Professor Shi Yafeng has been studying geographic sciences, especially in geomorphology, glaciology and Quaternary geology for more than five decades, and has been recognized as one of the organizers and leading scientists in interdisciplinary geosciences. In early 1950s, he organizing the studies on geomorphologic regionalization in China. He and his colleagues firstly proposed that three mega regions, i.e. the eastern fluvial landform region, the north western eolian and arid landform region and the Qinghsi Xizang Plateau glacial and cryogenic landform region, and 29 sub regions can be distincted. Since 1958, he had organized the glacier resource expeditions in Qilian and Tianshan mountainous areas. “Report of Expeditions on Modern Glaciers in Qilian Mountains” and “Glaciol and Hydrological Invention in the Urumuqi River, Tianshan” were published under his editorship, which was actually a formal proclamation of the staring of the studies in modern glaciology in China. In 1960s, he entended the glacier expedition into the highest mountains in the the world Himalaya. He firstly gave the systematical explanation about the characteristics of alpine glaciers in low latitudes, and the impacts of rapid uplifting of Himalaya on glacier history. These pioneering works were presented in two books edited by him namely “Scientific Expedition Report of the Oomolungma Peak Area: Modern glaciers and Geomorphology” and “Scientific Expedition Report of the Xixiabangma Peak”. In 1970s, he led a group to study on Batura Glacier in Pakistan and developed the method so called “Fluctuating ice discharge balance” in the prediction of the glacier termination change and sub stantially improved the accuracy of prediction. This was a key challenge for selecting a reasonable rehabilitation scheme for the International Karakorum Highway. In early 1980s, he collaborated with others questioned the long debated “Quaternary Glaciation Hypotheses in Eastern China” and provided extensive and reliable evidence to deny it. These works were reflected in the book “Problems on Quaternary Glacitions and Environments in Eastern China” chiefly edited by Professor Shi Yafeng. This book Marked the breathtaking progress and diversification of glaciology studies of China in late 1980s. Professor Shi also advocated and organized virginal studies in permafrost researches in the Qinghai Xizang Plateau, the debris flow in southwestern China, and hydrology in arid regions in northwestern China in 1960s. These studies contributed remarkably to regional development in China. During 1980s～1990s, Professor Shi Yafeng gained an insight into global change and enthusiastically promoted the research activities in China. He edited and published “Climate and Environments in China during the Holocene Megathermal Period” in which he suggested that precipitation from northern China to central Asia during mid Holocene was more plentiful than that at present due to the enhancement of summer monsoon. These viewpoints have been widely accepted in China. In recent years, Prof. Shi Yafent actively promoted and contributed to the studies on the hazards caused by the sea level reise and urged for more studies to seek countermeasures for the related problems.

Pressed by the northward movement of the Indian Plate, The crust in the Qinghai Xizang (Tibetan) area was uplifted during the Late Cenozoic, which exerts a great influence on East Asian environment. The Qinghai Xizang area was subject to two cycles of uplift and plantation in the Tertiary. The Plateau had raised up to about 2 000 m above sea level during 25 Ma BP～17 Ma BP. This, coupled with favorable continent ocean configuration at that time, may have triggered Asia monsoon, which replaced previously dominated planetary wind system and led to a big environmental shift in China in the Middle-Tertiary because of global cooling and plantation in Xizang. Monsoon decreasing resulted in intensification of the Asian dry climate after 8 Ma BP. The rapid uplift of the Qinghai Xizang Plateau in the Last 3 4 Ma had enhanced in a great deal again the summer monsoon, leading to moistening of the northern Plateau and even North China. On the other hand, the intensification of winter monsoon resulted in deposition of Loess at 2 5 Ma BP in North China. The subsequent tectonic movement of Mid-Pleistocene (0 8 Ma BP～0 5 Ma BP) might raise the Plateau up to about 3 000 m～3 500 m above sea level. This coupled with temperature drop by periodicity shift in the earth orbit, and led the Plateau to enter the iceosphere and form the maximum glaciating with ice cover of over 500 000 km². Because of strengthen of winter monsoon, desert in the northwest China was enlarged and Loess sediment expanded far to the lower reaches of the Yangtze River. Intense tectonic uplift happening in the last 150 ka, led to significant change of many local drainage system in the plateau. Based on 150 ka high resolution climatic records from the Guliya ice core, lake cores at Tianshuihai and Zoige, and loess profile at Linxia, it is found that the climate in MIS 5e was especially warm. The ice core record indicated temperature at 125 ka BP was 5℃ higher than that of today and climate was unstable in MIS 5e in the plateau, which was revealed also by loess and lake core records. The deglaciation since 15 ka BP was characterized by a clear identification of the Younger Dry as event at ac.. 12 ka BP followed by largely fluctuation rising of temperature with its warmest peak at 7 ka BP, corresponded with increase of precipitation, expansion of lakes. After 5 ka BP, temperature decreased again in fluctuation, accompanied with environmental deterioration. Repeated leveling indicates that the Plateau is still in rapid rising at an average rate of 5 8 mm/a in present time.

Tropical cyclones are the most devastating natural disasters causing severe losses of lives and property. One question that merits analysis is whether the frequency, intensity and location of tropical cyclones will change with future global warming. Based on data for 1949～1994 on the frequencies, locations of formation of the tropical cyclones and the monthly average sea surface temperatures ( SST ) in the latitudinal and longitudinal zone of 5°×5° that occurred in the northwestern Pacific, this paper focuses on the temporal and spatial changes of the cyclones and the relationship of cyclone frequency and the SST . The study shows that the frequency of the storms increased unsteadily from the mid 1950s to the late 1960s and again from the mid 1970s to the 1980s. The frequency decreased unsteadily from the late 1960s to the mid 1970s and again in the early and the mid 1980s and the early 1990s. Such changes corresponded roughly with the changes of SST in the waters of 10°N～30°N, 125°E～170°E. Since the 1970s, high SST corresponded with high storm frequency and low SST corresponded with low cyclone frequency. Before the 1970s, the changes were inconsistent which could have been due to the incomplete statistics of storm frequency because of the lack of weather satellite coverage. Spatially, high SST corresponded roughly with high storm frequency in the waters north from of 20°N and west from 140°E, and with low storm frequency in the waters south from 10(N and east from 140°E since the 1970s. Additional correlation calculations indicate that the relationship between SST and cyclone frequencies was nonlinear ( P =ae^b(SST-26)) for the 25 year period 1970～1994 in the waters of 10°N～30°N, 125°E～170°E, and the correlativity of long term change of 5 year running average SST and the frequencies is better than that of their annual change. The relationship between long term change in SST and cyclone frequencies is the best when the change in cyclone frequencies lags behind that of SST for about one year, the calculated parameters being a =10.89,b =0.88 and the correlation coefficient exceeding 0.76. This means that once the SST of northwestern Pacific rises with global warming, the frequency of tropical cyclones will also increase, and the trend of increase will be more obvious in the waters north from 20°N and west from 140°E.

In August 1998, an international symposium on the world deltas was held in New Orleans, Louisiana, USA. This symposium attracted discussion about more than 25 deltas from around the world with emphasis placed on those that are most densely populated and impacted by humans. Keynote papers printed details about the physical, biological, engineering and socioeconomic aspects of six deltas including the Mississippi, Nile, Ganges Brahmaputra, Rhine Meuse, Changjiang and Po. The main purpose of this symposium was to inform scientists, engineers and decision makers about information that is currently available and to provide them a basis for working in such environments.

Soil erosion in the loess hills of Shaanxi and Shanxi Provinces is extremely serious, caused mainly by unreasonable land use. This study is the first attempt to use system dynamics theory to simulate the changes of land use and soil erosion in a large region. The system dynamic model used in this study contains 35 flow level equations, 70 flow rate equations, 185 subsidiary equations, 45 table functions and 196 parameters. The model consists of six sub models, each interacting with the others. Through the feedback loop they are connected with all variables and parameters. The functioning of the model depends on feedback mechanisms. The model was run on a PC computer with the software of PD (Professional DYNAMO) Plus. The period of simulation is from 1980 to 2040. The results were tested using collected data for 1980～1990. The efficiency of the model is 91 5%. The model was used to simulate the dynamic changes in soil erosion in three different types of land use management programs through 2040: intensified management, traditional management, and steady development management. The simulation results of the three types of programs were analyzed by a comprehensive evaluation method using fuzzy and multi objective targets. The main conclusions of the study are: First, in the loess hills, land use for farming, forestry and animal husbandry changes in a nonlinear dynamic fashion and they affect one another. Thus the adjustment for land use structure should be made gradually to maintain dynamic balance. Different types of land use should be balanced. Second, land use pattern is the key factor affecting soil erosion. Reasonable land use and effective management can reduce erosion to a minimum. Otherwise soil erosion will intensify and the overall land use benefits will decline. Third, the levels of soil erosion in the region vary from slight, moderate to severe, and the rate of erosion is related to land use pattern and the intensity of land use management. There has been an increase in slight and moderate erosion in the recently developed land due to intensified human activities in the contemporary period. Soil erosion can be reduced to a minimum but it can never been stopped completely. Fourth, intensified management is the best approach for a balanced development of farming, forestry and animal husbandry and for sound ecological equilibrium in the region.

