This paper presents the results of bias corrections of Chinese standard precipitation gauge (CSPG) measurements for wind-induced undercatch, trace amount of precipitation and wetting loss. Long-term daily data of precipitation, temperature, and wind speed during 1951-2004 at 726 meteorological stations in China were used for this analysis. It is found that wind-induced gauge undercatch is the greatest error in most regions, and wetting loss and trace amount of precipitation are important in the low precipitation regions in Northwest China. Monthly correction factors (corrected/measured precipitation) differ by location and by type of precipitation. Considerable inter-annual variation of the corrections exists in China due to the fluctuations of wind speed and frequency of precipitation. More importantly, annual precipitation has been increased by 8 to 740 mm with an overall mean of 125 mm at the 726 stations over China due to the bias corrections for the study period. This corresponds to 5% -72% increases (overall mean of 18% at the 726 stations over China) in gauge-measured yearly total precipitation over China. This important finding clearly suggests that annual precipitation in China is much higher than previously reported. The results of this study will be useful to hydrological and climatic studies in China.

Annual distribution characteristic of river sediment concentration is closely related to climate change and anthropogenic activities and directly affected by the process of hydraulic resource development in the Lancang-Mekong River basin. Using monthly suspended sediment concentrations in 1987-2003 at the three hydrological stations, Jiuzhou, Yunjinghong and Chiang Saen stations, on the mainstream of Lancang-Mekong River, the uneven distributions coefficient, the concentrated rate and period, and the range of variation of annual distribution characteristics of sediment concentrations at the three stations were studied and then the responses of these characteristics to the progress of upstream dam constructions were analyzed. The research results show that: (1) Annual distribution of sediment concentration at Jiuzhou hydrological station agrees with the variation trend of local climate changes, its uneven distribution coefficients (AUDC) increased at the studied period and maintained the characteristics of natural river; AUDCs at Yunjinghong and Chiang Saen stations seemed to be affected by anthropological activities such as dam constructions rather than climate changes, however, sediments of two downstream stations responded differently to the dam constructions. AUDC at Yunjinghong station decreased from the periods of 1987-1992 to 1993-1996, then abruptly increased during 1997-2003; while AUDC at Chiang Saen took on a trend of moderate decreasing in the whole period. (2) Rates and distribution periods of sediment concentrations at Jiuzhou station almost kept stable during the studied period; but those at Yunjinghong and Chiang Saen changed evidently, their distribution periods delayed asynchronously during the dam construction processes (1987-1992, 1997-2003), in the period of Manwan dam construction, distribution periods at Yunjinghong station delayed 5-6 days, while Chiang Saen delayed almost half a month, and in the Dachaoshan dam construction, delayed period at Yunjinghong station was longer than that at Chiang Saen. (3) Relative and absolute ranges of sediment variations at Jiuzhou station both increased against the time phases; those at Yunjinghong decreased, while relative ranges of sediment variation at Chiang Saen decreased in the earlier years then increased later, but absolute range was decreasing all the time. Different responses of annual distributions of sediment concentration at three hydrological stations to dam constructions indicate that their driving factors of sediment variation are absolutely different. These discoveries of annual distribution characteristics of sediment concentrations present new scientific evidences for assessing the impacts of dam constructions on upstream and downstream sediment change as well as its transboundary effects.

The concentrations of heavy metals Ba, Pb, Cu, Zn and Co in snow pit collected in September, 2005 from the accumulation area of the East Ronbuk Glacier (a.s.l. 6523 m), which lies on the northern slope of Mt Qomolangma, were determined by inductively coupled plasma mass spectrometry (ICP-MS), concentrations (pg/ml) of heavy metals vary from Ba2-227, Co2.8-15.7, Cu10-120, Zn29-4948, and Pb14-142, respectively. And δ¹⁸O determined by MAT-252, the time period of the snow pit spanning from autumn 2005 to summer 2004. Seasonal variations of the concentrations and δ¹⁸O are observed, of which Pb, Cu, Zn and Co are much lower in summer monsoon season than that in non summer monsoon season, suggesting that different sources of heavy metals contributed to the site. EF_c (crustal enrichment factors) is Co 3.6, Cu 27, Pb 33 and Zn 180, respectively. Higher EF_c values of Pb, Cu and Zn suggest that Pb, Cu especially Zn mainly contributed by anthropogenic sources.

Based on historical documents, modern survey and statistics, as well as the result of predecessor studies, the trend and main process of forest dynamics are recognized. The forest area and forest coverage rates for each province of China from 1700 to 1949 are estimated backward by every 50 years. Linking the result with modern National Forest Inventory data, the spatial-temporal dynamics of Chinese forest in recent 300 years (A.D.1700-1998) is quantitatively analyzed. The study shows that in recent 300 years, the forest area in current territory of China has declined 0.95 ×10⁸ ha (or 9.2 percentage points of coverage rate) in total, with a trend of decrease and recovery. Before the 1960s, there was a trend of accelerated descending. The forest area was reduced 1.66×10⁸ ha (or 17 percentage points of coverage rate) in 260 years. While after the 1960s, there has been a rapid increase. The forest area increased by 0.7×10⁸ ha (or 8 percentage points of coverage rate) in 40 years. The study also shows that there is a significant spatial difference in the dynamics of forest. The amplitudes of increasing and decreasing in western China are both smaller than the ones in eastern China. During the rapid declining period 1700-1949, the most decrease appeared in the Northeast, the Southwest and the Southeast, where the coverage rate in most provinces dropped over 20 percentage points. In Heilongjiang Province, the coverage rate dropped 50 points. In Jilin Province, it dropped 36 points. In Sichuan Province and Chongqing Municipality, it dropped 42 points. In Yunnan Province, it dropped 35 points. During the recovery period 1949-1998, the western provinces, municipality and autonomous regions, including Ningxia, Gansu, Inner Mongolia, Sichuan-Chongqing, Yunnan, Tibet, Xinjiang and Qinghai, etc., the increase rates of the coverage are all below 5 percentage points, while the eastern provinces, municipality and autonomous regions (except Heilongjiang, Hubei and Jiangsu-Shanghai) have achieved an increase over 5 percentage points, among which the Guangdong-Hainan, Guangxi, Anhui, Beijing-Tianjin-Hebei, Shandong, Henan, Zhejiang and Fujian have an increase over 10 points.

