Spatial pattern of tourism resources is an important factor for regional tourism development. Exploring its own rules has significant meaning for the development of the tourism industry. The study on spatial pattern of tourism resources in the Yangtze River tourism belt is useful for understanding the features and spatial structure of the resources, and is important for scientific planning of the overall layout of regional tourism industry and the achievement of sustainable tourism development. In this article we selected spatial data of 4649 tourism resources of 12 class in the Yangtze River tourism belt across 9 provinces and 2 cities, including World Heritage Sites, 5A-class scenic spots, National Nature Reserve, National Forest Park, National Scenic Area, National Wetland Park, Historical and cultural city, National key units, Classic Red Tourism, Chinese Traditional Villages, Characteristic Landscape Tourism Famous Town (Village) and Spa. With the ArcGIS 10.2 software, we use spatial autocorrelation analysis and kernel density estimation method to analyze the spatial pattern of tourism resources of the Yangtze River tourism belt. The results suggest that the overall tourism resources, human tourism resources and cultural tourism resources of the Yangtze River tourism belt show a significant resource clustering feature in space. The overall tourism resources distribute mainly along the line of 30° latitude, The Yangtze River waterway, the G318 Highway, Shanghai-kunming High-Speed Railway and Shanghai-Wuhan-Chengdu High-Speed Railway; The high-level tourism resources which including the World Heritage, 5A-class tourist attractions and National Nature Reserve are the most densely populated in the Yangtze river delta; Natural tourism resources are centralized in the Shanghai-Wuhan-Chengdu high-speed rail and Shanghai-Kunming high-speed rail; Human tourism resources are mainly concentrated in Yangtze River Delta, ancient Huizhou area, northwestern Yunnan province and Southeastern Guizhou province where Minorities live. According to the features and distribution of tourism resources, we summarize some suggestions:based on the Yangtze river waterway, the G318 Highway and the high-speed, constructing the east-west tourism fast channel and boutique tourist route; using the policy advantages, uniting multilateral organization to create cross regional tourism cooperation community, activate the market; developing characteristic tourism and tourism new hotspot in homogeneous tourism resources concentrated area.

Based on the panel data of urban tourism development and its dynamic factors in Pan-Yangtze River Delta from 2002 to 2012, we analyzed spatial effect of urban domestic and inbound tourism development through Spatial Panel Durbin Model. The results show that:(1) Urban inbound and domestic tourism development in the Pan-Yangtze River Delta has obvious spatial dependence effects and positive spatial spillover effects. If the level of urban domestic and inbound tourism development increases by 1%, it will directly promote the level of regional domestic and inbound tourism development growth by 0.257% and 0.312%; (2) The spatial spillover effects of dynamic factors on urban domestic and inbound tourism development are also obvious. The level of tourism resources endowment, tourism service facilities, tourism traffic facilities and economic development has an obvious spatial spillover effect on inbound tourism development, where the level of tourism service facilities and economic development has an obvious spatial spillover effect on domestic tourism development; (3) The spatial interaction of affordable factors are positive spillover effects, where the attractable factors are negative spillover effects. Because of the difference of the spatial interaction and its intensity of affordable factors and attractable factors, the spatial pattern shows constant evaluation.

Landscape pattern in urban tourism destination is influenced by both the development of urban tourism and landscape process. As a result, analysis from integrated perspectives should be applied in landscape optimization. Taking Nanjing as an example, using the ArcGIS software, the evolution and optimization of landscape pattern in the study area were studied through the model of the evolution of urban tourism destination and the cumulative cost distance model. The results showed that:(1) From 1991 to 2003, the proportion of tourism landscape reduced while the non-tourism landscape increased, and the proportion of tourism landscape increased while the non-tourism landscape reduced during 2003 to 2014. This change showed the landscape evolution experienced a transition from the background type to the tourism type, expressing the characteristic of growth tourist destination. (2) In recent years, the degree of landscape fragmentation continued to aggravate. The landscape shapes of water of lake and river, parks and green space, and the construction became more and more regular, while the forest land and transportation became more complex. (3) Landscape pattern optimization can be achieved by building ecological sources, critical ecological positions and corridors, which can strengthen the spatial relationship of ecological network. (4) The focus of the optimization of the landscape type is to further transform the non-tourism city landscape into the typical tourism ones, in order to enhance both tourism and ecological value of landscape.