In the flood prevention system of the Huaihe River basin, the areas of flood diversion region play an important role. They protect the vast regions in the middle reaches of the Huaihe River, reduce the threat of flooding, and divert the effects of flood peak in the lowed part. The gradient of the river channel in the middle reach is very small, averaging 0 5‰, which can cause flooding that frequently threat urbane and villages on the two sides of the Huaihe River. The flood diversion region are economically backward and poor, and in some placer farming is possible only once in 1～2 years. Thus such area are a major problem in the Huaihe River basin. At present, there are 4 flood water retaining areas and 18 flood water transportation areas in the middle reaches of the river, which are total area of 3080 hm2, and a population of 148 million. These 22 areas occupy a transition zone between two types of monsoon climates, classified by abrupt changes of the cold and warm seasons and my alternation occurrences of drought and waterlogging. In these areas, about 60% of the farmland are low yielding producing only 3 700 kg/a·hm²～4 500 kg/a·hm² of grain. Farming has suffered from prolonged backward wheat soybean rotation, insects, weeds, and waterlogging, to improve gain yield, a system of wheat rice multicropping is proposed to reach the target of more than 6 700 kg/a·hm². The wheat selected for cultivation is Bie-Nong No.6. This strain is characterized by high yield, short stacks, early ripening and disease resistance. The yield of it can be as high as 8 000 kg/a·hm², the rice seed selected is No.180, an early ripening and xerophilous strain can be planted directly in the wheat stubble and it does not require frequent irrigation. It is also a high quality, with average yield of 6 000 kg/a·hm² and a maximum yield of 7 500 kg/a·hm². For the dry regions of the Huaihe Rivers where water saving farming is essential, the introduction of this rice strain is indeed an innovation.

Wind erosion, aeolian sand transportation and deposition are complex and closely linked aeolian processes. An aeolian region can be looked upon as an open system that exchanges aeolian sand with its surrounding regions. A region’s aeolian erosion and deposition play decisive roles in the development of the region’s aeolian landform. More specifically, the relationship between importing and exporting aeolian sand, or the balance between boundary mass input and output of sand, determines the resultant aeolian landform. The rate of regional erosion and deposition or the intensity of erosion can be expressed as: Re=We/S, where We is the balance between boundary mass input and output during a certain period of time and S is the areal size of the region. Field observation was conducted in the Shanxi-Shaanxi-Nei Monggol border area and different formulas between wind velocity and the transport rate of sand drift on different types of land surface were established. The rate of transport increases rapidly with increasing surface sand mobility and wind velocity. Three specific factors that have decisive effects on sand transportation are identified as wind velocity ( V ), blowing time ( T ) and wind direction. An analysis of the data on erosive wind collected by local weather stations determined the total annual quantity of sand transport flux on different types of surface. Vector analysis and charting ascertained that the prevailing direction of sand transport flux is mainly from the northwest to the southeast, with varying azimuth angles of between 288 7° to 303 6°, revealing clearly that the basic direction of aeolian sand movement and encroachment in the region is southeastward. Based on the interpretation of boundary land surface types from TM imagery, the boundary balance, quantity and intensity of aeolian erosion and deposition in different counties in the study area are calculated. The results show that the entire study area, which lies in the transitional belt between sand desert and loess plateau, is an aeolian erosion region. The region’s annual quantity of aeolian sand erosion is 109 million tons and the average intensity of sand erosion is about 160 t/a·km². Erosion increases from the sandy loessial region in the southeast to the desert region in the northwest as rainfall decreases, wind velocity increases, and vegetation cover declines. Thus the quantity of regional aeolian erosion increases with increasing interaction between the erosive agents of wind and water.

Unlike the imperial rule before 1840 and the socialist period after 1949, Beijing during the first 30 years of the twentieth century experienced colonialism and internal social change which caused a corresponding change in the city’s social areas. Based on the city’s population data and several social surveys conducted after 1906 and using notions of classic urban ecological theory and the preindustrial city, this paper examines the internal spatial structure of Beijing during the late Qing and the early Republican period. With the erosion of the Qing authority, the decline of the Manchu noblemen, and the rise of modern citizenship and new professionals in the commercializing society, dramatic social transformation taking place in Beijing caused major changes in the city’s traditional internal spatial structure. Three types of social spatial patterns emerged that bear some resemblance to the characteristics of the preindustrial city and the urban structure in Latin America. First, social spatial differentiation took place. While the elite were concentrated in the central zone where they enjoyed spacious housing and good urban facilities, the urban poor were compelled by rents to move to the periphery where the living conditions were much inferior. Second, family size and family economic condition were positively correlated, with the city center having more large families and the periphery more small families. Relatives, concubines and servants living in the same household were all counted as members of the same household. Third, in terms of ethnic distribution, the declining Manchu nobles and numerous Bannermen had to sell their houses to make a living and to leave their comfortable princely estates and the Inner City for the fringe areas, especially in the northern part of the city. The Hui had a lowly social status who were clustered in the Ox Street and the Flower Market areas, both in the urban fringe. Since the Qing established its capital in Beijing, the Inner City was known as the Tartar City because it was occupied by the Bannermen. Since the downfall of the Qing, social segregation and residential integration appeared, the patterns of which were determined mainly by the residents’ economic status. In addition, international forces further complicated the patterns of social spatial differentiation. Typical examples include the emergence of the new commercial center at Wangfujing, the formation of the high-class residential area in Dongdan, the appearance of a banking district in Xijiaominxiang, and the presence of the Legation Quarter in the adjacent Dongjiaominxiang area.

The bottom of Poyang Lake, China’s largest fresh water lake, slopes northward toward the Changjiang (Yangzi) River into which the lake drains. The lake’s outlet is located at the lake’s northern end at Hukou, Jiangxi. In earlier times, the river had been further north but it shifted its course southward, eventually reaching Hukou. This caused the river water to flow to the lake basin, forming natural levees and sand bars which obstructed the lake’s discharge into the river and affected the lake’s later development. However, it is not clear when did the river swing southward to reach Hukou, when did the river water begin to flow to the lake and what was the intensity of the lake’s back flow. This preliminary study is concerned with these questions which thus far have escaped the attention of researchers. High resolution data are obtained from Hole ZK 2, which is located at the outlet and which sometimes submerges during the dry season. The maximum water level at the site reaches 14 m during the flood season. The data indicate that lacustrine facies has dominated the modern sedimentary environment. The earliest date when the river reached Hukou was determined by searching for sediments that would provide clues to river lake interactions. At the depth of 7 48 m, the hole’s profile shows a boundary. Dated by 14 C to approximately 2 360 a BP, the sedimentary environment immediately above and below the boundary line was one of open lacustrine, but the clay minerals, heavy minerals, granularity, magnetic susceptibility and colors are all different, indicating different sources. The location of the hole suggests that the old sediment source was the ancient Ganjiang River (or the river system of the Poyang Basin), and sediment source changed only when the Changjiang water reached the site, bring different sediment. The age of the boundary is determined to be about 2 360 a BP when the Changjiang shifted southward and reached modern Hukou. This date is the date when the lake first experienced back flow from the Changjiang. In the fine sediment in the outlet waterway at Hukou, those from the ancient Ganjiang River have low Xlf while those from the Changjiang have high Xlf. When the back flow intensity is strong, the proportion of sediment from the Ganjiang becomes less while that from the Changjiang increases with higher Xlf, and vice versa. Moreover, the 20 year running average of the back flow for 1950～1984 matches perfectly with the Xlf curve, providing further evidence that Xlf may be used as a proxy indicator of the intensity of the back flow. The Xlf curve indicates that the history of the lake’s back flow can be divided into three stages. During 2 360 a BP～1 550 a BP, the intensity was weaker than that of today. From 1 550 a BP to 880 a BP, the intensity increased in a wavy pattern, with the greatest intensity occurring around 880 a BP. From 880 BP to the present, the intensity has fluctuated violently in six complete cycles, each lasting about 115 years.