Phenological observation data of plant communities from 1982 to 1996 at 5 sites and a method for phenological cumulative frequency modeling were used to determine the beginning dates of local phenological seasons and their corresponding threshold values of normalized difference vegetation index (NDVI) in each year. Then, a year-type cluster analysis of NDVI profiles at each phenological station during 1982 to 1999 and a spatial cluster analysis of NDVI profiles for all of the pixels within the study area year by year were employed to fulfill a spatial-temporal extrapolation of vegetation phenological seasons. Consequently, we obtained spatial-temporal patterns of the beginning date of vegetation phenological seasons and the length of the vegetation growing season in the deciduous broad-leaved forest area of warm-temperate eastern China from 1982 to 1999. The results show that (1) the annual mean beginning dates of vegetation phenological seasons and the mean lengths of the vegetation growing season indicate changes of a spatial pattern mainly following latitude and altitude; (2) the beginning dates of the phenological spring dominate a significantly advanced trend over the entire area during 1982 to 1999, especially in North China Plain, whereas the beginning dates of the phenological summer, autumn and winter dominate a significantly delayed trend, also mainly in North China Plain, which causes a significant lengthening of the vegetation growing season in North China Plain; (3) linear trends of the beginning dates of vegetation phenological seasons are consistent with linear trends of seasonal air temperatures in North China; (4) the vegetation growing season lengthening revealed by the current study is consistent with the phenological growing season lengthening of the individual tree species in Europe, and the satellite-derived growing season lengthening in Eurasia and temperate China.

There has been an increasing use of predictive spatial distribution of main communities or dominant species at the regional scale for ecological restoration planning, biodiversity conservation planning and regional management decisions on the Loess Plateau. Understanding the spatial distribution of main communities at the regional scale is also essential in assessing the impacts of environmental change or human impacts on vegetation distribution. Based on the investigation of steppe cover that correlates with its habitat, this study focused on the prediction of steppe distribution in the Zulihe River watershed of the west-central Loess Plateau. First, we investigated the cover of steppe of 365 samples. Secondly, important variables (annual mean precipitation and annual mean temperature) that affect vegetation distribution were spatialized as functions of elevation and horizontal coordinates via a GIS. Thirdly, the niche spaces of steppe were determined by incorporating the spatially-distributed main variables with the steppe cover that came from field investigation and remote sensing data (Landsat TM image). The niche spaces defined then were extrapolated over the study area. Comparing the distribution of predicted steppe with its present distribution, we found that the natural vegetation in the region has been destroyed and disturbed by human activities. As a result, soil erosion has been aggravated. The ecological conditions in the study area have been progressively degraded by human-accelerated soil erosion. The object of the study is to provide important baseline for ecological restoration in the Zulihe River watershed.

Quantitative analysis was performed on filling and scouring process for the river reach between Makou and Tianjiazhen, the middle Yangtze River with the help of GIS and DEM technology. The research results indicate that the river reach between Makou and Tianjiazhen was dominated by the scouring process, and the magnitude of scouring is increasing over time. The intensity of scouring process is more in the deep and narrower river reach than shallower and wider ones. The river reach in the Makou and Tianjiazhen river knot is in frequent scouring and filling process, however the river reach upper to the Makou and lower to the Tianjiazhen river knot is in moderate scouring and filling process. The river reach just upstream or downstream to the river knot (e.g. Makou and Tianjiazhen river knot in this research) is dominated by filling process and the river reach in the river knot is dominated by scouring process. Research results indicate no changes happen in the boundary of the river but the scouring and the filling magnitude in specific river channel is strong. The filling and scouring process of the study river reach is greatly impacted by the sediments and water from the upstream of the study river reach. The construction of the Three Gorges Dam just upstream to Yichang will cause further decrease of the release of sediment load to the middle and the lower Yangtze River basin, which will further intensify the scouring process of the river channel in the study river reach.

Soil organic matter (SOM) turnover was studied for the Chongmingdongtan Salt Marsh in the Yangtze River Estuary, based on the analyses of stable carbon isotope (δ¹³C), grain sizes and contents of particulate organic carbon (POC), total nitrogen (TN) and inorganic carbon (TIC) for three cores excavated from high marsh, middle marsh and bare flat. Results indicated that correlations between soil POC contents and δ¹³C values of the salt marsh cores were similar to those of the upper soil layers of mountainous soil profiles with different altitudes. SOM of salt marsh was younger than 100 years, and originated mainly from materials of soil erosion in the catchment basin of the Yangtze River. Diagrams of TN-C/N, POC-TIC and POC-δ¹³C for the cores indicated that turnover degree of SOM from the salt marsh was generally low, and tendencies regarding SOM turnover were clear from bare flat to high marsh. Bare flat showed characteristics of original sediments, with minor SOM turnover. The turnover of SOM was discernable in the high marsh and middle marsh, and the mixing degrees of SOM compartments with different turnover rates increased with evolution of the muddy flat. The exclusive structure of muddy layer and sandy layer originated from dynamic depositional processes on tidal flat made great difficulties for vertical migration of dissolved materials, and SOM turnover was then constrained.

The analysis results of 58 surface sediment samples collected in February 2003 showed that the median diameter of surface sediment (D₅₀) was decreased gradually from Jiangyin to the mouth of the Yangtze River Estuary, and the value decreased from 217.8 μm of Jiangyin to 12.1 μm of the mouth of South Passage. In Turbidity Maximum, the coarsest surface sediment was in North Channel, and the average median diameter was 126.2 μm. The coarser surface sediment was in North Passage, and the average median diameter was 48.4 μm. The finest surface sediment was in South Passage, and the average median diameter was 14.2 μm. While in the mouth of the estuary, the coarsest surface sediment was in North Passage, then in North Channel, and the finest surface sediment was in South Passage. The median diameter of the mouth of North Passage was 22.4 μm, and that of North Channel 16.5 μm, and South Passage 12.1 μm. The sorting of surface sediment was mainly moderately sorted to very poorly sorted, and the greater the median diameter, the smaller the sorting coefficient. The skewness was mainly nearly symmetrical to very positive. The kurtosis was mesokurtic to extremely leptokurtic. The composition of surface sediment in upper Hengsha island was mainly sand, while in the mouth of the estuary was mainly clayey silt. In Turbidity Maximum, it was much more complex and variational. There was positive relationship between the D₅₀ of surface sediment and the ebb tidal duration, value of V_e/V_f and residual current speed in channel of the Yangtze River Estuary, and the ebb current condition was the main dynamic factor to determine D₅₀.

Satellite images acquired in 1984, 1991, 2000 and 2005, respectively, were used to interpret the boundaries of cities and towns in the south of Jiangsu province along the Yangtze River on the basis of our boundary definition. And then regional urban spatial sprawl pattern was uncovered quantificationally by some indexes, such as fractal dimensions including radius dimension, grid dimension, correlation dimension, boundary dimension, compactness index, and spatial autocorrelation index including global Moran I and local Moran I. The result shows that: (1) fractal dimension metrics and spatial autocorrelation index, respectively, can quantificationally reveal regional urban spatial distribution features and spatial sprawl process and cluster pattern, and their fuse can measure regional urban spatial sprawl pattern and uncover the process of how to form the urban cluster; and (2) at the initial stage of reform and open-up policy, the urban sprawl in the south of Jiangsu province along the Yangtze River was spontaneously point-pattern centered around big cities, and then point-axis-pattern dominated by policy-driven urban spatial fuse and fill, at present cluster-stretched pattern based on regional industry spatial organization driven by economical radiation and contact, and a big urban group is coming into being. Especially since 2000, spatial structure features of an urban group are more and more evident with time elapsing. Additionally, the main feature of urbanization in the study area is that industrialization pushes urbanization based on the background of global economy. In the study area three centric cities have existed since the 1980s, however, their spatial polarizations have a less and less effect on others.