Wuhan city has developed rapidly since the 21st century, and the urban area is increasing. With high-resolution remote sensing images at different phases and object-oriented methods of interpretation, we derived the land cover map of Wuhan city in 2000, 2005, 2010 and 2014. And the spatial and temporal characteristics of Wuhan urban expansion have been analyzed in terms of urban expansion intensity index(UⅡ), the city center coordinates and fractal dimension. It showed that the UⅡ of Wuhan city was 1.41 in 2000-2014, and the annual growth rate of construction land reached 10.7%; The annual growth rate of non-core urban construction land was 20.3%, much larger than the main city; And each central district had different UⅡ in 2000-2014, the UⅡ of Qiaokou District was 2.01 while that of Wuchang District was only 0.54. The construction land area of Hongshan District increased 112 km² which was significantly higher than other main city in 2000-2014 years, its urban expansion area accounted for 59.2% in urban expansion area of the main city, so the largest contribution to urban expansion of the main city was the growth of construction land in Hongshan District; With the expansion of Wuhan city, the fractal dimension of Wuhan City increased in 2000-2014, suggesting that urban spatial morphology became complicated, and the impact on urban form of urban interior renovation was greater than the city extended outward in 2010-2014, but the urban form of Wuhan city is still in instability; In 2000-2005 years, the form of urban expansion in Wuhan city was "cone-shaped", and the form of urban expansion in Wuhan city was "plume" in 2005-2010 years while the form of urban expansion in Wuhan city was "massive" in 2010-2014 years; The gap in different directions of urban expansion intensity in Wuhan city became smaller in 2000-2014, and the extended mode of Wuhan city was core-radial in general, but it is gradually turning circle formula expanded configuration and Wuchang District and Jianghan District are turning content-type. At last, natural conditions, economic development, population growth, transportation facilities, policy and urban planning were the main driving forces of Wuhan city's expansion in 2000-2014 years, and transportation facilities had the largest driving forces while the population growth had a greater impact on urban expansion than the economic development. With the construction of transport facilities, the effect of natural conditions and economic development to urban expansion in Wuhan city is gradually weakened, but the policy and urban planning still determine the general direction of urban expansion of Wuhan city.

To promote structural reform of the supply front in the Wuhan Metropolitan Area and improve production efficiency of land, each county's TEP and the contribution of capital, labor, land, as well as intermediate inputs to economic growth were analyzed before and after the establishment of "Resource-economical and Environment-friendly Society" experimental zone of the Wuhan Metropolitan Area (2000~2006 and 2007~2012) using the Solow residual method. The results showed that the economic development in Wuhan Metropolitan Area mainly relied on input of capital, labor and technology progress, while the contribution of land expansion to economic growth was relatively low. Remarkably, intermediate inputs had a weak effect on economic growth. Furthermore, "Resource-economical and Environment-friendly Society" experimental zone establishment elevated the land element contribution to economic growth. Although the output efficiency of urban land expansion was enhanced, but it was still weak and less increased compared with the input of other factors. We also observed that the gap between ΔE_S and ΔTEP was inversely proportional to the dependence of the corresponding industrial structure on land. The following conclusions can be drawn:on the one hand, the government should continue to promote technical innovation in the economic development, put efforts into labor force training, and attract high quality talents, so as to make the economic development direction from extensive growth to intensive growth. On the other hand, firstly, the government should push forward the corresponding policy to support industry development involving land and improve the participation degree of the land on the economy. Secondly, the government should make master planning of urban land expansion in the whole Wuhan Metropolitan Area and develop corresponding policy of land supply and economic development according to the difference of the industrial structure and resource superiority. Finally, the government should pay more attention to intensive land utilization on the principle and mechanism aim of structural reform of the supply front.

