sampling method to analyze the differences of trees’carbon storage and carbon sink in various land cover type along highway, and to map these results at the same time. The results showed that total carbon storage by trees adjacent to highway was149780.99t, equivalent to mean figure of2.40kg C m-2. Carbon sequestration by trees along highway was20162.28t C yr-1, corresponding to0.324kg m-2yr-1. The only significant difference in carbon density and net primary production per area was apparent between trees adjacent to forests and countries. Carbon storage of trees along different highways could vary widely, which is derived from the areas of trees along highways. Carbon density of trees along highway fell in between wild forests and urban greenspaces, however NPP per area was higher than them. Thus planning and management of forests adjacent to highway could be another method to promote CO2absorption second to the development of alternative energy sources and management of wild forests and urban forests.(5) Urban expansion and associated land cover change drives natural habitat loss, threatens biodiversity globally; while in the urban areas, the heterogeneity of urban landscape could create diverse habits, which are conducive to the protection of biodiversity. However, species homogenization of flora in an urban areas globally has reduced species diversity and hampers its function, thus there is an urgent need to develop strategies and principles to protect species of different sources and different groups. Heterogeneity in protection effect for species results from differences in human management and microenvironments, and thus making habit suitability and survival activity distinct for species in different green spaces in urban areas. Visual interpretation and stratified random sampling method are used in this paper to map urban park, to analyze microhabitats in different green spaces in build-up area, and to explore the possibility for biodiversity protection in build-up area. When the endangered plants introduced into build-up area, there is an need not only to consider the external conditions which is used for plant growth, but also some characteristics of endangered plants themselves, such as the pollination characteristics and the minimum spaces required for reproduction. The results showed that10kinds of microhabitats appeared in road green space, which were disturbed intensively. At least16kinds of microhabitats in park green space, which has some characteristics of natural microhabitats. The amount of microhabitats owned by commercial or residential green space was between them. Therefore, park green spaces could be used as priority areas for the protection of endangered plants. |
PDF Full Text Request