The sunrise and sunset hour angles on a slope are determined by considering the sun’s horizontal hour angles and the non horizontal hour angles of the slope. Such coordinated relationship varies with latitude φ, slope angle α, slope azimuth β, and solar declination δ . When latitude φ and slope azimuth β meet the condition of |sin β |cos φ >sin23 45°, the coordination of sunrise and sunset hour angles has only one form for any solar declination δ . But when |sin β |cosφ≤sin 23 45°, the coordination will vary with slope azimuth β . For any given solar declination δ and latitude φ, there are two slope azimuths β¹ and β² (where | β¹ |=arcsin(sin δ /cos φ ), | β² |=π-| β¹ |). Between β¹ and β² and beyond β¹ and β², the coordination relationships are different. This detailed study provides a way to determine the concrete form of coordination for any slope azimuth. Based on the coordination of sunrise and sunset hour angles on a slope, we deduce a series of formulas that are used to calculate the daily duration of sunshine and the daily extraterrestrial solar radiation on the wall surface. From the results of calculation, we obtain the distribution of daily duration of sunshine and daily extraterrestrial solar radiation with slope azimuth. We find that there is a pair of slope azimuths, where the daily duration of sunshine in summer half of the year is equal to that in the winter half when the absolute values of solar declination are equal. These slope azimuths are called “equivalent duration of sunshine azimuth (EDSA)”. There is also a pair of slope azimuths where the daily extraterrestrial solar radiation in the summer half is equal to that in the winter half at the same time. These slope azimuths are called “equivalent insulation azimuth (EIA)”. Together, EDSA and EIA are called “equivalent solar radiation azimuth (ESRA)”. ESRA is a function of latitude (only. The value of EDSA can be expressed as | β |=arccos(tg φ ). The value of EIA can be expressed as | β |=arctg(cos²φ /sin φ · Y_φ ), where Y_φ is a solution of the equation y +tg² φ ·tg y =0.

Massive influx of foreign direct investment (FDI) since her opening to the outside world has made China the largest FDI receiving nation in the developing world. The distribution of FDI in China is highly uneven because of complex regional differences. On the whole, FDI has been concentrated in the coastal provinces and along the Changjiang River, namely in the provinces of Guangdong, Jiangsu, Fujian, Shanghai, Shandong, Liaoning, Zhejiang and Hainan. From the 1980s to the 1990s, there was a major shift from the south to the north and the middle reaches of the Changjiang River. This regional pattern is explained by coefficients of elasticity. Using SAS, this paper calculates the relative importance of three sets of factors affecting the distribution of FDI. The traditional factors used include comparative labor cost, market capability and capital productivity. The aggregate factors include infrastructure, the level of economic development and cumulative FDI. And the risk factors include cumulative FDI, the degree of openness, and geo economic factors. The following conclusions were reached. (1) The regional patterns of FDI reflect the existing unbalanced patterns of development. (2) The key factors influencing the patterns of FDI are information cost, market capability and risk factors. (3) A change in the selected factors affected the comparative advantages of regions, which in turn affected the regional patterns of FDI. (4) FDI has moved beyond the early exploratory stage of seeking quick profit to the mature stage aimed at market expansion. (5) The basic patterns of FDI distribution are difficult to change because of the inherent differences among regions.

Contemporary China’s urbanization has evolved along two lines: the growth of cities following urban economic development and population concentration, and rural urbanization based on the growth of towns in rural areas. The latter,conceptualized as “urbanization form below” has been propelled by the rise of rural enterprises, resulting in rapid rural employment transformation and the growth of towns. Rural urbanization since the reforms has developed rapidly. In 1995, the share of the designated towns’ nonagricultural population accounted for 9 5% of the nation’s total population, an increase of 5 3 percentage points over that of 1978. This nonagricultural population was 12 5% of the total rural population which was 7 5 percentage points higher than 1978. It constituted 42 5% of the total nonagricultural population of all cities and towns for the nation, rapidly catching up with the share of urbanization contributed by cities. Urbanization from below has been driven by the activities of rural governments and peasants. Three stages of development can be recognized. After the initial period of growth during 1978～1983, it experienced rapid take off during 1984～1988. It entered a period of expansion since 1990. However, the level of rural urbanization has lagged far behind the level of rural nonagriculturalization as well as the nation’s urbanization level in general. As rural enterprises will remain the mainstay of the China’s rural economy for a long period of time, they will continue to cause rural labor transformation and rural urbanization. But with increasing economic liberalization in urban economy and industries moving out the cities, the fact meaning of “urbanization from below” will have to be broadened to include all domestic and overseas forces affecting town growth that are non state in nature.

The method involves the following steps. (1) The study area is first divided into several regions, each with relatively similar features such as soil, topography, ground water level, location attributes and income level. (2) In each region, a number of reference plots are identified, each bordering several other plots. The characteristics of a reference plot such as its physical features, land management, major crops, crop yield, plot size and shape should typify those of the adjacent plots. (3) Calculate the price of each reference plot based on its net annual income using the equation T=(O-I)/Sr where T is the price of a reference plot, O is its gross annual income, I is total annual expenses, S is its areal size, and r is the rate of return. (4) Establish a list of factors affecting land price in each region and determine their weights based on the Delphi measurement method. (5) Rank the factors and assign appropriate points to them. (6) In a region, delineate valuation land plots based on physical features, plot location, field infrastructure, major crops and crop yield. (7) Calculate the total points for each valuation land plot using the equation V = F _i * W _i where V is total points of the plot, F i is the points assigned to each factor affecting the price of the plot, and W i is the weights of the factors. (8) Calculate the average value of land price per point in a region using the equation K=1/n∑ni=1 P_i/F_i where K is the average value of land price per point, P i is the value of price of the reference plot, F i is the total points of the reference plot, and n is the number of reference plots in a region. (9) Calculate the price for all of the valuation land plots in a region using the equation P=V*k where P is price of a valuation land plot, V is the total points of the plot, and K is the average price per point of the region in which the plot is located. (10) Check the calculated land price through a comparison between the price obtained by using the approach described in this paper with the price calculated by income approach for the same sampling plot. (11) Rank the calculated land prices for all plots in the study area according to the limitation values of each class that are obtained from a statistical analysis of land price distribution against the area of all land plots in the study area. With the assistance of GIS technology, the method was used to appraise the price of cultivated land in Yudai Township, Liuhe County, Nanjing City. The results show that the approach is acceptable and has the advantages of being simple and practical. However, several points should be noted. The number of cultivated land regions to be delineated from a study area and the selection of the factors affecting land price must be in line with the local conditions. Also, the effects of such factors as variation in land management and income on land price must be considered when indices of income and expenses are used to calculate the net annual income. Moreover, to ensure the accuracy of land price appraisal, it is necessary to select an appropriate rate of return which is in line with the level of economic development of the study area.

The Siberian High is one of the most important circulation systems influencing the climate of eastern Asia in winter. In this study, the intensity index of the Siberian High is defined as I=? 苮72n=1P_nδ_ncosφ_n? 苮72n=1δ_n cosφ_n where P_n is the sea level pressure at grid n , φ_n is latitude, δ_n =1 when P_n ≥1 028 hPa, and δ_n =0 when P_n <1 028 hPa. This index indicates the level of anomaly from the average atmospheric conditions in the core region of the Siberian High. A long time series of the intensity index for the last hundred years was established on the basis of monthly mean sea level pressure data for the northern hemisphere obtained from CRU/UEA. Power spectrum analysis reveals that, for about four decades, the index showed decadal variability. In the 1960s, the Siberian High’s intensity was the highest. During the late 1980s and the early 1990s, the intensity was the lowest. The intensity was strongly correlated with the winter temperature of East Asia. A great deal of the temperature variance in the region 110°E～140°E and north of Bangladesh can be explained by the intensity index. Our calculation of the temperature data for 65 winters shows that 43 6 percent of the variance of North China’s winter mean temperature can be accounted for by the intensity of the Siberian High. The possible impact of global warming on the variability of the High was also investigated using the atmosphere ocean coupled general circulation model (HADCM2). The result of CO² increase at the rate of 1% per year on global warming was compared with the result of control integration. It was found that in the last hundred years the impact of global warming on the Siberian High was not evident and that the High’s own natural variability played a more important role. Continuous warming in the future will significantly reduce the intensity of the system. However, the change will not be linear. With increasing Co², variability at 5～6 year inter annual scale can be expected to become more prevalent.