Guiyang is a typical karst area in our country where water and soil erosion is serious and ecological environment is vulnerable. This research identifies ecological sources on the foundation of ecological importance, landscape connectivity and ecological demand. Ecological resistance surface is built by means of land use/cover, anthropogenic disturbance and nighttime light. Based on MCR model to recognize ecological corridor network, we built the landscape ecology security pattern of administrative region and downtown in Guiyang. The results showed that:(1) the area of the most important and important ecological patches was 5393.43 km² which were mainly distributed in the north and west. Landscape connectivity of these patches was generally good. Garden land was the best and forest land was the worst. Area with higher ecological demand was in the surrounding of built up area with dense population. Ecological sources accounting for 18.56% of the total area were located in the north area or somewhere like Hongfeng Lake and Baihua Lake in the southwest. (2) Ecological corridors were 933.58 kilometers long and the key corridor was 287.6 km long spanning from northeast to southwest. The current corridor that mainly consists of river system is all over the city. Nighty-eight percent of the potential corridor and the key corridor were forest and they were distributed in the hilly areas with good ecological environment. (3) According to the landscape ecological security pattern of downtown, ecological sources were located in some conservation areas such as Baihua Lake, Hongfeng Lake, Xiangzhi Valley and Xiangsi River. As the key corridor of downtown, Huaxi Wetland and Nanming River extended from south to north. Based on these results, this research put forward some ecological management measures which require simultaneous and considerate preservation, construction and planning to provide scientific and workable guidance for ecological environment constructions in Guiyang.

The health assessment of soil, which is an important factor of forest ecosystem, is of great significance for maintaining soil health and forest ecosystem functions. On the basis of systematic investigation and analysis of soil characteristics for 8 types of forest vegetation, the assessment criteria were selected, ranging from species diversity and various soil properties, including coefficient of species diversity, litter layer thickness, soil humus layer thickness, soil thickness, soil bulk density, clay content, organic matter, pH, cation exchange capacity (CEC), total nitrogen, hydrolysable nitrogen, available phosphorus, available potassium, and phosphatase activity. The difference tests and correlation analysis of various criteria data were conducted with the SPSS software (version 19.0), and the weight of each index were determined by proper scoring function equations. Then, the soil health index was calculated through weighted synthesis, and soil health was assessed for 8 type forest vegetation. The results show that the order of health indices for various forest soils is:mixed coniferous broad-leaved forest (0.78) > evergreen and deciduous mixed forest (0.72) > shrub (0.69) > evergreen broad-leaved forest (0.67) > deciduous broadleaved forest (0.64) > bamboo forest (0.59) > Pinus massoniana forest (0.53) > Pinus taiwanensis forest (0.46).