According to the theory of abrupt climatic change, the regional economic development process in Jiangsu province since 1978 is divided into three periods in this paper. Then the paper employs urban centrality index and time cost for transportation, instead of the city scale and the distance factor, to modify the original potential model. Based on the grid arithmetic of GIS methodology, the paper identifies the scope of urban economic effect regions (UEERs) for central cities at different ranks in Jiangsu province of the main years. Accordingly, Jiangsu province can be separated into different ranks of UEERs. Consequently, the characteristics of UEERs in Jiangsu province since 1978 can be summarized below: (1) the spatial structure changes from South-North to East-West; (2) the UEERs in the southexpands towards north; and (3) the new central cities in southern Jiangsu province occur. Furthermore, the mechanism of evolution of UEERs in Jiangsu is also analyzed in this paper. The development of transportation axes, the diversities of economic development level and the disparities among the input of social production factors are the three main reasons accountable for the changing patterns of UEERs. The effects of transportation are measured by density of transportation network of each county or city which is calculated by different weighed values for highways, expressways and so on. The economic development level measured by GDP per capita is a direct way to reflect the disparities of capability for central cities. And the influence of social production factors is weighed by analyzing the index of total population, fixed assets investment per capita and the number of the labor force. It is concluded that transportation axes play the most important role in the spatial structure formation of UEERs among the three factors; besides, the capital-intensive factor input has greater influence on UEERs than the labor-intensive one does to a great extent.

This paper selects the statistics of 183 cities in China, in the years of 1991, 1995, 1999 and 2003. We adopt the Natural Breaks with Arcmap software to divide these cities into five ranks according to their urban transportation development levels, and use the SPSS to analyse the development level rank disparity change which characterizes the spindle structure—the plus number of the cities which are better and worse than the average level is less than 32% of the total. Meanwhile, more than 68% of the cities' transportation development levels are close to the average one. Cities with higher urban transportation development level than the average present the "centralized - decentralized - decentralized - centralized" characteristics in provincial spatial distribution, while those with lower level than the average present the decentralized and irregular spatial pattern. The disparity of the urban transportation development level rank in China appears a U-shaped changing pattern—it decreased in 1991-1995, increased in 1995-1999, and then decreased in 1999-2003 again, but the fluctuation range was not great in each period. Cities with urban transportation development level going up spatially centralized in Henan, Shandong, Jiangsu and Liaoning, while those with degressive development level spatially are centralized in Heilongjiang, Jilin, Shaanxi, Hubei and Hunan. Applying to the relative analysis it can be found that, in China, there exists a significant positive correlation betweent the development levels of urban transportation and economy. At the same time, the economic structure is relevant to the urban transportation development level. In 1991-2003, the urban transportation development level exhibited a distinct correlation with the gross industrial output value per capita at 0.01 level, and the correlation coefficient took on an ascending trand.

The triangle method, as an intuitive platform for illustrating sustainability status and trends in economic development, seems to hold promise as an analytical management tool given its simplicity, ease of use, and flexibility. The triangle model can serve as an intuitive platform for illustrating existing sustainability status and trends in economic development coincident with the implementation of various measures. The intuitive aspect of the model makes it easy to understand and use by researchers and policy-makers from many different levels. The method's flexibility in the selection of individual indicators and basic analytical units, its ease of understanding and use by experts and managers, and its simplicity in calculating and presenting status or trends of sustainability would seem to offer the potential for widespread use. Based on the interrelationships among economic development, resource-energy consumption, and environmental pollution, in conjunction with ecological performance and the triangle method, a definition for sustainable economic development was proposed, and a novel triangle method was designed to evaluate economic development sustainability. Chuzhou City is located at the middle east of China and lies in the middle zone among the country's three major zones of the eastern zone, middle zone and western zone. It is adjacent to the economically developed Yangtze Triangle Region. It had a total population of 435.62 ×104 in 2005. It covers a total area of 135.23 ×104 hm² with four counties, two cities and two districts under its administration. Because Chuzhou City is located between the economically developed eastern coastal areas and economically backward western areas , the study of economic development is of typical to a certain extent. As a case study, the triangle method was applied to assess the sustainability status and long-term trends of Chuzhou's economic development. The results show that economic development in 2005 represents a relatively commonly state of sustainability, and the seven political regions in Chuzhou City reflect sustainability positions ranging from weakly sustainable to commonly sustainable. The sustainability in the southeast is better than in the northwest. The Chuzhou's economic development between 1975 and 1998 reveals a common sustainable trend, while that from 1998 to 2005 demonstrates a relatively strong sustainable trend. Chuzhou's unremitting efforts in environmental protection and in the increasing integration of environmental considerations within the economic development process over the last three decades are contributory to these status and trends.

By using the method of MEDD, from two aspects of shape and mobility, this paper explores the evolution of spatial distribution dynamics of provincial economic growth in China from 1978 to 2004. The research results indicate that the shape of provincial economic spatial distribution has changed from stratification to convergence and then to two peaks state in China. Besides the gap among the economic growth levels of provinces within a convergence club has varied from widening to narrowing again, whereas the gap among the economic level of different convergence clubs has varied from narrowing to widening again, indicating that the economic inequities between the two convergence clubs have continue to expand. After the 1990s the regional economic growth became more energetic. Transition matrices, unconditional and space-conditioned dynamic stochastic kernel and contour plots reveal that spatial interactions influence the spatial distribution of provincial economic growth in the future. The spatial context is a very important factor in affecting the distribution of economic growth. It is very likely that spatial proximity stimulates spatial interaction. The economic growth of a region depends on the performance of its neighboring provinces. Provinces with wealthier neighbors have a greater chance to become prosperous. The spatial spillovers from the economic development of provinces having close geographical positions promote provinces with similar economic level to cluster spatially.

This paper focuses on market potential by taking it as an indicator of market accessibility from a viewpoint of location-market connection, based on the analysis of district-level data in China. The results indicate that market potential index (MPI) degrades from the coastal areas to the inland areas. Furthermore, disparity of market potential between the coastal areas and the inland areas is still expanding unceasingly. Manufacturing industries concentrate in the regions with high market potential. A significant spatial correlation between market potential index and regional economic performance is observed at regional level. Such a spatial correlation has verified that market accessibility is one of the determinants of location choice for manufacturing industries. It also has revealed that there is a circulatory causality between market potential and regional development. The circulatory causality may spur the coastal areas to keep predominance in future and then lead to a continuous expansion of regional disparity. As the inter-regional economic exchange is significant to location-market connection, location analysis should take into account the inter-regional economic exchange. Western China is restricted to develop external linkage with the coastal areas due to long distance to the latter and expensive transport cost. Therefore, the western areas should make a point of expanding internal market to improve market accessibility.