Wince 1991, some researchers have proposed a hypothesis claiming that the continental shelves of China were once a desert environment during the low sea level period of the last glacial maxima in late Pleistocene. This hypothesis has been used to interpret the origin and formation of the remaining deposits in the shelf regions. The hypothesis’ main evidence includes the disintegration of the integrated marine stratum, extensive mixed deposits, endless erosional surface texture of repose angle, buried dune groups and ventifacts, etc.. We argue that although it is logical to use such evidence for large scale analysis of the geological environment of China’s continental shelves during the late glacial maxima, it is inadequate to prove the validity of the hypothesis. Moreover, the hypothesis does not consider the impact of the post glacial marine transgression on the aeolian san which might have been on the continental shelves. The paper explores four specific issues. (1) Lacking biological evidence, the criteria used to analyze aeolian sand facies (ancient sand dunes) and desert environment (ancient desert) should be comprehensive and based on multiple indices. The selected indices should be able to diagnose the processes and environmental features. However, the hypothesis’ indices such as the “buried sand dune groups” are mostly multiorigin indices that are not unique to desert environment. (2) Although ventifacts (aeolian gravel) are an indicator of intensive aeolian process, they do not surely represent arid desert environment. A comparison of the characteristics of the egometry and surface textures of the gravel indicates that the gravel from the seabed of the South Huanghai Sea differ significantly from the typical ventifacts of the modern desert environment. Hence they should not be regarded as one and the same thing. (3) Under the dry and cold climate conditions during the glacial period, aeolian sands (dunes )were difficult to be cemented and diagenized, and they may be easily destructed or regormed by post glacial marine transgression. Therefore, such sediment structures as “texture of repose angle” and such sand dune morphologies as the “buried sand groups” are not likely to be preserved extensively. In conclusion, the remaining deposits on the continental shelves of China should not be regarded as ancient desert deposition.

The lower reach of the Jinsha River, the study area of this paper, lie in the Hengduan Mountains in the transitional zone between the Qinghai Tibetan Plateau and the Yuannan Guizhou Plateau. Stretching for 786 km from Panzhihua to Yibin, the topography of the area rises and falls sharply. The area’s geology is complex and tectonic movements new and active. The area has rich water resources. The Ministry of Electric Power has planned 4 large scale hydroelectric stations with an installed capacity of 36.9 million kW. The area has experienced repeated landslides and rock mud debris flows that have seriously affected its engineering work and economic development. Remote sensing interpretation and site verification reveal that the area has 400 landslides with volumes >10 8 m³. These landslides are distributed mainly along the middle section of channel bank and in the river’s tributaries where the elevations range between 400 m～ 3 200 m. The forms of the landslides are various and complex. Those with high rushing speeds and large scale clastics are the most damaging to engineering work and the environment. At present, about 70% of the landslides are partially active. The study area’s drainage basin is >0.2 km² in size. In the first grade tributaries of the river, there are 438 large scale debris flows whose deposit fans are 0.01 km²～3.1 km². These debris flows are distributed mainly in the study area’s middle section where the elevation is 500 m～4 000 m. The areal size of 80% of the debris flows’ drainage basins is 1 km²～50 km². The study area has 299 viscid debris flows, accounting for 68% of the total, with only 11% of the total being dilutive. The numbers of gullies where debris flows of extra large, medium and small scales may occur are, respectively, 6, 136, 183 and 103. Numerous new debris flows are being developed in the area. Landslides and debris flows have caused serious damages to the area and strongly impacted the environment. The massive volume of silt the river carries from the area posses a major problem to the Three Gorges Dam downstream.

Suburbs are a transitional zone between city and country with the characteristics of both. Peripheral to and administered by a city, suburbs have nonetheless always been an important part of cities. Historically they represent an important stage in the development of cities. During the Qing Dynasty (1644～1911), the suburbs of Beijing were known as chengshu (literally, “belonging to the city”) whose administrative boundaries were demarcated during the mid Yongzheng reign (1723～1735). But because of inconsistent historical records, scholarly views regarding the locations of the boundaries have long differed. Using information and maps on Beijing’s suburbs recorded in the local gazetteers compiled during the Qing, this study first ascertains that Beijing’s chengshu were indeed its suburbs. After raising questions concerning the areal extent and the boundaries of chengshu recorded in documents, the approximate outer boundary of Beijing’s chengshu is identified by relying on the data from the imperial edicts of Emperor Yongzheng. The edicts provide information on the city’s garrison and police district that corresponded to the city’s suburban area. This boundary was stable until the Republican period (1911～1949). The significance and function of the boundary are then discussed. Whereas China’s administrative systems and regional divisions appeared quite early in history, regional divisions at different administrative levels were usually controlled by long and customary traditions without any clearly defined legal boundaries. For example, during the Sui (581～618) and the Tang (618～907) Dynasties, administrative regions known as sizhi badao (literally, “four reaches and eight arrivals”, referring to the four cardinal directions plus the additional subdirections of northeast, southeast, southwest and northwest), were merely very general administrative areas of different levels without any rigid administrative boundaries. The Qing more clearly demarcated the boundaries at different administrative levels to maintain border peace and for better social control, including the delimitation of Beijing’s suburban boundaries. In the history of urban growth in China and elsewhere, this development is important. After their boundary had been delimited, the suburbs of Beijing provided space for the Manchu Bannermen to concentrate in who had come from the city, for the construction of imperial gardens and villas, and for the development of handicraft industry, trade and services.

The pathogeny of Kaschin Beck Disease and keshan Disease have not known yet by now. Some scholars think that it is resulted from the biological toxin and some think that the chemigeological factor is their pathogeny. Author agree the later view. By co operative study with the medical workers for several ten years, we found that the geographical distribution of Kaschin Beck Disease and Keshan Disease coincided with the distribution of low selenium zone in China. So that the pathogenical doctrine of selenium deficiency in environment about these two endemic diseases was suggested. It was thought that Kaschin Beck Disease and Keshan Disease were actually environmental diseases. The prevention and cure of them therefore should started with improving local environment for the permanent control of these two diseases. Based on this consideration, this study increased selenium intake of local residents by improving the cycle of nutrional material such as selenium in food chain, and finally control these two endemic diseases. The serious affected area of Kaschin Beck Disease and Keshan Disease in loess plateau, Binxian County, Shaanxi province was selected for the example study on spraying Se on leaf of wheat, vegetables and adding Se into diet (bread) to control these two diseases from 1992 to 1995. The result of the example study shown that the Se concentration in wheat sprayed selenium raised to 0 081 μg/g, it was obviously higher than that (0 009 μg/g) of wheat unsprayed selenium. The Se in diet from experimental village reached 0 076 μg/g and higher than that (0 014 μg/g) of control village. The average Se intake of residents in experimental village raised from 13 40 g/person. day to 60 39 μg/person.day. The X ray positive rate of Kaschin Beck Disease in experimental village came down from 43 33% to 6 45% and came up to the natioal control standard of Kaschin Beck Diease. This indicated that it was obvious effective to prevent Kaschin Beck Disease and Keshan Disease using trace element selenium, and that selenium really relates to the prevalence of these two diseases.

Louisiana has lost in excess of 4 000 km² of its coastal wetlands in the last 70 years. Much of this loss is directly attributable to man’s activities. A major coastal restoration effort is needed or many coastal communities will have to be abandoned. Recognizing the seriousness of the wetland loss problem, several federal and state programs were initiated in 1991 to attempt to address the problem. However, it was imperative that the state develop a long term comprehensive coastal wetland creation/restoration plan to aid the wetland restoration programs. The plan herein described was first presented to these agencies in 1993. Various aspects are now being implemented. The plan promotes diverting sediment and fresh water from the Mississippi River. It attempts to simulate natural delta growth processes by creating river diversions, re establishing former distributaries and restoring restoration of Louisiana’s barrier islands.

The study of land resources in China and abroad since the 1970s has evolved into an important and comprehensive field of applied research in earth sciences. This field encompasses the study of land types, land assessment, land productivity potential, land capacity to carry population, land use and land cover, land planning, and land information system, etc. During the 1990s, a key focus of study has been the changes in land cover. More recently, in an attempt to better understand the changes in land quality, an international collaborative project has been initiated to develop land quality indicator systems. Participating agencies include the World Band, United Nations Food and Agriculture Organization, United Nations Environment Program, United Nations Development Program, and the Ministry of Agriculture and Agri Food of Canada. This paper introduces the this project in the hope that it will arouse the interests of Chinese scholars at a time when the nation faces extremely serious problems of land degradation, especially in environmentally fragile regions. The project is based on the theoretical framework of pressure state response to measure the pressure on land resources, the impact of the pressure on land quality, and societal responses to the pressure. Different land quality indicators have been established through the analysis of land quality problems in typical agricultural ecological zones of the world in the tropical and subtropical regions, including the humid tropical rainforest region, seasonally dry tropical region, Savanna region, subtropical region, semi arid region, and arid region. In each region, the indicator system takes into consideration such variables as land quality, land use, agricultural production, population, resources and societal responses. On the other hand, other indicator systems are designed to evaluate a specific problem, such as land degradation. Such indicator systems can be disaggregated into pressure indicators, land condition indicators and societal response indicators. Although designed mainly for use at the regional and national levels, some systems have been developed for use by local places and farms.

