Valuation Of Ecosystem Services Based On Remote Sensing And Landscape Pattern Optimization In Beijing

Ecosystem services values assessment is the basis of the sustainability of ecosystems. It is of great significance to optimize the prediction of the ecosystem cover type's number and space distribution. This paper/dissertation needs to be (according to) based on the historical ecosystem's land cover conditions, in combination of the principle of the space suitability, to maximize the Ecosystem services values.In Beijing, the topography is relatively complicated. The ecological landscape is diverse, the population and the buildings are highly dense, and the industries are also quite concentrated. Unreasonable combinations of ecosystem types will bring the ecosystem enormous pressure; the social and economic development results in insufficient intensive use of the ecological environment resource, and particularly there are cases that people try to pursue economic benefits at the cost of eco-environmental benefits. All these problems make the conflict between the local people and eco-environment more pronounced.Hence, it is still of great significance to study the prediction and optimization of the ecosystem types and the landscape patterns, in order to change the current problematic value systems, to raise the people's awareness of environmental protection, and help them to reasonably use the resources intensively.Remote sensing technology has quite unparalleled advantages when compared with other observation methods, in terms of its spatial macro, wide viewing angle, multi-resolution (spectral and spatial), multi-time phase, regular periods, and rich information. So, the remote sensing technology can not only provide the information of ecosystem macro-space distribution, but also provide detailed local information and the dynamic information over time and space, thereby making itself one powerful tool to study the ecosystem.This study took maximized the ecosystem value as a goal to predicted and optimized the ecosystem types and the landscape patterns in future used the remote sensing technology to carry out the assessment of the Ecosystem services values based on the ecosystem type spatial distribution and ecosystem environment quality parameter in Beijing during 1978-2010. The major contents and results in this dissertation are as follows: (1) Analysis of the landscape pattern changes in ecosystem typesAfter analysis of the geographical features of Beijing combined with Level-II national classification system, this paper defines the Beijing's ecological value assessment in the following six types of ecosystem:farmland, forest, grassland, water area, city, and desert. This paper also employed the 16 views (scenes) Landsat remote sensing images with middle resolution as data source, in combination of the 166 field survey point data in 2010; the neural network based on characteristic vector method was used to extracted the ecosystem type's space distribution which was defined in 4 stages (periods) for the 32 years from 1978 to 2010.In the meantime, Used transition matrix and landscape index to analyzes the change patterns of the ecosystem types and the internal structure, as well as the space change patterns of the landscape in the past 32 years in Beijing, On the basis of this analysis, this paper concluded the following:during the period of 1978-2010 the area of arable land, desert decreased considerably, by 42.00% and 58.15% respectively; the forest area and city have increased substantially, by 35.87%,39.43% respectively; the grassland changed enormously and the water area maintains stable, but the arable land and desert normally have been changed into forest and residential use land. Analyzing from types and landscape index, this paper drew the following conclusions:the three ecosystem types which were forest, farmland, and city, cover the highest percentage of the landscape. The degree of landscape fragmentation and fractal dimension were less than that of the grass, waters and desert, and the concentration was higher. So it is the dominant landscape type. From the point view of landscape level, the patch density index, the average fractal dimension, as well as the falling down of the diversity index and the going up of the condensation index, have indicated that the research area has experienced a greater impact from the human's behaviors. This results in smaller and smaller landscape fragmentation, more and more regular landscape shapes, and also equilibrium of the landscape patches in the landscape as a whole.(2) Analysis of estimation and change in ecological environment quality parametersThe amount of soil erosion and NPP are the basis and necessary prerequisite for estimating the Ecosystem services values, also they are the important parameters to characterize the quality of the ecological environment. Using the study of the previous two chapters as the data source, which included its space meteorological, soil, terrain data and ecosystem land cover spatial pattern in different historical periods, this paper had calculated Beijing's amount of soil erosion and NPP's spatial and temporal distribution for a period of 32 years with the methods of Universal Soil Loss Equation (USLE) and light use efficiency model. Later, it carried out analysis of the spatial and temporal pattern change of these two ecologic environmental quality parameters and the influential factors, and concludes the following:in the past 32 years, the result showed that:the slightly soil eroded area covers 70% to 80% of all the area in Beijing. Specifically, mildly or slightly eroded area amounts to 90% of the whole erosion area, and in some areas of the western part and northeastern part, moderate and above erosion has occurred. Among all the ecosystem types, the soil loss ranks as follows:desert>forest>grassland>arable land. In addition, the erosion has shown an ascending trend from 1978 to 2010 peaking at 2000 but beginning to drop down subsequently. The soil loss in different slopes has shown such a pattern:the greater the slope, the more the soil loss, and in the curve of the overall erosion amount and the slope, it shows an uptrend trend first and then a downward trend. The NPP average and the total amount has decreased over these years, decreasing to 1348.09 gC/m2.a in 2010 from 1703.46 gC/m2.a in 1978, while the total amount down to 22.25 TgC/a in 2010 from 37.66 TgC/a in 1978. The NPP amount for different ecosystem types has shown such a pattern:forest> arable land>desert> grassland> city> water area. The affect NPP production important factors are precipitation, solar radiation, and temperature.The NPP has a positive correlation with precipitation and solar radiation, but negative correlation with the temperature.