The relief degree of land surface (RDLS) is an important factor in describing the landform macroscopically. Under the new proposed concept, based on the macro-scale digital elevation model data, by using ARC/INFO software, the RDLS of 10km×10km grid size is extracted and mapped in China. Then this paper systemically depicts the distribution rules of RDLS in China and its correlation with population distribution by analyzing the ratio structure, spatial distribution and altitudinal characteristics of the RDLS. The distribution rule is elaborately expatiated in three separate ways: the ratio structure, the accumulative frequency, and the change along with the longitude and latitude, which clearly reflects the regional topographic framework of China. The result shows that the majority of the RDLS is low in China, for more than 63% of the area in China with the RDLS lower than 1 (relative altitude ≤ 500 m). As for the spatial distribution, in general, the RDLS of the west is higher than that of the east and so is the south than the north. Specifically, the Hengduan Mountains and the Tianshan Mountains regions have the highest RDLS, while the Northeast China Plain, the North China Plain and the Tarim Basin have the lowest ones. The RDLS of 28^oN, 35o^oN and 42^oN as well as of 85^oE, 102^oE and 115oE accords well with the three topographic steps in China. The RDLS of China decreases with the increase of longitude and the change clearly illustrates the landform characteristics that most of the mountains are located in the west and most plains in the east of China. The RDLS of China decreases with the increase of latitude as well and the trend shows that there are more mountains and hills in South China and more plains and plateaus in North China. In the vertical direction, the ratio of high RDLS increases with the increase of altitude. Finally, this paper analyzes the correlation between the RDLS and population distribution in China and the result shows that the RDLS is an important factor affecting the distribution of population and most people in China live in low RDLS areas. To be more specifically, where the RDLS is zero, the population amounts for 0.83% of the total; where the RDLS is less than 1 (relative altitude ≤ 500 m), the population reaches 20.83%; where the RDLS is less than 2, the population amounts for 97.58% of the total; and where the RDLS is bigger than 3, the population only amounts for 0.57%. That is to say, more than 85% of the population in China lives in areas where the RDLS is less than 1 and less than 1% of the population lives in areas where the RDLS is bigger than 3. The correlations between the RDLS and population distribution of eight regions in China are different. The correlation is obvious in northeast, north, central and south China, while it is nearly nonexistent in Inner Mongolia and the Qinghai-Tibet region.

It is of great importance to analyze urban commercial spatial structure from the viewpoint of consumers' behavior. During the latest ten years, with the development of suburbanization, the spatial structure of urban commerce in Beijing has greatly changed. Under this background, Beijing residents' shopping behavior has greatly changed. Based on more than five hundred questionnaires, the authors try to analyze the features of the evolution of spatial structure of shopping behaviors of Beijing's residents in the latest 10 years and their mechanisms. The average distance of residents' shopping trip was shortened in Beijing in the latest 10 years. The change of the spatial structure of shopping behavior of each kind of goods differs from each other. So does that of the shopping sites of each kind of goods in Beijing. The change of the hierarchical structure of shopping behaviors of Beijing's residents is also very interesting, for the hierarchical structure of the inner suburban residents descended while that of the core residents had little change. In a word, the features of the changes of the spatial structure and hierarchical structure all reflect the tendency of decentralization and polycentric urbanization of the development of Beijing's commerce. It should be pointed out that Beijing's residents tend to buy goods of low hierarchy near where they live, such as vegetables and the daily use articles, while they tend to buy goods of high hierarchy in shopping malls, large supermarkets and shopping centers for electric equipments, for these places are more attractive. The change of macroscopic environment, that of shopping suppliers and that of consumers are all significant factors propelling the evolution of the spatial structure of Beijing residents' shopping behaviors. Based on these factors, the authors put forward a theoretical frame of the mechanisms of the evolution of urban commercial spatial structure.

The essential part of geographical cellular automata (CA) is to provide appropriate transition rules so that realistic patterns can be simulated. Transition rules can be defined by a variety of methods, such as multicriteria evaluation (MCE), logistic regression, neural networks, and data mining. The solicitation of concrete knowledge (transition rules) is often difficult for many applications. There are problems in representing complex relationships by using detailed rules. This study demonstrates that the case-based approach can avoid the problems of the rule-based approach in defining CA. The proposed method is based on the case-based reasoning techniques, which don't require the procedure of soliciting explicit transition rules. The knowledge for determining the state conversion of CA is inexplicitly embedded in discrete cases. The lazy-learning technology can be used to represent complex relationships more effectively than detailed equations or explicit transition rules. This paper presents an extended cellular automaton in which transition rules are represented by using case-based reasoning (CBR) techniques. The common k-NN algorithm of CBR has been modified to incorporate the location factor to reflect the spatial variation of transition rules. Multi-temporal remote sensing images are used to obtain the adaptation knowledge in the temporal dimension. This model has been applied to the simulation of urban development in the Pearl River Delta which has a hierarchy of cities. Comparison indicates that this model can produce more plausible results than rule-based CA in simulating large complex regions.

Mapping distribution of population has arisen as an important issue in the fields of geographical and relative researches, due to the necessity of combining with spatial data representing socio-graphic information across various spatial units, such as to evaluate the total numbers of people at environmental health risks or died in natural disasters. However, most existing solutions to this problem focus on selection and quantification of influencing factors and rarely take into account the correlation among selected factors. And much expertise is needed in modeling process to formulate the relationships between influencing factors and population data successfully. It usually not only produces information redundancy but increases the complexity of the problem. This paper explores a novel approach to transform population data from census to grid by integrating genetic programming (GP), Genetic Algorithms (GA) and Geographic Information Systems (GIS). A set of natural and socioeconomic factors which contribute to population distribution are identified and quantified under GIS environment. And then GP and GA are severally applied to build and optimize the population model in the hierarchical form, allowing for the computation of the relevant population data error. The experiment proves that the proposed method performs much better than stepwise regression analysis and adapted gravity model approaches. The GP/GA-based method is the first to introduce such computational intelligence techniques as GP and GA to generate gridded population maps, hence it is a methodological innovation in interpolation of population data.

Land use/cover change (LUCC) is widely recognized as one of the most important driving forces of global carbon cycles. The influence of LUCC from native forest to plantations, secondary forest, orchard and arable land on soil organic carbon (SOC) stores and quality was investigated by using "space for time" method in mid-subtropical mountainous area of southern China. The results show that LUCC has led to great decreased SOC stocks and quality. Considerable SOC and light-fraction organic carbon (LFOC) have been stored in native forest (142.2 t hm^-2 and 14.8 t·hm^-2, respectively). When native forest was converted to plantations, secondary forest, orchard and arable land, the SOC stocks decreased by 25.6% -28%.7%, 38.0%, 31.8% and 51.2%, respectively. The LFOC stocks decreased by 52.2-57.2% when native forest was converted to woodland plantations and second forest, and by 82.1% -84.2% when to economic plantation, orchard and arable land. After the conversion, the ratios of LFOC to SOC (0-60 cm) decreased from 13.3% to about 3.0%-10.7%. The SOC and LFOC stored in the upper 20 cm soil were more sensitive to LUCC when compared to the subsurface soil layer. Comparison suggests that the effects of the above mentioned LUCC on carbon release and quality deline were greater than the global average level, which could be explained by the vulnerable natural environment and special human management practices. Thus, it is significant to protect vulnerable environment, restore vegetation coverage and afforest plantation reasonably in mountainous area in mid-subtropics, for enhancing soil carbon sequestration, mitigating elevated atmospheric CO₂ and sustainably developing ecological services.