According to the spatial and temporal distribution of the brown reddish soil, red earth and relicular mottled red clay, the locations change of subtropical and tropical zones since late Pliocene in China are discussed in this paper. From the change of the limit of natural zone, the climatic variation and its amplitude in southern China can be deduced respectively. The north limits of the northern part of subtropical zone in N₂, Q₁², Q₂¹ and Q 3 were 41°～42°N, 46°N, 42°～45°N and 35°～40°N, 30°～33°N for the southern part of subtropical zone and 25°～32°N, 30°～32°N, 25°～30°N and 24°～26°N for tropical zone. Comparing with the present limit of the northern part of subtropical zone was of the largest amplitude moving northward, but the other two zones were of a similar moving amplitude during the four periods. The limits of the three kinds of red earth during Q₁² period are latitude of 12°～13°, 8°～12° and 5°, that showing the most warm period of Quaternary in China. Comparing with the climatic condition at present, a rising amplitude of annual temperature in 1℃～4℃ during the four warm periods in general is predicted. North China and the area of the middle lower reach of Changjiang River occured obvious environmental change especially during the period of Q₁² with a variation of two natural zones, and was a confrontation state each other between the southern part of subtropical and tropical zone in central China during the four periods, but as tropical zone always in south China. The north limit of tropical zone had moved to 30°N during three periods, so there was tropical for a long term in south east China causing extensive development of the relicular mottled red clay especially during the period of Q₂¹. Until late Pleistocene central and east China is of a stable zone of the southern part of subtropical, thus a corresponding scope of the present tropical and red earth distribution is formed finally.

This paper reveals the mechanism of population migration along the coastal urban regions in China. This is the first large scale survey on floating population ever conducted in large and medium cities including Beijing, Tianjin, Tangshan, Langfang, Nanjing and Kunshan. The paper first reviews the intellectual foundation of demographic studies on migration. With the survey data and further analysis, it examines the characteristics of population migration in Chinese cities. China is now undergoing a large scale of rural migration into the urban domain with mechanistic increases in the number of floating population. The motive for migration is predominately economic. Family type migration is getting more popular than individual migration. From an economic perspective, low value added, semi skilled jobs are taken up by the first time migrants, and there is also the corresponding shift to service and managerial type of work for the local residents. Chain type migration with upward movement on the job ladder is observed for the second time migrants. From a social viewpoint, jobs for migrants are mostly untenured and without welfare provision as compared with those available for the urban dwellers. Migrations in cities coalesced together to form cliques and syndicates. This has resulted in a more segregated urban morphology.

The flora of North China (excluding Qingling Mts.) includes about 151 families, 914 genera, and 3 925 species of seed plants. A comparison of paleobotanical material and present day flora provided information on their evolution and sources. The evolution of North China’s flora and vegetation since the Cretaceous may be divided into the following five periods: 1. In the Early Cretaceous, floras were dominated by ferns and gymnosperms comprised mainly of Lygodiaceae, Ephedraceae, Cycadaceae, Ginkgoaceae and the genus Brachyphyllum of coniferae, etc. A few of the angiosperms that appeared initially are Tricolpollenites, Quercoitites type of family Fagaceae and Salixpollenites of family Saliaceae, etc. The vegetation was arid tropical subtropical in aspect. 2. During the Late Cretaceous, angiosperms increased greatly, becoming about equal in number to ferns and gymnosperms. They are mainly composed of Exesipollenites, Gothanipollis, Plicapollis , etc. and formed a subtropical mixed forest and scrub vegetation. 3. In the Paleogene, floras were dominated by angiosperms and hydroherbs such as Potamogeton . The earlier ferns and gymnosperms were greatly depleted or became extinct. However, the coniferae developed including the families Pinaceae, Taxodiaceae, and the genus Podocarpus . The vegetation was a subtropical forest landscape. 4. In the Neogene, angiosperms increasingly flourished with a great number of Amentiferae, temperate elements, and herbs. The differentiation between the northern and southern parts of North China became obvious. The vegetation is a warm temperate mixed needle and broad leaved forest and deciduous broad leaved forest in the north, and a warm temperate subtropical evergreen broad leaved forest in the south. The xerophilous deciduous broad leaved forest and steppe appeared in the late Neogene. By then, the present flora and vegetation were basically formed. 5. In the Quaternary, climatic fluctuations induced plants to migrate and reassemble resulting in warm temperate forest and open forest steppe alternating during glacial and interglacial periods and in the increase in herbs and the expansion of steppe. After the late glacial, the flora and vegetation were similar to that of the present because of a warming climate. The sources of the present flora of north China may be recognized as taking four paths. It is mainly of autochthonous origin with evolution from the Paleogene to the present like various ancient woodplants, temperate and subtropical or tropical elements. Herbs occurred mostly since the Neogene. Besides, many species are from subtropical tropical regions of southwest and south China, especially from Henduan Mts. Some are from the Central Asia Mediterranean and Europe Asia steppe as well as northeast Asia and the Siberian temperate cold area due to the expansion of steppe as the climate became cold during the Neogene and the Pleistocene.

In collaboration with a coastal engineering project, an investigation into the structures and spatial distribution of the seabed sedimentary stratification of the Shenzhen Bay has been undertaken. Shenzhen Bay is located between Shenzhen City and Hong Kong and in the east coast of the Pear R. Estuary, with an area of 112 km 2. The seabed topography of the bay is rather flat, with most water depth less than 5 m. The bay is developed along a tectonic fault depression, therefore, the configuration of the coast is controlled mainly by fault tectonics. The bedrock of the Shenzhen Bay is composed of granite of Yanshan Period. The surface of the bedrock had undergone intensive weathering with a weathering layer thickness of about 5 m～10 m; some of them have become laterite. The climate of the region is tropical monsoon. East wind is dominant throughout a year. Typhoon affects the area often, with a frequency of 2 times each year, normally in September and October. The tides here are irregular and semi diurnal, with tidal ranges of 3 4 m and 1 36 m for maximum and mean tides, respectively. Flood tidal current velocity is almost equal to that of ebb, at about 70 cm/s. The sediments in the Shenzhen Bay are mainly from the offshore, with marine material taking about 95%; this implies that the region is an estuarine embossment with low sedimentation rates. The fieldwork undertaken includes two parts. First part is geomorphologic and Quaternary survey in the bay and along the coastline, with the objectives of investigating the initial geomorphologic characteristics and the outlines of environmental evolution. The second part if the survey of the seabed stratification using GPS(model 1008-586), Geopulse(5210A) and chartered steel hull ship which has a load capacity of 200 tons and light draght of 0.5 m, along the grid lines in NS and WE. The seismic profile records were interpreted, on the bases of comprehensive analysis of seabed geomorphology and Quaternary Geology, and core data. Five sedimentary units were found in the seabed sedimentary stratification in the Shenzhen Bay. (1) Marine muddy layer: it is the fine materials deposited in the quiet water environment, with well developed horizontal bedding structures; (2) Fluvial sub clay layer: it is a kind of rather solid sediments, with milky white and grayish green strips, and originated from flood plain sediments. It has been deposited in water for quite a long time, and then emerged out of water to be in the oxidizing and leaching environment; (3) Fluvial sand layer: it is composed mainly of pure coarse sand, with a mean grain diameter of 0 5 mm～1 0 mm. The sands are mostly white, with clean surface and many fresh, sharp edges; (4) Layer of mixed sand and clay: it is the alluvial/diluvial sediments on the surface of the weathering crust; (5) Layer of weathered rock: it is the transit layer between the deposited sediments and fresh granite. All the data were input into computer, and processed using GIS software ARC/INFO. The outputs are various data sets and images about the spatial distribution of seabed sedimentary stratification. The study gives an good example to show that the application of geophysical survey and GIS in coastal water has the characteristics of high speed, high resolution, and illustrative displaying effects, and is promised to have bright future in the studies on regional geomorphology and Quaternary Geology.

It has great significance, both theorefically and practically, to study the change of the spatial structure of port system. In the West, Geographers interested in this problem have focused on two aspects the competition of hinterlands or forelands among ports and the degree of traffic concentration in port system. But scholars at domestic have not study this problem from view of change of cargo flow up to now. The main purpose of this paper is to study the dynamic features, formation causes and change tendencies of the spatial structure of the Changjiang River port system. Based on annual cargo throughout of 24 principal ports on the Changjing River from 1952 to 1995, author calculates Gini coefficients of the port system in eight base years. Gini coeffient in 1952 as only 0.209 5 that was rather low, then there appeared an steady tendency to rise. In 1980, Gini coefficient reached its maximum which was 0.528 0. But an evident drop took place at the early of 1980’s. According to above calculating results, author holds that there are four dynamic features of spatial structure of the port system, (1) The spatial structure of the port system has become more concentration from 1952 to 1995 by and large; (2) It is based on much deconcentrated port distributions that the spatial structure tends concentration; (3) There is obvious difference for the change of spatial stracture in 4 phases; (4) An evident deconcentration process took place at the early of 1980’s. A systematic analysis has been carried out on the formation causes of the change of the spatial structure of Changjiang River port system. The paper holds that the following factors have profound implications on its evolution: economies of scale, uneven economic development along the Changjiang River, the change of the prot system’s technological conditions, the development of transport network, and the transformation of the nation’s economic policies. Compared with some port systems in the West, the Changjiang River port system is still in the primary fromation stage. The present spatial state of the port system is unfavourable to elaboration of the whole benefit itself. In the author’s opinion, from now on the basic trend will be continued toward the concentration in the spatial structure but the developing velocity will be slowed down. Nanjing, Wuhan and Chongqing will grow faster than the other ports.