Forest classification plays an important role in understanding the structure and function of forest ecosystem. Shennongjia Forest district belongs to the northern subtropical monsoon climate zone with vegetations distributed in different vertical belts. Because of its high altitude, complicated climate, rich forest types and fragmented landscape, it is difficult to classify the vegetation types in forest region of Shennongjia district. This paper constructed a time series analyses for vegetation classification. The time series of factors collection consisted of the vegetation indices, the first principal component (PC1) and topographic factors, which were calculated from 24 HJ-1A/B CCD images of 2013 and DEM. The vegetation indices collection included NDVI (Normalized Difference Vegetation Index), DVI (Difference Vegetation Index) and RVI (Ratio Vegetation Index). The topographic factors collection included altitude, slope and aspect. By means of the C5.0 Intelligent decision tree algorithm, seven categories were classified in the Shennongjia forest district. They were:cold-temperate coniferous forest, temperate coniferous and broad-leaved mixed forest, temperate deciduous broad-leaved forest, broad-leaved evergreen and deciduous mixed forest, north-subtropical evergreen broad-leaved forest, shrub and meadow. The results showed that:(1) Using the intelligent decision tree algorithm, the overall accuracy of the classification reached 72.7%, with a Kappa coefficient of 0.67; (2) the Vegetation Index values of north-subtropical evergreen broad-leaved forest, the broad-leafed evergreen and deciduous mixed forest, temperate deciduous broad-leafed forest and the temperate coniferous and broad-leafed mixed forest were higher than the cold temperate coniferous forest, scrub and meadow in June to August; the window period of forest vegetation classification was identified as June to August. Using the time series of factors collection can show up "the window" and provide a theoretical basis for the vegetation classification; (3) the time series of PC1, NDVI and altitude as well had a significant contribution to the forest classification; the time series of DVI can help to distinguish between north-subtropical evergreen broad-leaved forest, broad-leafed evergreen and deciduous mixed forest, temperate coniferous and broad-leaved mixed forest, temperate deciduous broad-leaved forest; while slope, aspect and the time series of RVI had much less help in the automatic generation of classification rules, thus, they were of little value to vegetation classification. Overall, the time series of data derived from 30m-resolution HJ-1A/B CCD imagery can accurately categorize the Shennongjia forest vegetation into seven classes based on the C5.0 decision tree method. As an intelligent decision tree classification algorithm, the method has the advantages of identifying more categories and is of high application value. There is still space to improve the classification accuracy, i.e., a more considerate classification system of Shennongjia Forest, the increasing of field samples at mixed vegetation belt, and the use of higher spatial and temporal resolution images.

The spatial variation patterns of various vegetation habitat suitability categories under global and local scales are main characteristic and symbol for vegetation habitat management and restoration in the upper reaches of Min River. Based on spatial distributions of four class vegetation habitat suitability categories (class one was suitable vegetation habitat whose habitat suitability index was highest, class two was sub-suitable, class three was unsuitable, class four was worst suitable whose habitat suitability index was lowest), in this study we established a lacunarity index and used multilevel bayesian models to quantify spatial cluster and variation of diverse vegetation habitat suitability categories from global (the whole upper reaches of Min River) and local scales (five counties including Heishui County, Li County, Songpan County, Mao County, and Wenchuan County). The detailed results were as following:(1) The maximum lacunarity indices of comprehensive class (Ⅰ~Ⅳ), class three (Ⅲ), and class four (Ⅳ) occurred in the Wenchuan County, the spatial variation scale of above three habitat suitability categories were 3000 m (30 times of grid size, the grid size is 100 m), 2900 m, and 2800 m, respectively. It is attributed to driving factors including the lower elevation, flat topography, higher precipitation and lower evaporation. The class two (Ⅱ) and class three (Ⅲ) appeared in the Songpan County owing to higher altitude, steeper slope, lower precipitation, and higher evaporation. The spatial variation scale of previous two habitat suitability categories were 3000 m and 2800 m, respectively. It was suggested that the habitat management of Wenchuan and Sonpan counties was sensitive to sampling grid size (habitat management unit) because of the spatial variation of different habitat suitability categories represented by topographical, meteorological, soil, and vegetation driving environmental factors. The habitat suitability category of maximum lacunarity index appeared in Heishui County, Li County, Wenchuan County, and whole watershed was class Ⅰ. However, the habitat suitability category of maximum lacunarity index appeared in Mao County, and Songpan County was class Ⅳ. (2) The maximum lacunarity dimensions of Ⅰ~Ⅳ, Ⅰ and Ⅱ occurred in the Wenchuan County, Songpan County and Li County, respectively, which meant their spatial structures were in the discrete state. The maximum lacunarity dimensions of Ⅲ and Ⅳ occurred in Mao County. The county administrative unit had the largest impact on spatial variation and character scale of class Ⅱ and had smallest impact on spatial variation and character scale of class Ⅰ~Ⅳ. The analysis of spatial variation of vegetation habitat suitability could provide the theories for vegetation habitat management in the upper reaches of Min River.