In this paper, a new and advanced time series analysis method--the empirical mode decomposition (EMD) method is presented into climate analysis field. It is a method developed from analyzing nonlinear and non-stationary data. The oscillations of different scales or trend in the signal are decomposed into a number of characteristic intrinsic mode function components. This decomposition method is adaptive, and, therefore, highly efficient. Thus, we could extract the variation trend from the data. The testing results indicate that EMD method is the best one for extracting data trend at the present time. The climatic trend is very important in temperature change. Therefore, how to eliminate periodic oscillation in temperature change and obtain variation trend is the important process for estimating and comprehending global climate warming. In this study, temperature variation trends of monthly mean temperature data for 740 stations over China from 1951 to 2001 are diagnosed by EMD method, and there are three types of variation trends: ascending trend, descending trend, and fluctuation. Hereby, several temperature change regions have been divided in China and fields have been spatially classified. Simultaneously, the temperature variability of every station is calculated by this method, and classification chart of long term trend and temperature variability distribution chart of China are obtained, supported by GIS, 1km×1km resolution. The results show a large scale warming trend in China, especially Northeast China, Inner Mongolia, Gansu, the west of Qinghai and Tibet, and the north of Xinjiang witnessing marked warming, as opposing to the descending trend patches distributed over Southwest China and the middle and lower Changjiang valley.

Using global Palmer drought severity index (PDSI) database, the seasonal and annual PDSI mean values of 56 grid points covering the period 1953-2003 over northwestern China is analyzed by rotated experience orthogonal function (REOF). The results show that the seasonal and annual means display similar anomaly regions, and furthermore there are five main anomaly regions in the northwestern China based on the annual PDSI mean, i.e., the northern and southern parts of Xinjiang, northern part of plateau, western part of Inner Mongolia and eastern part of northwestern China. The opposite changes of dry-wet conditions are discovered after comparing the curves of feature point series and binomial fit between the eastern and western parts of the northwestern China: The western part of the northwestern China influenced by westerly, a trend of wetting gradually occurred since 1980. On the contrary, a trend of drying gradually occurred in the eastern part of the northwestern China which is mainly influenced by Asian monsoon, and especially in summer and autumn, there is notable trend of drying over the southeastern part of the northwestern China. The northwestern China is obviously affected by westerly, Asian monsoon and Tibetan Plateau. There are differences among the responses of the dry-wet conditions to the global warming in different regions.

Sediments from inland enclosed lakes in arid regions of northwestern China provide ideal documents in reconstructing high-resolution history of aeolian dust and aerosols. Sugan Lake in northern Qaidam Basin on the Tibetan Plateau is an enclosed shallow lake mainly fed by groundwater, in which annually laminated varves are well preserved since mid- and late Holocene. Surface sediments from Sugan Lake were collected and analyzed for grain size. Frequency curves of samples indicated multi-components in the surface sediments. Weibull function fitting employed to the grain size distribution suggested that there are four components from clay to sand in the surface sediments characterized by their respective modes. Super-fine component has mode size around 1 μm, fine-grained component has mode of 5-10 μm, coarse-grained component has mode of 50-100 μm, and the mode of sand component is about 500 μm. The main components in the surface sediments appear in fine or coarse grains. Particles trapped in the lake ice in winter and dust storms during the same period present the similar distribution pattern in grain size. The main component of aeolian dust from lake ice is composed of particles sized between 15 and 20 μm, and sand component (with mode size of 452.9 μm) appeared, indicating that sand particles were transported to Sugan Lake during the winter. Dust samples from dust storm period (May) or non-dust storm period (November) prevalently consist of three components without sand grains. During strong dust storm events, however, four components including sand are identified, the main component of which still has a size of 100-200 μm. Our results show that the sand component of the surface sediments in Sugan Lake has been transported by ambient strong dust storm activities; the coarse grains have most likely recorded the regional dust storm events; fine component seems to have indicated the intensity of water supply in the basin; super-fine component as extensively deposited in the Chinese loess, may have represented the background level of atmospheric dust and aerosols in the arid region of northwestern China. Therefore, the sediments from Sugan Lake have documented the regional aeolian dust and dust-storm events and have the potential in reconstructing the history of aeolian dust activities.

In this study, short-term gully retreat was monitored from the active gullies selected in representative black soil area, using differential GPS. With the support of GIS, multitemporal digital elevation models (DEM) were constructed in light of the data collected by GPS and used for further analysis. Based on the analysis of multi-temporal DEM, we discussed the erosion-deposition characteristics of a gully and a developing model for black soi l gully area of Northeast China was proposed. The results are: (1) The analysis of the monitored gully data in 2004 indicated that the retreat of gully head reached more than 10 m, gully area extended 170-400 m², gully net eroded volume 220-320 m3 and gully erosion modulus 2200-4800 t·km^-2·a^-1. (2) Compared with the mature gully the initial gully grows rapidly, and its erosion parameters are relatively large. The erosion parameters have not only to do with flow energy, but also with growth phase. (3) There are significant seasonal differences in gully erosion parameters. The extension of gully area and width dominates in winter and spring without marked net erosion while changes mainly occur in gully head expansion and net erosion in rainy seasons. (4) It is remarkable for freeze-thaw erosion in the black soil area of NE China. The gully wall of SG2 extended 0.45 m under freeze-thaw effect in 2004, and the distance of gully head retreated maximally 6.4 m. (5) Due to freeze-thaw action and snowmelt, gully is primarily in the interior adjustment process in winter and early spring. There are much more depositions compared with that in autumn, which can almost happen throughout the gully, while erosion mostly occurs near head, esp. for gullies having a relatively long history of development.

The chemical weathering intensity and element migration features of the Xiashu loess profile in Zhenjiang are studied in this paper. Based on the comparison with other aeolian-dust deposits, i.e. Quaternary loess-palesol and late-tertiary red-clay profiles in the Loess Plateau, and aeolian dust red earth in Xuancheng of Anhui province, we draw the following conclusions: (1) The Xiashu loess profile underwent moderate chemical weathering, which was stronger than that of the loess and palesol in Luochuan, much weaker than that of the aeolian dust red earth in Xuancheng, and similar to that of late-tertiary red-clay in Xifeng. The chemical weathering differences among the studied aeolian-dust deposits were mainly induced by the distinction of the climate condition, which imposed important influence on the geochemical environment through the mean annual temperature and annual precipitation. It seems that the annual precipitation has more important influence on the chemical weathering process. (2) The major elements such as Si, K, Na, Ca, Mg, Mn and P were migrated and leached, while the elements Fe and Ti were slightly enriched during the chemical weathering process. According to the migration ability, the major elements are ranked in the following order: P > Na > Ca > Mg > K > Fe²⁺ > Si > Mn > Al > Ti > Fe³⁺. The migration features of the major elements reveal that the Xiashu loess finished the primary process of chemical weathering characterized by leaching of Ca and Na, and almost reached the secondary process characterized by leaching of K. Except the elements Sr and Ga, other trace elements such as Th, Ba, Cu, Zn, Co, Ni, Cr and V were enriched during the chemical weathering process. And this enrichment might be caused by both the biogeochemical process and the adsorption of trace elements by clay mineral and organic materials during the chemical weathering. (3) The difference of element migration down the Xiashu loess profile reveals that the climate was warm and wet at the early-middle stage of the Middle Pleistocene (before 0.24 Ma), however, bearing the feature of an alternate dry and wet fluctuation. At the end of the Middle Pleistocene, it became dry and cool, which resulted in the weakest weathering and element migration. At the early stage of the Late Pleistocene, the paleoclimate became warm and wet again, which led to rather strong pedogenesis and the formation of the paleosol S1 in the Xiashu loess profile. As a whole, the paleoclimate generally became drier and cooler in this region from the beginning of the Middle Pleistocene.