Land use changes may influence a variety of natural phenomena and ecological processes, including soil conditions, water runoff, soil erosion and biodiversity. Irrational land use is one of the main reasons for the soil erosion and nutrient lose in the loess hilly area. The Yangjuangou catchment in the Loess Plateau of China, with typical loess hill and gully topography, was selected as the study area. The study focus on the affects of land use changes on soil erosion, the distribution of soil nutrient and soil moisture from catchment, and land use type at three spatial scales. Aerial photography interpretation and field survey mapping were used to produce land use maps in 1984 and 1996. GIS was used for data storage, analysis and display from a comparison of land use areal changes in 1984 and 1996. It was determined that the area of forest and grassland increased 42% and 5% respectively and slope farmland decreased 43%. Land use changes result in a decrease of soil erosion by 24%. Three types of typical land use structure during 15 years in the loess hill slope were selected in order to study the effect of land use structure on the distribution of soil nutrients and soil moisture. From the bottom to top of hills, the patterns of land use types are: grassland slope farmland forest, slope farmland grassland forest and slope farmland forest grassland. By measuring the contents of the total N, total P, available N, available P, organic matter of soil and soil moisture in 0 cm～70 cm depth, the results show that the land use structure type of slope farmland grassland forest has high contents of soil nutrients and low antecedent soil moisture. This indicates that this land use structure has a better capacity for retaining soil nutrient and a high efficiency for soil conservation. The analysis of soil nutrient and soil moisture in different land use types showed that the content of soil nutrient are: forest>grassland>slope farmland, while the content of soil water are: forest

Peat forming and accumulation is one of main approaches of natural ecosystems on surface of the Earth to store the solar energy. A series of simultaneous field spot fixed experimental study were conducted at 4 wetlands respectively in different environment in Northeast China and North China from 1993 to 1996. Based on the experiments, it is researched that the relation of decomposition of herbaceous wetland plants and peat accumulation to the physiological properties of plants and environmental condition. Aquantitative model for relation of peat forming to the climate is established. Two species of mire plant, i.e. reed (Phragmites communis) and sedge (Carex schmidtii) were selected for the nelon bag test to determine the decomposition rate in 5 different kinds of microenvironment in the wetlands. The experiments show that the substantial composition and the anaerobic environment of roots are both not favorable to decay. The sequence of the micro environments on decomposition of plants is water surface>underground>groundsurface>underwater>abovewater. The dynamic change of decomposition is very obvious. In the first year, there are two decomposition rate peaks occured respectively during 0～10℃ period in spring and 20℃～20℃ period in summer. The changing curve of decomposition in the second year presented a monopeak form. When the temperature is much high for the humidity, the decomposition is fast; when the humidity is much high for the temperature, the restrained decomposition is advantageous for peat to form. Finally, a model for the global peat forming climate was established, that is V=(S-56 % )/T=0 009 28 in whichV : the humidity thermal ratio of peat forming climate; S: multiyear average of relative humidity during the >0℃ period; T: multiyear average of air temperature during the >0℃ period. When S >56%, and ( S -56%)/ T >0 009 28, the climate environment is favorable to peat forming; when S <56% or S >56% while ( S -56%)/ T <0 009 28, the climate is not favorable to peat forming.

Heterogeneity is the most essential landscape characteristic and one of the most important objects of landscape research. In this paper, we use six types of landscape remote sensing maps along with a new procedure that combines average information levels and a lacunarity index, to the analysis of the landscape heterogeneity of the eastern part of the Zhujiang delta. Heterogeneity types and spatial distributions of landscape elements in working areas are very significant. Arable land and orchards are microheterogeneous elements with a homogeneous distribution and a low heterogeneity level. Developing areas are macroheterogeneous elements with luster distributions and high heterogeneity levels. A woodland looks like a microheterogeneous element, but its distribution is more uneven. Natural environmental changes and human disturbances are the main causes of the significant differences in heterogeneity and spatial distribution among the landscape elements. Homogeneous management of agricultural production converts arable land and orchards into nonheterogeneous elements, but as areas decrease there is a gradual increase in their heterogeneity. Vigorous human disturbances in a local area, like non-agricultural land exploitation, causes the difference in the spatial distribution of towns and other developing areas. Since geomorphology is a medi-scale impact factor, the woodland has more of a cluster distribution pattern because of its effect. Water’s high heterogeneity level is produced by a river’s linear characteristics after most of its surface area is changed by construction. Its results can be used for classifying the microheterogeneity of landscape elements. The lacunarity index method, although losing some detailed structural information, provides results that are not impacted by any local special element distribution. We also gain important structural information such as area change and autocorrelation. Using the two methods simultaneously in heterogeneity studies compensates for the weaknesses of each other and improves the results.

Glaciers on the Xizang (Tibetan) plateau play an important role in the earth’s climatic system. Regular surveys of glacier change is almost impossible in some areas of the region, and the use of remote sensing techniques is at present the primary, and in may places the only, means of measurement. GIS provides an efficient tool to analyze the status and the changes of glaciers. In this study, we use 1973～1974 serial RBV, MSS and TM Landsat data and GIS to construct glacier maps and to measure glacier changes for the Bukatage Peak located in the middle section of the Kunlun Mountains. Between 15 and 20 ground control points were selected from topographic maps and identified on each image. The root mean square error (rms) of an independent set of points was selected and used to assess the accuracy of coregistration. The maximum verification error from coregistration of all images was 53 m in the X and 67 m in the Y direction. The details of glacier change such as the velocities of glacier movement at different times were obtained by using GIS tools, based on the results of the serial images’ coregistration, classification, vector analysis and calculation. The result shows that there was no significant change in the study area as a whole, but the northern glaciers advanced while the southern glasiers retreated at the velocity of between 50～105 ma-1. Climatic record for the area shows that both temperature and precipitation had a rising trend from the mid 1950s to the late 1980s. This study suggests that the southern cirque/valley glaciers, especially the bukatage Glacier, responded more rapidly to temperature than precipitation change because of their small areal extent and volume and their south slope positions where more powerful solar radiation occurs. Because of emission from surrounding slopes, the valley bottoms receive more thermal radiation than the unobstructed areas. On the other hand, the northern glaciers, especially the West Bukatage Glacier, are larger in size and receive less solar radiation and thus are less affected by temperature than by precipitation. The northern glaciers show an advancing trend because precipitation, mainly snowfall, increased. The deficiencies of this method and the problem areas for future research are presented in the final section of the paper.

Since 1991, some scholars have proposed the hypothesis that at the end of the late Pleistocene when the sea level was low and glaciation reached its maximum, the environment of China’s continental shelf was once a desert. The idea has been used to interpret the formation of the remnant deposits in the shelf region. The main evidence includes the presence of uniform disintegration of marine strata, extensive mixed deposits, long erosional basal plane, repose angle type of structure, buried dune groups and ventifacts. We contend that whereas the idea has some merit when the environment at that time is seen at the macro scale, the evidence submitted thus far is not convincing. Moreover, the idea does not take into consideration the possible impact of aeolian sand on the environment during marine transgression in the post glacial period. On this issue, we offer three views. (1) In the absence of biological evidence, to distinguish aeolian sand facies (ancient sand dunes) from the desert environment (ancient desert) should follow the principle of using multiple indices, and the indices selected should be able to diagnose landform processes and environmental features. But thus far much of the evidence proposed to support the hypothesis such as the “buried sand dune groups” is not unique to desert environment. (2) Although ventifacts (aeolian gravel) do indicate strong aeolian activities, they are not the only indicator of an arid desert environment. A comparative analysis of the grave forms and surface texture of the gravel from the seabed of the South Sea with the ventifacts of a modern desert environment shows that they differ markedly. They should not be regarded as the same thing. (3) Under the dry and cold glacial climate, aeolian sand (sand dunes) does not speedily cemment and diagenize into rocks. In addition, aeolian san (sand dunes) was highly susceptible to damage and change by marine trnasgression during the post glacial period. Thus it is not likely that sediment structure such as repose angle type of structure and sand dune morphology such as buried sand groups could be preserved in large areas. In short, the remnant deposits on China’s continental shelf should not be seen as ancient desert deposits.