Studies on the status of fish resources and feeding habits of Ptychobarbus kaznakovi of the Zengqu River, in the upper Jinsha River, were conducted in May and September, 2013. A total of 10 fish species belonging to 2 orders and 3 families and 5 genera were found in the Zengqu River. The Cyprinidae included 7 species accounting for 70% of species; Sisoridae included two species accounting for 20% of species; Cobitidae included one species accounting for 10% of species. The diet of P. kaznakovi consisted of 53 genera or species, which included phytoplankton, protozoa, arthropod, etc. Bacillariophyta had the highest occurrence rate percentage, index of relative importance and quantity percentage among the all prey species. Aquatic insects had the highest mass percent among the all prey species. There was an obvious seasonal difference in feeding intensity of P. kaznakovi, which had relative lower feeding intensity in the end of the rainy season. Index of relative importance of plant in the diet increased with the growth of body length of P. kaznakovi. The present study indicates that the fish composition of the Zengqu River was simple, and Cyprinidae fishes were the dominant fish species. P. kaznakovi is omnivorous fish with seasonal variation. It is suggested that ecological compensation measures such as fish inhabitant conservation or reconstruction should be done during the hydropower cascade development to protect the fish in the Zengqu River.

The absorbed phosphorus is one of the major surface contaminants, which not only leads to aquatic environment degradation, but also plays a critical role in water eutrophication. In order to provide theoretical information for agricultural non-point source pollution (ANPS) management and marine ecological safety, the major purpose of this research is to assess the absorbed phosphorus loads under ANPS in the coastal areas. An estimation model of adsorbed phosphorus load was constructed by integrated soil erosion factor, sediment delivery factor and phosphorus enrichment factor together. The model was used to assess the load capacity of adsorbed phosphorus from 2000 to 2010 in the coastal areas of Jiangsu Province using GIS. Finally, the dynamic change features of absorbed phosphorus loads under different land uses were analyzed. The results showed:(1) Absorbed phosphorus load in the coastal areas of Jiangsu province increased significantly from 2000 to 2010, the average phosphorus load modulus increased from 105.89 kg/km²·a to 201.67 kg/km²·a, while the total volume of absorbed phosphorus load increased from 3284 t to 6255 t. (2) During the ten years, the hot areas of absorbed phosphorus load in the research region have narrowed, presenting a state of convergence, while the total amount of absorbed phosphorus load in the hot areas increased significantly, which indicated that the spatial polarization of phosphorus load in the research area was more obvious. (3) Within the ten years, there was a significant growth of the average phosphorus load modulus under different land use backgrounds. According to the total volume of phosphorus load in the same year, different land use types in coastal areas of Jiangsu Province ranked in the following order:paddy field > dry land > forest land > grass land > unutilized land > garden plot. The research results indicated that there was significant difference in the phosphorus load among different land use backgrounds. Therefore, the reasonable land use structure adjustment and landscape pattern optimization is beneficial to preventing and controlling phosphorus pollution.

Poyang Lake is the largest freshwater lake of China and also one of the only two large-scale water-carrying lakes that takes in and sends out water in the Yangtze area. Extreme droughts and low water levels have frequently occurred in dry seasons of Poyang Lake in the last decade which affected the function of wetland ecosystems and their important factors. How to get balance between relieving the drought situation so as to use the wetlands wisely and protecting biodiversity is a major concern both domestically and internationally. This research analyzed the existing problems of Poyang Lake wetland ecosystem regarding hydrology, river-lake relationship, water quality, waterbird habitat and fishery resources, and systematically reviewed the external and internal factors causing these problems. Our results suggest that the lower water levels in dry seasons of Poyang Lake have negative effects on wetland ecosystems of Poyang Lake. In the area, the hydrological regime changed, as dry seasons arrived earlier and the duration extended as well as water levels in dry season lowered than that before 2003. The interaction between Poyang Lake and Yangtze river was altered; Wetland ecosystem degraded such as water quality declined, wetland vegetation degraded, threats to habitat of waterbirds increased and recession of fishery resources. Changes of Poyang Lake wetland both were affected by external drivers, composed by upstream reservoir group (including Three Gorge Dam), controlling reservoirs within the watershed, climate change, water consumption, pollutant emissions, overfishing, sand mining and water-control activities et al., but also by the constraints of internal drivers, including changes of water flow, velocity of flow, flooding duration, and water depth. Based on the scenarios of "business as usual" and "water level regulation", future trends of wetlands were predicted, and the uncertainties in the current study were pointed out. Predicted impacts on local ecosystem of different regulatory schemes through model simulation and scenario analysis were suggested in case to optimize the regulatory schemes and minimize their adverse impacts.