River channel sedimentation in the lower reaches of the Yangtze River can be affected by both rising sea level due to global warming and the construction of reservoirs in the upper river that reduces the amount of sand discharge. To evaluate the dynamic changes of sedimentation and erosion of the Jiangsu reach of the Yangtze River (about 330 km in length) from 1958 to 2003, databases of five periods Digital Elevation Mode (DEM) of underwater topography were designed and constructed on the basis of the Jiangsu River Relief Map in 1959, 1970, 1985, 1992 and 2003 (1:25000-1:60000). The result indicated that the main course of Yangtze River in Jiangsu Province had experienced an obvious switch from sedimentation to erosion around 1985 because of the decreasing amount of sediment yield from the upper Yangtze River after that year. The sedimentation process in the main course of Jiangsu reach of the Yangtze River demonstrated propulsive phenomena of "down-reach-ward accumulation". Between 1985 and 2003, the erosion rate of the lower segment was greater than those of the middle and upper segments. The reason is probably the bigger shortage of sediment yield due to the decrease of sediment source from the upper reach and the strengthening protection of the bank erosion in the lower reach.

This paper presented the vertical mass flux profiles of different grain size groups in aeolian sand transport, which is the hot issue of non-uniform sand transport research but is rarely studied recently. The sand samples of vertical mass flux were further investigated using a particle size analyzer that were measured in a wind tunnel experiment conducting with non-uniform sands and five different stream velocities. Results show that the vertical mass flux profiles of coarse sand groups are similar to that of all sands measured in the wind tunnel, exhibiting an exponential decay distribution with a positive deviation occurring in the near-bed region; however, the vertical mass flux profiles of fine sand groups decayed exponentially in the whole height. Based on the results above, this study simulated the vertical mass flux profiles of different grain size groups in terms of equilibrium saltation theory. It is revealed that in the non-uniform sand transport, the launch speed and angle distributions of coarse sand are exponential function; the launch speed of fine sand is exponential function but the launch angle of it is Gaussian function. The mean sand launch speed occur between 0.4u* and 2.2u*, decaying with the increase of wind speed (u*) and grain size.

Based on the three temporal satellite remote sensing TM images of Tongshan county located in Xuzhou city, basic data about land use of Xuzhou coal mining area are obtained with the neural network classification module from 1987 to 2003. According to the basic data of land use, the land use change is quantified by way of computing the index models of regional land use change. The following results were obtained: (1) The areas of construction land and water body have continually increased from 1987 to 2003. In this period the area of construction land has increased by 4761.09 hm². On the contrary, the area of cropland has constantly decreased with a decrease of 27528.21 hm². (2) From 1987 to 2003, both the construction land and unused land were chiefly transformed into cropland. From 1987 to 1994, the water body was chiefly transformed into woodland. The cropland was chiefly transformed into unused land and construction land. The woodland was chiefly transformed into cropland. From 1994 to 2003, the water body was infrequently transformed. The cropland was chiefly transformed into unused land and woodland. (3) From 1987 to 2003, the variation index of land use degree and the dynamic degree index of land use demonstrate that land use in Tongshan county was in a decline period and the changes were drastic.

Analysis of land use/cover change in six periods from 1954 to 2005 was conducted with the aid of remote sensing and GIS techniques on the watershed scale. The landscape diversity of wetlands within Honghe Nature Reserve and the surrounding area in the watershed was calculated by using landscape diversity index to reflect landscape diversities and their changes in historical periods. At the same time, the landscape connectivity of the Honghe Nature Reserve with the surrounding area of the watershed was calculated to study the effects of landscape isolation on landscape diversity of the wetlands. Results show that the large-scale land reclamation started in the 1980s resulted in an annual increase of 101 km² of agricultural land. The direct land use/cover change brought about the decline of wetland landscape diversity in watershed, and also impacted the diversity of wetlands in Honghe Nature Reserve. Decline of landscape connectivity between wetlands within the nature reserve and the surrounding area of the watershed due to land use changes has seriously segmented the original relationship between the nature reserve and its surrounding area, which directly resulted in the decline of wetland landscape diversity within the Honghe Nature Reserve. At present, the decline of landscape diversity of the wetlands in the Honghe Nature Reserve has serious negative effects on wetland structure and its ecological function. So it is necessary not only to conserve the wetlands in the nature reserve, but also to protect the wetlands in the surrounding area. We should restore certain area of the marsh wetlands to keep a high diversity of the landscape and landscape structure of the wetlands within the nature reserve, as well as to restore certain area of the wetlands in the surrounding area of the nature reserve for maintaining a high landscape connectivity in the watershed so that good conservation benefits of the nature reserve in the Sanjiang Plain can be achieved.

Heterogeneity of landscapes is supposed to be of significant importance for biodiversity and ecological function in urban ecosystems.However, land use activities caused primarily by urbanization process has a profound and far-reaching impact on stability of landscape and differences of landscape structure at the regional scale. Therefore, in this paper, the northwest region of Beijing is chosen as a study area and the technologies of remote sensing (RS) and geographical information system (GIS) is applied, and meanwhile the priority is been given to understanding of the spatial-temporal characteristics of landscape pattern change through visually interpreting Landsat TM images of 1989, 1996 and 2005 respectively. It is believed that there are a series of landscape ecological problems because of city expansion, such as low landscape ecological connectivity, simplified landscape structure and more obvious fragmentation of green land patch in the plain areas. Moreover, spatial distribution of green land is unbalanced. In order to solve those problems, according to the principle that the movements of flow, energy and material in a landscape are related to some factors such as distance, time, impedance, etc., this study adopts an accumulative cost distance model, combines with ecosystem services and spatial interactions of landscape types, analyzes the spatial difference of the ecological function and the compactness of landscape structure in the study area, and further discusses the landscape pattern optimization proposal.