Internet is becoming more and more important in geography research. This makes it possible for geographers to do their research by international cooperation and thus to probe deeply into the essence of geography. Geography resources in the Internet include the following aspects: 1 Virtual geography libraries. This section gives many useful web addresses of virtual libraries, geography links and resource guide. 2 Remote sensing images and maps. They are important means to express geographical information. Some of those resources in Internet are introduced here such as space and aerial photographs, digital images, atlas and cartography. 3 Academic organizations web site. Many academic organizations in geography have established their services in the Internet. 4 Electronic publications. There are some electronic publications known as E Periodical in geography are available for subscription. 5 Electronic forums. The usenet and the mailing lists in various professional geographic fields provide the place for geographers to share their ideas and discuss common interests. Some of them are completely user driven. 6 Software resources. GIS and other geographic analysis software web sites are the main topic here. 7 Educational resources. Online universities and geography courses will bring about great changes in geography instruction, especially by means of multimedia and distance education. Some non profit organizations have set up their web sites to encourage and help teachers to use the Internet in their classrooms. In this paper the authors not only enclose many classified geography resource addresses but also point out some ways to access the Internet efficiently.

Cellular Automata (CA) have attracted growing attention in urban simulation because of their capability in spatial modelling which is not fully developed in GIS. This paper discusses how cellular automata (CA) which is increasingly used in urban simulation can be extended and integrated with GIS to help planners in the search for sustainable urban forms. A constrained cellular automaton is developed to produce optimal development patterns and thus reduce wasteful use of land resources. The cellular automaton is built within a grid GIS system to facilitate easy access to GIS databases for constructing the constraints. The essence of the model is that constraint space is used to regulate cellular space. Local, regional and global constraints play equally important roles in affecting the modelling results. Agricultural suitability, available land resource and economic and population growth could be used as key constraints. The paper discusses how to combine these constraints in CA modelling to enhance the efficiency in using land resources. In addition, ‘grey’ cells are defined to represent the continuous degrees or percentages of urban land development during the iterations of modelling for more accurate results. The model can be easily controlled by the parameter k using a power transformation function for calculating constraint scores. The CA model can be used as a useful planning tool to test the effects of different scenarios of urban development. The protection of valuable agricultural land is important in many developing countries where cities are growing rapidly. In China, some cities are expanding without properly considering land suitability and environmental impacts. Recent rapid land development has been accompanied by the loss of large amount of agricultural land. This study demonstrates that CA can be used to simulate urban development based on constraints that reflect our environmental concerns. The objective is to produce sustainable urban forms as alternatives to the existing development patterns. The study has suggested that CA can be used as a useful planning tool to model urban development under complicated and changeable environmental factors.

This paper focuses on the changes of the basic nonbasic economic activities in Guangzhou since the middle of 1980s. The attention is initially paid to the composition and change of urban basic nonbasic economic activities. Four research findings have been achieved. ① The general change tendency was that the scale of urban basic economic activities was smaller than that of nonbasic ones and showed the tendency of becoming smaller with time. The secondary sector showed the shrinking scale of basic activities but the tertiary sector had a reverse change. ② The scale of the basic economic activities of industry shrank and experienced a shift of the basic part being bigger than nonbasic part towards smaller than nonbasic, but industry was still the major economic basis of Guangzhou. ③ The proportions of basic and nonbasic parts of many tertiary trades changed, i.e. as the former rose, the latter fell, or the basic part had a relatively high proportion all the time, which strenghthened the status of the basic activities in urban economic development. ④ The fact, the basic employment (basic population) of producer services increased with a faster speed than the total employment and accounted for nearly 10% in the total employment, indicated that the producer services were becoming one of the important components in the urban economic basis and had the huge potential and possibility of occupying the regional market. Then the analysis was turned to the external function of the producer services based on questionnaires. The results showed that: ① In the producer services of Guangzhou, there indeed existed basic economic activities of exporting services to its surrounding regions (Guangdong Province, the whole country and the overseas). ② The differences of the external functions among the trades of producer services were obvious. The finance and insurance sectors had smaller proportions of providing services for the areas outside of the city than other sectors like information consultant, computer application, and scientific research and technological services. ③ The producer services in Guangzhou serviced mainly for Guangdong Province, which showed that the development level and scale of producer services were limited and their international market was narrow. Thirdly, the economic base theory of urban development was discussed. The research results showed that services not only provided services for cities themselves but also export services to the areas outside cities. And producer services mainly provided services for the areas outside cities (i.e. mainly being basic economic activities) which indicated that producer services have the function of engineering economic growth. The substance of development of basic economic activities lies in the tradability of services. Several factors, information technology, transnational enterprises of services and transformation of government function, have resulted in the services’ tradability. All these changes made the relationship between basic economic activities and urban development more complicated than before. A re explanation to the traditional economic base theory, paying more attention to interaction between urban economic activities and the location significance of the theory, was made. With the technological progress, the impacts of basic economic activities on urban development by stages have been identified. All these imply that the classic theories have not well explained the urban development in the post industrial society and it is very necessary to make the re explanation to the classic theories and build new theories.

The 17 years mean (1979～1995) NCEP/NCAR reanalysis radiation data are used to estimate the global monthly mean surface albedo. For clear sky conditions, the monthly mean surface albedo can be obtained by the ration of monthly mean reflected radiant flux and monthly mean total radiant flux at surface, and a correction is applied which takes account of the effect of solar zenith angle. The results show that the seasonal cycle and the spatial distribution of global surface albedo are correspondent to the characteristics and variations of the earth’s different underlying surface. The zonally averaged surface albedo shows an apparent annual variation. A comparison of the results of present method with other studies indicates that the values of sea surface albedo at high latitudes in this work tend to be slightly higher than the others. While the distribution of surface albedo in other regions are basically consistent to the others. The surface albedo in China also well represents the characteristics and seasonal cycle of complex underlying surface. Compared to the other similar works carried by Chinese scientists, the time and spatial distribution are accordant but the values in some regions are slightly higher. Verified by observation data of Qinghai Xizang (Tibetan) Plateau Meteorological Scientific Experiment in 1979, it is found that the surface albedo in the Qinghai Xizang (Tibetan) Plateau region obtained from the NCEP/NCAR data is closer to the observation data than the others. From the above results, it is suggested that the global monthly mean albedo estimated from the NCEP/NCAR can be applied into the global climate model, further correction for the albedo will depend on the sensitivity numerical experiment of the climate model.

Semi arid climate, poor vegetation cover and highly erodible loess can be regarded as the main natural factors responsible for the formation of hyperconcentrated flows in the Loess Plateau, China. The degree, to which the hyperconcentrated flows develop, especially their extreme suspended concentration, shows a significant areal differentiation. This is closely related to the grain size composition of suspended sediment. To study the relation between hyperconcentrated flows and the surface material, the mean annual suspended concentration C mean and C max suspended concentration C max have been plotted against the >0 05 mm and <0 01 mm percentages ( R >0 05 mm, R <0 01 mm) in suspended load, respectively. For each curve obtained, there is a peak value. The peak value appears where R >0 05 mm is around 40% or R <0 01 mm is around 20%. This fact means that there exists an optimal grain size composition of suspended load, which makes suspended concentration reach the highest value. The formative mechanism for this optimal grainsize composition can be explained by the special physical and mechanical properties of hyperconcentrated flow as a solid liquid two phase flow. When the fine sediment is much more but the coarse particles are not sufficient, the flow may enter a laminar state; when the coarse particles are much more but the fine sediment is not sufficient, the coarse particles can not be set in suspension effectively. In these two cases, suspended concentration cannot reach very high. Obviously, between the two ends, an optimal grainsize composition exists at which the suspended concentration may be the highest. This optimal composition is related to the combination of three surface material types, i.e. loess, wind blown sand and weathered material of bedrock, and it occurs as a result of seasonally alternating wind and water processes operating in this area. It was found that in the middle Huanghe River basin, suspended sediment is coarser than that of the loess in the basin, but finer than the wind blown sand and bedrock weathered material. So, it is some proper combination of materials supplied from these sources that forms the optimal grainsize composition. During winter and spring, wind carries coarse aeolian sand to gullies and river channels. It has been observed in the field that, driven by strong wind, aeolian sand dunes on floodplains may approach the river bank and then collapses into the channel. This process is enhanced by bank erosion. When floods occur in summer, the aeolian sand previously supplied to the channel may be carried downstream by stream flow. During the floods that come from loess covered areas and lack of coarse sediment, the previously stored coarse aeolian sand may be set suspension by the fine sediment rich water flow and thereby its suspended concentration increases sharply, leading to the formation of hyperconcentrated flows.