(3) The quantitative calculation and change analysis of Ecosystem services valuesConsidering the region ecological environment characteristics of Beijing and the availability and reliability of the data, this paper chose the Ecosystem services values evaluation index system suitable for remote sensing, which included six services functions as producing organics, nutrient recycling, reserved water, soil conservation, the pollution's absorption and decomposition, and air conditioning.Based on the study of chapter 4 as the Ecosystem services values indicators, including the soil erosion and NPP in Beijing in different historical periods as research basic data, For each service functions, We calculates the annual change of the service value per unit area for each item in the four periods of the 32 years from 1978 to 2010, and analyzes the Ecosystem services values's change patterns in different ecosystem types. From the in-depth study of the composition of the Ecosystem services values and their spatial and temporal patterns in Beijing:among the total values, the reserved water contributes the most to the service values, followed sequentially by the nutrient cycling, production of organic matter, soil conservation, air conditioning, and the pollution's absorption and decomposition. In the 32 years, the total Ecosystem services values have shown a descending trend:the greatest is 1425.48×108 Yuan/a in 1978, the least is 919.82×108 Yuan/a in 2010, a decline of 35.47%. As regards to the service value per unit area for each item, it shows a similar trend as the service total amount between 1978-2010:the greatest in 1978 and the least in 2010, ranging between 8.70 Yuan/m2-5.61 Yuan/m2, with a change of 35.45%.For the space distribution of the Ecosystem services values, it showed such a pattern:high in northern part but low in southern part, high in northwestern part but low in southeastern part, and high in mountainous areas but low in plain areas. The Ecosystem services values for different land use/cover types showed such a pattern in the descending rank:forest>arable land> city> desert> water> grassland, but the service value per unit area ranks as:arable land> forest>water area> grassland> desert> city.(4) the study of the prediction of the ecosystem's landscape pattern change and the gray optimizationAccording to the ecosystem type's landscape space distribution in different historical periods, used the method of CA-Markov for the prediction of 2020's landscape. Meanwhile, it used Markov to predict the transition probabilities, and the CA to realize the space transition simulation. The CA-Markov model can simulate the ecosystem types' landscape pattern change trend preferably in Beijing, but this trend might look unreasonable from the point of view of ecological benefits. In order to maximize the ecological system serve value benefits of Beijing, and make a reasonable structural composition of the various ecosystem types, this paper used a combination of gray system and cellular automata to realize the space simulation with the gray optimization theory. Drawing upon the gray linear programming algorithm, this paper set the maximization of the Ecosystem services values as the objective function, to optimize the area combination of the various ecosystem types; then we used the gray correlations to filter the assessment indexes of the space adaptability for various ecosystem types, and determines its weight respectively; then, it adopts the gray concentration to assess the space suitability and generate the space suitability rules for different ecosystem types. Finally, we input the area optimization solution and the space optimization solution into the CA model, to realize the landscape pattern optimizations of the ecosystem types in 2010 for Beijing. This optimization model can basically achieve the purposes of optimizing the ecosystem types in terms of number and the space distribution, and finally the set goal of the optimization of the ecosystem type's landscape patterns.The main features and innovations of this paper:1. It adopts dynamic assessment method of the spatial and temporal data based on the time sequence, instead of the past static assessment method of the per unit area amount to derive the total Ecosystem services values. The study used the time-series remote sensing images, soil, weather, and socio-economic benefits as the data sources, remote sensing technology, GIS space analysis, and computer programs were used. Chose the soil erosion and NPP as the parameters to assess the ecosystem environment quality, it dynamically estimated the per unit area Ecosystem services values in 32 years from 1978 to 2010 in Beijing, and analyzed the change of ecosystem environment quality, the composition and change patterns of the Ecosystem services values.2. This paper also coupled the gray theory and the cellular automata, to predict Beijing's ecosystem landscape patterns and optimize in terms of number and space. First, we used the gray linear programming algorithm, set the maximization of the Ecosystem services values as the objective function, and the ecosystem environment and the social economic data as the independent parameters, to optimize the area combination of the various ecosystem types. Then used the gray correlations to filter the assessment indexes of the space adaptability for various ecosystem types, determines its weight respectively, the gray concentration was used to assess the space suitability and estimates the space suitability rules for different ecosystem types. Finally, it used the number and space optimization solutions as the basic rules in the CA model to predict and optimize the ecosystem system landscape patterns in 2010 for Beijing.