资源描述
本科毕业设计（论文）外 文 翻 译翻译二Spatial-Temporal Pattern and Driving Forces of Land Use Changes in Xia menQUAN Bin, CHEN Jian-Fei, QIU Hong-Lie, M. J. M. ROMKENS, YANG Xiao-Qi,JIANG Shi-Feng and LI Bi-ChengABSTRACTUsing Landsat TM data of 1988, 1998 and 2001, the dynamic process of the spatial-temporal characteristics of land use changes during 13 years from 1988 to 2001 in the special economic zone of Xiamen, China was analyzed to improve understanding and to find the driving forces of land use change so that sustainable land utilization could be practiced. During the 13 years cropland decreased remarkably by nearly 11 304.95 ha. The areas of rural-urban construction and water body increased by 10152.24 ha and 848.94 ha, respectively. From 1988 to 2001, 52.5% of the lost cropland was converted into rural-urban industrial land. Rapid urbanization contributed to a great change in the rate of cropland land use during these years. Land-reclamation also contributed to a decrease in water body area as well as marine ecological and environmental destruction. In the study area 1) urbanization and industrialization, 2) infrastructure and agricultural intensification, 3) increased affluence of the farming community, and 4) policy factors have driven the land use changes. Possible sustainable land use measures included construction of a land management system, land planning, development of potential land resources, new technology applications, and marine ecological and environmental protection.Key Words: driving force, GIs, land use change, remote sensing, XiamenINTRODUCTIONLand use and land cover are prominent ecological symbols within the surface system of the earth. Land use refers to human manipulation of the land to fulfill a need or want. Meanwhile, land use change may involve either a shift to a different use, such as from rice paddy to aquaculture, or an expansion and intensification of an existing form, such as from subsistence to commercial farming (Matson et al., 1997). Land cover, defined as the physical surface condition of the land, is likely to change as a result of land use change (Turner and Meyer, 1991). Furthermore, land use influences the environment mainly by land cover, and thus land use and land cover are inter-related.Land use/cover change (LUCC) is a core project of the International Global and Biology Plan (IGBP). It aims to improve understanding of the global dynamics of LUCC with a focus to improve the ability to project such change (Turner et al.; 1997). More and more people believe that it is a timely project to comprehensively assess the global environmental changes (Liu and Buhe, 2000a). LUCC studies the changes of natural land, socio-economic conditions, and human activities. Therefore, it requires the cooperation of natural and social sciences to link LUCC to global change (Turner, 1994). LUCC revolves around core problems of regional population, resources, environment, and development. Since the 199Os, the study of LUCC has been a subject of intense interest in academic circles. In recent years, some researchers have made great progress in LUCC studies (Meyer and Turner, 1996; Luo and Ni, 2000; Shi et al., 2002). However, few studies have been done to date in the southeastcrn part of Fujian Province, which experienced major economic development during the past 20 years. Currently, the rate of conversion of agricultural land in the southeastern coastal area of China to non-agricultural uses is increasing (Liu et al., 2003). Consequently, there is a need for more research in the southeastern Fujian Province, where rapid development has led to swift changes in land use patterns. In this work, land use spatial changes during 1988, 1998, and 2001 in Xiamen were studied using remote sensing (RS) and geographical information system (GIS) tools. The characteristics and rules of land use changes and their driving forces were analyzed quantitatively through models, which provided a scientific basis for decisions in regional resource and coordinated environmental development, whilst offered a typical case in land use change in one of Chinas “hot spots” of economic development. MATERIALS AND METHODSSurvey of regionXiamen, with an area of 1638 km2, is located in the southeastern part of Fujian Province, facing the Taiwan Straits. It has a southern subtropical monsoon climate, an annual mean temperature of 20.8 “C, and an annual precipitation of 1143.5 mm. The natural vegetation is a southern-subtropical monsoon rainforest, but human activities have destroyed most of this. Masson Pine (Pinus massoniana Lamb.) and Taiwan Acacia (Acacia confusa Merr.) have been planted in the upland and bottom flat land, under which a lateritic red soil has developed over time (Quan et al., 2004b and 2005a). In 2001, Xiamen consisted of seven administrative districts including Siming, Kaiyuan, Gulangyu, Huli, Jimei, Xinglin, and Tongan Districts with a total population of 1.31 million. When China began a policy of opening up to the world, Xiamen became one of the first four special economic zones because of its advantageous location. Since then the economy developed rapidly.Data and classificationLand use data were obtained from Landsat satellite images from 1988 to 1998 and 2001 with a spatial resolution of 30 m x 30 m. In addition, maps of Xiamens vegetation distribution, Xiamen City remote sensing images, Xiamen administrative district (2004), and land use etc. were collected for image-interpretation. Land resource investigation data (1988-2001) were also gathered for consultation. Two grades were established