There are 6 major rivers, named Wudanggou, Shuijiangou, Meidaigou, Dashuigou, Shuimogou and Halagou, from west to east, in Mt. Daqingshan, the upthrown block of the Daqingshan Piedmont Fault, which is an active normal fault. Based on paleoearthquakes having occurred along it, the Piedmont Fault can be divided into 5 segments, one of which has undergone unique faulting history. Accordingly, the 6 rivers, Crossing different segments of the Piedmont Fault, have differences in their landforms. The features of river valley landforms in Mt. Daqingshan reveal that the 3 middle segments are more active than the other 2 segments of the fault. In the river valleys, 2～5 terraces, having developed since 23 ka BP of Late Pleistocene, are identified. In the 4 west rivers’ valleys, the number of terraces is 4～5; the highest terrace is 80～120 m above the current river bed; all of the terraces have been displaced by the Daqingshan Piedmont Fault. But in the 2 east rivers’ valleys, the number of terraces is fewer, only 2～4; the highest terrace is 20～45 m; the latest terrace, formed in Holocene, has not been displaced yet. The model diagram of longitudinal profiles of the 6 rivers’ valleys is a parabolic pattern: h=alⁿ . The longitudinal profile of the 4 west rivers’ valleys has a convex parabolic pattern, n<1; but that of the 2 east rivers’ valleys has a concave pattern, n>1. The more active a segment of the Fault is, the more convex the parabolic pattern of the river’s valley crossing the segment is. From west to east, V_f value of the river mouth at the Piedmont Fault becomes higher. Based on the detailed landforms and their formation ages of the rivers’ valleys, we can drive some conclusions on the activities of the Daqingshan Piedmont Fault. Since 23 ka BP, the uplift amount of Mt. Daqingshan is the same as the elevation of terrace Ⅴ near the Fault, 80～120 m for the 3 middle parts, and 20～45 m for the east part. During Holocene, the uplift amount is about 17～18 m for the 3 middle parts and less than 6 m for the east part. Obviously, the slip rate of the Daqingshan Fault can be calculated.

Endemic goiter, a worldwide endemic disease which relates to geo chemical environmental factors, has a very long prevalent history in China. This thesis discusses the below four issues by analyzing a large amount of Chinese historical document. 1. Ancient Chinese descriptions on endemic goiter’s symptoms As early as 2 500 years before present, Chinese knew the facts that endemic goiter distributes mostly in mountainous areas and prevails mainly in female population. In Chinese historical documents, goiter was called “Ying” (goiter) at first, and then “big neck”. According to modern medicine, “Ying” is the third or fourth degree of endemic goiter, “big neck” is the first or second degree of endemic goiter. Chinese character “Ying” is composed of words “disease”, “ornament” and “female”, its initial meaning might relate to the thought of antiquity people of goiter as the ornament surrounding female’s neck. The fact shows ancient Chinese very early knew that female falls endemic goiter easier than male. By the Southern Dynasties (A.D.420～589), Chinese were able to distinguish between diffuse goiter and nodular goiter. Unfortunately, after the Southern Song Dynasty (A.D.1 127～1 279), goiter was mixed with tumor by many doctors. 2. Ancient Chinese knowledge on the links between goiter and environment Ancient Chinese believed that the prevalence of endemic goiter in some area related closely to the natural environment of this area and pointed out that water quality and landscape were important pathogenic factors of endemic goiter. About 2 000 years ago, they discovered that mountainous areas were main goiter’s epidemic areas so that they wrongly believed endemic goiter resulted from drinking rushing, sandy water and living in precipitous place. Although then doctors knew how to treat goiter sufferers with rich in iodine drugs, until the early of the 20th century, Chinese don’t knew that the real reason for endemic goiter’s prevalence is iodine deficiency. 3. The geographical distribution of endemic goiter in ancient China Because ancient historians and literature writer intentionally evaded recording diseases, the records on endemic goiter in China’s historical document are seldom seen and very fragmentary. But this paper yet collects enough historical data to recover the distribution image of endemic goiter in ancient China’s hinterland. By analyzing a lots of historical records, the author finds the serious prevalent areas of endemic goiter in ancient China mainly include: the Qinba mountainous areas in Shaanxi Province, the western mountainous areas of Henan Province, the west northern mountainous areas of Hubei Province, the Zhongtiao mountainous areas in Shanxi Province, the Taihang mountainous areas in Hebei Province, the Yimeng mountainous areas in Shandong Province, the Three Gorges area, the Minshan mountainous areas in the northwestern of Sichuan Province, the Dieshan and Liupan mountainous areas in Gansu Province, and so on. These areas are situated in the central part of China, so are earlier exploited and are more recorded. But it is not mean other areas without the records are not the prevalence areas of endemic goiter. 4. The causes of distribution change and the socio economic harm of endemic goiter Endemic goiter is a geo chemical environmental disease. Owing to only a little change of geo chemical environment in the past 2 000 years, the spatial distribution of endemic goiter in China remained relatively stability so that the serious prevalent areas of endemic goiter in the historical times was generally coincident with the distribution areas of endemic cretinism in the early of 1980s. But, by comparing and analyzing, we could still find, because of the improvement of translation condition and dietary composition, the distribution scope of endemic goiter reduced and its harm is alleviated and even eliminated in some areas. For example, the western part of Three Gorges area, the eastern foot plain of the Taihang Mountain.

The overland flow is defined as the flow of water over the land surface toward a stream channel and is the initial phase of surface runoff. The sediment yielding from overland flow erosion increases with the increasing slope if the slope is smaller than a critical value, and reduces following further increase of the slope if the slope is larger than the critical slope. With Energy Theory, Sediment Transportation Mechanism and data from laboratory and field experiments, the critical slope of overland flow erosion was studied in detail. Firstly, from the viewpoint of energy analysis, the detachment capacity of overland flow is directly proportional to the flow shear stress, and the shear stress is directly proportional to the kinetic energy of overland flow. A theoretical formula is derived which indicated that the energy of overland flow is closely related to the flow rate per unit width, composition of soil particle size, runoff depth and slope gradient. The maximum flow shear stress occurs if the slope is between 22° and 26°. Therefore, the conclusion from the energy analysis is obtained that the critical slope of overland flow erosion is about 22°～26°. Secondly, the analysis from dimensionless shear ratio and sediment transportation mechanism shows that the capacity of overland flow carrying sediment reach its maximum at the slope gradient approximate to the frictions angle of sediment under water. The sediment carried by overland flow is mainly composed by the silt and fine sand, whose friction angle under water is measured at 22°～27°. This means that critical slope of over land erosion is between 22°～27°, which is coincident with the critical slope derived from energy analysis. Thirdly, simulation experiment in laboratory shows that the maximum sediment yield is obtained at the slope of 23°～27°, which verified the result induced from the theoretical analysis. Finally, it is concluded that the critical slope gradient of sheet erosion and rill erosion is about 22°～27°, and that of rainsplash transportation is below 22°, and critical slope gradient of gully and gravitational erosion will be bigger than that of sheet erosion and rill erosion. Another conclusion of this paper is that the critical slope gradient is not a fixed value, but a range of values, and discussion of the critical slope of erosion should refer to concrete erosion way and boundary conditions.

The spatial structure of the source or market areas of tourists is basic to tourist geography. Although there are case studies on such issues as mathematical modeling of distance decay of tourists from a tourist site, its market potential and actual market structure, many questions remain. Take tourist distance decay for example. Whereas hypothetical theoretical curves have been suggested, no basic mathematical model based on case studies exists. In addition, the application of distance cumulative curve in tourist research is still limited to descriptions of the actual distribution of tourists, which is ineffective in analyzing the spatial structure of tourist market potential. Jiuzhaigou Conservation District, one of China’s most picturesque tourist areas with numerous colorful lakes and waterfalls, is recognized by UNESCO as one of the world’s natural heritage sites. Distance decay model is used to reveal the characteristics of its market area structure. Several types of distance cumulative curve are also used to reveal its market structure at varying distances from the site. A sample of more than 16 000 tourists was obtained from entrance registration which shows the they come from 235 cities and counties. Through regression analysis we calculated the main coefficients of the distance decay model and found that the best fit model is the Pareto model with a distance decay exponent of 0 48. The next best are the lognormal model and square root exponential model, while the normal model is unfit. As the relative coefficients of these models are not high enough, we used the gravity model to analyze distance decay versus the influence of population and economic growth. The result shows a distance decay exponent of 1.84 and a clear increase in relative coefficient to 0 70. Gravity model was also used to better understand how the sampled tourists declined proportionally with increasing distance from the site. The result shows the distance decay exponents range between 1 74 and 1 92. When per capita industrial and agricultural output is replaced with total industrial and agricultural output of cities and counties, the exponents range between 2 5 and 3 0. This study demonstrates that the several distance cumulative curves of the market areas of the site and the population of the market areas can be effectively used to analyze the spatial structure of the site’s ultimate market area potentials.