In the case of banking, this paper defines local financial sys tems as thos e financial ins titutions which s erve different spatial ranges located in an adminis trative region. We divide these banking ins titutions into two types : the exogenous ins titutions and the endogenous ins titutions . The exogenous ins titutions are national s tate-owned banks , and the endogenous ins titutions are local banks . The former ones contain the s tate-owned commercial banks , the policy banks , the national joint shareholding commercial banks ; and the latter ones include the city commercial banks and the rural credit cooperatives . With the two types combined, local financial sys tems experience three s tages : The firs t one is the budding s tage . In this s tage , the exogenous ins titutions dominate the financial s ervices in a region; the endogenous ins titutions jus t begin their financial affairs . The s econd one is the developing s tage . Local financial sys tems have the dynamic changes in this s tage . On one hand, sys tems become more attractive for the exogenous ins titutions and make the endogenous ins titutions grow up fast; on the other hand, sys tems develop so s low that the exogenous ins titutions choos e to leave the region. The third one is the mature s tage , the endogenous ins titutions develop to the exogenous ins titutions , and sys tems change to financial centers . Based on the data from the s tatis tics of the Central Bank, this paper analys es the spatial differences among provinces . The result shows that the eas tern part plays an es s ential role in both the ins titution numbers and the as s ets ; local financial sys tems in this part are in the mature s tage . While the middle and wes tern parts have a small portion compared with the eas tern, the sys tems in thes e areas are in the budding and s low developing s tage . In local financial sys tems , the branches of s tate-owned banks dominate local financial sys tems ; the foreign banks concentrate on important city nodes ; the national joint shareholding commercial banks centralize in certain areas ; the city commercial banks are unevenly dis tributed; and the rural credit cooperatives have little contribution to local economy.

This paper analyzes the urban inequality of Yunnan Province during 1990 to 2004. The authors find that the differentiations among the comprehensive strength of cities of Yunnan tend to increase since the 1990s. When most of the cities in Yunnan sprawl slowly and change unobviously, only a few highest-rank cities are in the process of polarization, presenting clear central-peripheral effects. Thus, the urban economy organization that has influenced the evolution of the urban-regional structure has demonstrated these following features: (1) The urban economic effect region (UEER) in Yunnan was splitted into the discretization patterns that they were fragmentized in geography before the 1990s; (2) since the 1990s, with the fast pace of the regional industrialization and urbanization as well as the growth and the radiation of the communication network of transportation, the medium- and small-sized cities have been agitated. Furthermore, with the unbalanced spatial expansion and the shrinkage of UEER, that is, UEER has gradually transformed from a simple discrete distribution structure into a three-layered-nesting structure. The UEER of Yunnan begins to reveal the obvious spatio-temporal order: from the miniature discrete patterns and the transitional patterns to the three-layered-nesting structure patterns. Finally, based on this spatio-temporal evolutive sequence, the authors analyze the three types of the spatial spread of the UEER in different spatio-temporal combinations: dominant-direction spread type, the multi-direction spread type and the abutting spread type.

This paper examines the micro foundations of geographical concentration of Chinese manufacturing industries in China at very disaggregated levels using the most recent economic census data. The empirical results indicate that natural advantages, agglomeration economies and institutional changes together influence industrial location in China. Overall, industries bearing higher transportation costs and difficulty to ship are largely dispersed. Resource-based industries follow the pattern of natural advantages and show less agglomeration but metal mineral consuming industries are agglomerated. Trading establishments and foreign enterprises are heavily concentrated, confirming the importance of globalization effects. However, local protectionism has indeed discouraged industrial agglomeration, but provincial governments are more likely to succeed in exercising local protectionism policies and imitation strategy compared to the county governments. Agglomeration economies have done a better job in driving the geographical concentration of Chinese industries at the county level than at the province level. Proxies for knowledge spillovers are highly significant at the county level. The findings suggest that the spatial scale matters in understanding industrial clustering, and economic transition and its consequence are also critical in explaining the spatial pattern of Chinese industries.

On the bas is of re-unders tanding the spatial scope and spatial contents of the Yangtze River Economic Zone (YREZ), applying the usual spatial data analytical methods , the natural characteris tic of spatial divis ion of the YREZ has been attained. The special band-shaped core-periphery-edge s tructure of the YREZ was formed after the four developing periods of independent development, regional concentration and connection, adminis trative economic region and spatial competion and connection and so on. The YREZ has three cores , one main axis , one secondary axis , four locational axises , one peripherical region and one edge region. The bas is of the formation and evolvement of spatial s tructure in the YREZ has been analyzed from complementarity, trans ferability and tervening opportunity and other parts by applying sptial interaction theory. To speed up optimization of spatial s tructure of the YREZ, we should bring forth new ideas in the regional organization of adminis tration. Then the es sential factors of spatial s tructure need to be optimized. One regional organization and adminis tration mis s ion, known as GCT model (G represents government, C s tands for city adminis trators and T is transnational-corporations or trans regional-corporations ), should be the mos t ideal organization model, which is composed of local government, city adminis trators and the mas ter enterprises pos ses s ing transnational-corporations or trans regional-corporations networks . When a mechanism of reaching unanimity through consultation and combination between the three bodies of local government, city and transnational or trans regional-corporations is set up, spatial adminis tration of the YREZ could be mos t efficient. Among the factors of spatial s tructure of the YREZ, the nodes and the pas sageway are the framework as well as the two mos t important factors optimizing spatial s tructure .

The number of touris ts is one of the key indexes in measuring the development of eco-tourism area . By s tudying touris t flow and its variation law and by analyzing the flow s eries to predict its trend, s cientific references could be provided for the area's planning, decis ion-making, protection, exploration and operational management. The development of eco-tourism area is subject to the interactions of a number of certainties and uncertainties , which is reflected via obs ervational touris t flow s eries . This paper es tablishes a touris t s erial analys is framework and puts forward its theoretic model. Taking Yancheng eco-tourism area for David's deer in J iangsu Province as a s tudy cas e , the paper applies various mathematic methods to dis compose , readjus t and synthetically forecas t touris t s eries . The Tramo/Seats method is us ed to undertake seasonal regulation analys is for extracting each component of time s eries , hence obtaining functional models at various s cales of long-term trend with s easonal, periodic and irregular variations . Moreover, the internal law of touris t flow fluctuation in eco-tourism area and its correlation between its influenting factors are analyzed, and thus we get the nois e elimination s eries which provides foundation for a forecas t s tudy. The synthetic forecas t results obtained by applying ARIMA model, Winter addition model, Winter multiplication model and Tramo/Seats model indicate an annual average increase of 19.5-3.14 thous and touris ts in the coming two years , coinciding with the former s eries . The differences in forecas t results between each method are relevant to the modeling thoughts and means of realization, this paper collects various methods to render forecas ts and analys is in order to gain more comprehens ive knowledge about touris t flow fluctuation. Furthermore , the various pos s ible forecas ts could offer references for capacity control, protection and exploration of tourism environmental resources and decis ion-making of tourism management.