Soil physical properties and anti-erodibility were determined under natural forest (TRL) and artificial plantations of Phyllostachys heterocycla (MZL), Sassafras tzumu (CML), Cryptomeria fortunei (LSL), Cunninghamia Lanceolata (SML) and Metasequoia glyptostroboides (SSL) which were converted from TRL in southern Sichuan Province. The methods of principal component analysis (ZCFM) and membership function (LSHM) were used to calculate the comprehensive valueⅠ(KSX-Ⅰ) and Ⅱ(KSX-Ⅱ) of soil anti-erodibility, respectively, to evaluate the impact of forest change on soil anti-erodibility and explore a simple and feasible soil anti-erodibility evaluation method. The results showed that natural forest converted to artificial forests, leading to worse soil physical properties and anti-erodibility, and decreasing content of soil organic matter, >0.25 mm water stable aggregate, mean weight diameter of water stable aggregate, aggregate degrees, porosity and physical stability index; on the contrary, the ratio of soil structure deterioration, unstable aggregate index, bulk density, dispersion rate and erosion coefficient increased. The KSX-Ⅰ and KSX-Ⅱ of soil anti-erodibility under different forest types ranked as TRL > CML > SSL > LSL > MZL > SML. Significant (p<0.05) correlations were observed between the indices of soil anti-erodibility. The KSX-Ⅰwas significant (p<0.05) correlated with KSX-Ⅱ, and both were significant (p<0.05) correlated with all indices of soil anti-erodibility. Our results suggest that soil anti-erodibility and water and soil conservation will become worse after natural forest change, and the protection of natural forest and its ecosystem should be strengthened; both ZCFM and LSHM were effective methods to calculate comprehensive value of soil anti-erodibility, while the latter is more simple and convenient than the former when the difficulty degree of calculation process was took into consideration. The results could provide references for developing new methods for comprehensive evaluation of soil anti-erodibility.

The analysis of ion concentration and composition characteristics was conducted based on samples from the Sanchahe River Basin, which were collected in dry and wet seasons from the Wujiang source area. The results showed that the dominate anion and cation were HCO₃^-and Ca²⁺ in the samples, accounting for 55% and 70% of the total anions and total cations, respectively, suggesting that the ion composition was similar with that of the karst area river basins. The comparison of temporal variations of the concentrations of major ions showed that, the concentrations of Ca²⁺、Mg²⁺、Na⁺、HCO₃^-、Cl^- in dry seasons were slightly higher than they were in wet seasons, while the concentrations of K⁺、SO₄^-2、NO₃^- were almost the same in these two seasons. And the spatial distribution types of the ions were diversified, showing that different small basins have different geological background, ecological environment, and different human activities, all of these could affect the concentration and composition of water ions. The Gibbs diagram indicated that the river water chemistry was mainly influenced by rock weathering. Triangle diagram showed that the river water chemistry was mainly affected by the carbonate rock, and sulfuric acid was widely involved in the rock weathering. Human activities also have some influences on the composition of water chemistry.