for the land use classification system. The first grade was divided into six classes: cropland (l), orchard (2), forestland ( 3 ) , rural-urban industrial land (4), water body (5) and unused land (6). The second grade was divided into 12 classes with the following names and codes: paddy field (ll), dry land (12), orchard (21), forest (31), urban, town and separated industrial land (41), rural land (42), salt field (43), reservoir (51), other water bodies (52), coastal beach (53), barren land (61), and other unused land (62).ProceduresFor land use data and conversions from 1988 to 2001, first Landsat TM images of three different periods were acquired, and then the GCP (ground control points) works module of Canadian PCI software was applied for making geometric corrections. More than forty ground control points were selected as references on a topography map of scale 1:50000. The Gauss-Kruger projection, which belonged to a kind of transverse equal-angle cylindrical projection, was used to correct images with its projection parameter as follows: central longitude 117”, Krassovsky ellipsoid, and false easting 500 km (Chang, 2002; Chen et al., 1998). Color composites were generated displaying bands 5, 4 and 3 as red, green and blue, respectively. An image enhancement was performed to increase the visual distinction between features in order to increase the amount of information that can be visually interpreted from the data. After image enhancement, based on field investigations, image interpretation symbols of different image elements were added. A global positioning system (GPS) receiver was used to collect the coordinates of sample sites. Additionally, the land use map of 1996 was digitized in GIS ArcView software before image-interpretation. This could be consulted in the process of personmachine alternating visual operations. The land use types were interpreted visually in the screen based on the TM images. Also, some additional errors were corrected based on auxiliary reference data and fieldwork. In the end, the smallest plot on the interpreted map corresponded to a scale of 1:100000, and field checking verified the accuracy of image interpretation of up to 90%. To determine the rate of land use change, the study period 1988-2001 was divided into two subperiods and the land use changes of the two sub-periods were compared. The first sub-period was from 1988 to 1998, called the earlier stage, and the second sub-period was from 1998 to 2001, called the later stage. The comparative analysis on land use change focused on the two sub-periods. Regional difference in land use characteristics were determined using the land use dynamic degree model that could be mathematically expressed by the following relationship (Liu and Buhe, 2OOOb): where S is the land use change rate over time t, Si is the ith type land use area at the beginning of the monitoring period, n is the number of the land use type, and ASi-j is the total area of the ith type land use that is converted into the other types of land use. The land use dynamic degree was thus defined as the time rate change of land use that was converted into the other types of land uses and that at the beginning of the monitoring period was part of the land use area subject to change. The dynamic degree represented, in a comprehensive manner, the change of land use in a given region. In order to understand the rate of regional land use changes and their characteristic differences, the land use dynamic degree was calculated for the administrative districts in Xiamen. The seven administrative districts were divided into three groups on the basis of the value of the land use dynamic degree. The first group belonged to the fast mode of land use change, in which the land use dynamic degree was from 25% to 40%. The second group belonged to the moderate mode of land use change where the dynamic degree was from 12% to 25% (including 25%), and the third group belonged to the slow mode with the dynamic degree from 8% to 12% (including 12%). After each group was encoded, the regional distribution map for the land use dynamic degree was made using GIS ArcView software. Regional differences in the rate of land use change were determined with the rate of land use change model as follows (Liu and He, 2002): where S is the rate of the ith type land use change during the monitoring period TI to Tz; Ai is the area of the ith type land use at the beginning of the monitoring period; and UAi is the area of the ith type land use that remains unchanged during this monitoring period. (Ai - UAi) is the changed land area during the monitoring period, i e . the total area of the ith type land use converted into the other types of land use; Thus, this model represented the time rate of change for one type of land use that was converted into another type of land use relative to the land use situation at the beginning of the monitoring period.Regional differences in the land use intensity comprehensive index were calculated using the mathematical expression given by the following relationship (Lai et al., 2002; He et al., 2002):where I is the land use intensity comprehensive index; G; is the gradation value of the ith ranking land use type; Ci is the area percentage of the ith ranking land use intensity; and n is the number of land index of land use was given by:where Ib and I, are land use intensity comprehensive indexes at time point b and a, respectively. These relationships were comprehensive representations of land use intensity. If the parameter AIb-, 0, the land use is continuely