The paper focuses on the environmental explanation of the selection of settlement sites belonging to different types of archaeological culture in the middle Holocene Epoch, Western Liaohe River Valley. Through Chi-square Test, the authors point out that the settlements in the target region were mainly distributed in the sloping field of 400-500 m above sea level and the distribution presents spatial and temporal changes. The settlements in 8000-6000 a BP concentrated on the sloping field of 400-500 m above sea level. The settlements in 6000-3000 a BP were extending from the core area of 400-500 m above sea level to the higher and lower directions. While showing the preference of the height of 400-500 m above sea level, the settlements in the Western Liaohe River Valley in the middle Holocene Epoch demonstrated a horizontal amassment in the sloping field 10-40 m away from the riverbed. From the environmental explanation of the selection of settlement sites belonging to different types of archaeological culture, the paper gets three main findings: 1) the Western Liaohe River Valley could be divided into two parts, the woodland and the grassland, according to the style of living of the people. The periphery of the woodland is the living space of the people 8000-6000 a BP, when gathering, fishing and hunting were the main life style. The grassland is the main distribution area of primitive agriculture. 6000-3000 a BP is the vital period for the development of primitive agriculture, during which the settlements had a trend expanding from the periphery of woodland to the grassland. The sequence of the settlement sites defines the spatial and regional features of environment. 2) Though the Holocene Epoch experienced a prominent warming period, the animal and vegetation resources were far from abundant, restricted by the zonality constraints of the Western Liaohe River Valley. The living resources reserves could only afford the living of the people in 8000-6000 a BP when the population density was very low. In 6000-3000 a BP, the growth of population had exceeded the environmental capacity of this area. To fulfill the survival requirement, primitive agriculture was developing gradually and new man-land relationship was established. 3) Using thickness of the cultural sediments to analyze the continuance of the settlements in the middle Holocene Epoch, the paper points out that the population density then was much sparser than modern times. Considering the environmental capacity and population density, agriculture cultivation in the middle Holocene Epoch did not cause significant environmental disturbance.

The forming proces s has been resumed according to his torical materials and researched results . That was a proces s which was from nongovernmental to governmental reclamation, from Han people's tenancy to Meng and Han people's joint reclamation, and from the edge to the interior of Ordos . The changing mechanism of the farming-grazing trans itional land use pattern in Ordos was discus sed based on the factors of natural conditions , population pres sure , economic benefit and amalgamation of nationalities , etc. The period of warm-humid climate in the Qing Dynas ty provided ideal conditions for reclamation; the frangibility of the ecosys tem in this area interfered with the exces s reclamation. The farmland exploitation expanded toward borderland because of the unprecedented increase in Chinese population, and the wandering and dropping of the per unit area yield of cultivated land. The gradient force of population in Ordos and peripheries , the land degradation of the Loes s Plateau to the repuls ion for population promoted the reclamation in Ordos . Both nongovernment and government action s promoted the land reclamation in running after interes t; at the same time, the scrambling for interes t between the central and the local, Meng and Han, despots and paupers prevented the course of reclamation. The change in Meng and Han population s tructure and the common interes t relation built in the proces s of reclamation impelled the national amalgamation. The agreement and repuls ion in the national amalgamation proces s not only s trengthened the s tatus of agriculture, but consolidated the s tockbreeding as well. The farming-grazing trans itional land use pattern in Ordos was jus t formed and developed in the balancing proces s of these driving factors and obs tacles.

Rapid fragmentation of natural ecosystems by anthropogenic activity is the important form of natural landscape evolution on earth surface today. In an ecological sense, fragmentation involves dividing up contiguous ecosystems into smaller areas called 'patches'. The development of fragmentation indices mostly parallels the development of landscape ecology. This review examines landscape indices and their usefulness in reflecting the effects of ecosystem fragmentation. Then on the basis of Geographic Information System technology and landscape ecological method, the paper defines the concept of ecological land and measurement indices of fragmentation and urbanization size and spatial configuration, and then quantitatively explores fragmentation properties of ecological land and their relationship with urbanization in the Yangtze River Delta area. Our findings are summarized into three points of view in the end of the paper. (1) Ten fragmentation indices such as MPS, LPI, CAI, NP, PD, FD, SI, MNN, AI and CO based on landscape ecology are firstly handled by selecting the method of Principal Component Analysis of SPSS, and then there are three principal components extracted. The first principal component is defined as size component highly loaded by indices of MPS, LPI, CAI, FD, SI, and the second principal component defined as numeral component highly loaded by NP and PD, and the third component defined as spatial configuration component highly loaded by MNN, AI and CO. According to three principal components the general fragmentation model of ecological land F has been set up. Then the Cluster Analysis is selected to analyze the relationship between F and urbanization level U, urban land ratio UA and Aggregation Index of urban land UI. Lastly there are seven categories areas to be clustered for fragmentation of ecological land in the Yangtze River Delta area, which include two high fragmentation areas (F1, F2), three moderate fragmentation areas (F3, F4, F5) and two low fragmentation areas (F6, F7). (2) Based on the spatial scale of county area and the remote sensing image of a 30-metre resolution, fragmentations of ecological land are the combined representation of urbanization, ecological land properties and the other disturbance resources. The general mode of ecological land fragmentation F has no evident relationship with urbanization level and urban land ratio, whose relative coefficients respectively are 0.303 and 0.432, but has a very clearly negative relationship with urban land Aggregation Index, whose relative coefficient is -0.807. (3) The conclusions of ecological land fragmentation areas and their connectivities with urbanization provide scientific basis for selecting spatial strategies of ecological land safety patterns. At the same time it is very meaningful to guide scientifically urban spatial development direction. The following researches about this topic mainly are from three aspects: exploring influencing factors for ecological land fragmentation by use of high-revolution image data in typical areas; analyzing fragmentation and its connectivity with urbanization based on the buffer zone around urban area; and revealing the dynamical process of ecological land fragmentation caused by urbanization.

Land use changes in different regions have profound spatial connections. In order to reveal the spatial linkage among land use changes of various provinces in China, the conception of land use advantage index is defined firstly referring to regional advantage theory in economics, and a method (Di = Bi ×Si Bmi ×Smi ) is established to measure the index. Based on the index, land use data, and production value published in statistical yearbook, landuse advantage of different regions in China and its evolution between 1996 and 2003 are analyzed. The results show that the provinces with remarkable advantage in built-up land, including Shanghai, Tianjin, Beijing, Jiangsu, Zhejiang, Shandong and Guangdong, strengthen their advantage in built-up land to an extent higher than the national average. Meanwhile, the advantage in agriculture land decreases in most of those provinces. The provinces with relative advantage in agriculture land, including Hebei, Fujian, Hainan and Jilin, strengthen their advantage in agriculture land to an extent exceeding Guangxi, Yunnan, Ningxia, Gansu, Xinjiang and Inner Mongolia. Advantage index increases in forestland or grassland in Henan, Anhui, Hunan, Heilongjiang and Sichuan. Those results imply that there are three evolvement mechanisms in the evolution process of land use advantage. They are polarization mechanism, gradient extrapolation mechanism and ecological blocking mechanism. By exploring the results of the existing researches on land use changes in China, it is further confirmed that the three mechanisms, at the national scale, not only dominate the evolution process of regional land use advantages in China, but also dominate the changes of land use pattern in China.