Quantitative evaluation the effects of climate change and human activities on river flow is of great importance for water resource management and utilization. As the water source of China's South-to-North Water Transfer Project in the middle line, Danjiangkou reservoir's function may be severely affected by the decrease of runoff into the reservoir. In this paper, runoff data from 1960 to 2012 and climatic data from 20 meteorological stations of Hanjiang River basin between 1960 and 2012 were used to analyze and compare the changing trends of runoff, precipitation and temperature by using the Mann-Kendall test method and the Pettitt test method, aiming at finding reasons for the decrease of reservoir inflow. Then, the climate elasticity model is adopted to estimate the degree of climate change effect on flow decrease. Our results indicated that climate change was the dominant reason for the inflow decrease in spring and autumn. In spring, climate change accounted for 67% of the decrease of reservoir inflow and human activities accounted for 33%; in autumn, climate change accounted for 88% of inflow decrease and human activities accounted for 12%.

This paper aims to conduct a 3D geological model of a mountainous area in Chongqing by means of a software platform, namely, Groundwater Modelling System (GMS, after comprehensive analyzing data on geological maps, geological cross-section maps and borehole data. The simulation results showed that:sandstone-mudstone interbed of the deposition sequences and the distribution of different lithology were consistent with the existing area data; the ratio of the model profile area to the actual geological sections area was about 1.15, the absolute relative error of the main lithology area percentage were less than 15%. We built a 3D hydrogeology model based on the geological model for groundwater numerical simulation, which reflected the value of 3D geological model in the field of groundwater numerical simulation. This 3D geological model not only provides reference for geological structure of mountain region, but also lay a foundation for the future work related to groundwater research.

Drought, one of the most devastating natural hazards, has caused tremendous damage in Southeast China. It is of great importance to evaluate and investigate the spatial and temporal evolution of drought in Southeast China. Based on the monthly precipitation and temperature data in Southeast China during 1962-2012, this study applied the standard precipitation and evaporation index (SPEI), linear analysis, Mann-Kendall test, wavelet analysis as well as rotated empirical orthogonal function (REOF) methods to analyze drought dynamics in Southeast China. The results showed that:(1) A significant rising trend of drought frequency has been detected in Southeast both at annual and seasonal scales, especially since 2000; autumn and winter presented frequent occurrence of drought in Southeast while the trend of drought in autumn has become more significant; there was no obvious mutation point in the annual SPEI; 2 to 8 years period of oscillation has been found in the area when researching the annual and seasonal variation of drought. (2) According to the front six spatial modes disassembled from REOF, we divided Southwest China into six different areas with totally different aridity and wetness features:the eastern Yunnan Plateau, Hanzhong Basin, the eastern mountainous area, the western Yunnan Plateau, Sichuan Basin and the Guizhou Plateau. Among the six different areas, the eastern Yunnan plateau has an obvious dry trend, while Hanzhong basin and eastern mountain area have a wet trend (but not obvious); also, the variation of the six areas generally exists 2-6 years period of oscillation. The study will contribute to exploring the spatial-temporal variability of dryness conditions in Southeast China while allowing a prediction of possible future drought distribution and trends under global climate change.

Based on the low-flow data in 1960~2009 in the Three-River Headwaters Region (TRHR), the trends of monthly runoff and the recession coefficient in the runoff recession process were analyzed, and the low-flow process was reconstructed by the monthly runoff recession coefficient. The results showed that:(1) the monthly runoff showed a decreasing trend in the low-flow period both in the Yellow Rivers Headwaters Region (YERHR) and the Lantsang Rivers Headwaters Region (LARHR), and the trend showed a weak increase in the Yangtze Rivers Headwaters Region (YARHR). The low-flow was controlled by summer-flow, and both of them kept a consistent changing trend. (2) The runoff recession coefficient in runoff recession process showed an increasing trend both in the YERHR and YARHR, which means that the runoff recession process was slowed. This is due to increasing groundwater capacity under long-term climate warming. The trend of runoff recession coefficient was not obvious in the LARHR. (3) the winter (from November to February) runoff recession process can be reconstructed by mean monthly runoff recession coefficient and the runoff of October in the TRHR with high simulation accuracy, the difference between the calculated and observed value were less than 8%. The study can provide beneficial reference for understanding the characteristics of low-flow under the background of climate change and hydrologic forecasting in the Three-River Headwaters Region.