developing in the region; on the contrary, the land use is regressing. Land use can be divided into several rankings according to their change intensity to natural “equilibrium status” (Wang and Bao, 1999). In the gradation index system, unused land was assigned the factor 1; forest and water body were factor 2; agricultural land, including cropland and orchard, was factor 3; and rural-urban industrial land was factor 4. So, the calculation represented the range andintensity of land utilization (Wang et al., 2002). To determine the driving forces of cropland change a comparison of the Landsat acquired data was made with the statistical data obtained from the Land Resource Survey Office. The statistical data of the Xiamen cropland area for each year and the corresponding social-economic data were also collected and analyzed. The social-economic data included general population, total agricultural output value, GDP, etc. There were 23 indexes and the data sets covered each year for the 1988 to 2001 study period. The information was calculated on the basis of no change in the prices from 1990. Then a correlation analysis was conducted between cropland and the other factors to assess cropland change. Driving forces of other land use types were also analyzed to help develop strategies for sustainable development.RESULTS AND DISCUSSIONLand use data and conversions from 1988 to 2001The spatial-temporal land use changes in Xiamen are shown in Table I and Fig. 1. The data indicated that three land use types increased while three decreased from 1988 to 2001. Rural-urban industrial land had the largest increase with 10 152.24 ha followed by orchard with 1635.84 ha (Table I). Due to hydro-technical construction projects, water body increased by 848.94 ha on the whole while coastal beach land decreased by 2 139.07 ha (Fig. 1). Among the land types, cropland decreased by 11 304.95 ha, while forest and unused land decreased by 727.90 and 604.15 ha, respectively (Table I).Land use conversion was common among the various types. About 52.5% of the cropland area lost was converted into rural-urban industrial land, and 27.9% and 16.6% were converted into orchard and water bodies, respectively (Table 11). During the study period, many paddy fields were converted to rural-urban industrial use, orchard, reservoir, and hydro-technical construction sites. In addition, part Fig. 1 Net changes of land use in Xiamen from 1988 to 2001 with paddy field ( l l ) , dry land (12), orchard (21), forest (31), urban, town and separated industrial land (41), rural land (42), salt field (43), reservoir (51), other water body (52), coastal beach (53), barren land (61), and other unused land (62). The number in the parenthesis respresents the ranking code of land use type of the second grade according to the land use classification system.of the orchard area was converted into cropland and rural-urban industrial land. However, the total orchard area increased because of conversions from cropland (62.9%) and forest (20.8%), respectively. About 50.8% of the forest land lost was converted into orchard. Rural-urban industrial land and cropland made up the rest of the converted forest. About 53.6% of the water body lost was converted into ruralurban industrial land (Table 11). In the meantime, the conversion of some cropland into reservoir and hydro-technical construction land led to the increase in the total water body area. The increase in ruralurban industrial land was most noticeable (Fig. l), which came from other land use types, especially cropland. 62.1% of the lost area of the unused land was converted into forest, and the rest was converted mainly into orchard (Table 11).Comparison of land use changes between 1988 to 1998 and 1998 to 2001The rate of land use change for the study period 1988-2001 is shown in Fig. 2 with a larger decrease of cropland ( i e . paddy fields, 10575 ha) during the earlier stage than the later stage (730 ha). The decrease in the earlier stage was about 14 times that of the later stage while the period of observation in the later stage was about one third of the earlier stage. Thus, the data suggested that the disappearance rate of cropland had slowed. During the two sub-periods both of the rural-urban industrial land use changes increased (Table I) and the ratio of the increase between the two stages was about 12.4:l. Land areas for orchard and water bodies increased during the earlier stage while no major change took place during the later stage. Forest showed over twice as great a change in the first period than the second. In short, the basic rule of land use change in Xiamen during the period 1988 to 2001 was that changes in the early stage were greater than those in the later stage. This meant that changes in land use gradually decreased with time suggesting a more rational utilization of land resources.Fig. 2 Net changes in land use areas of Xiamen for the earlier stage (1988-1998) and the later stage (1998-2001) with paddy field (ll), dry land (12), orchard (21), forest (31), urban, town and separated industrial land (41), rural land (42), salt field (43), reservoir (51), other water body (52), coastal beach (53), barren land (61), and other unused land (62). The number in the parenthesis represents the ranking code of land use type of the second grade according to the land useclassification system.The land use dynamic degree for the seven administrative districts in Xiamen is shown in Fig.3. The overall comprehensive land use dynamic degree of Xiamen for the period from 1988 to 2001 was 21.1%. Fig. 3 indicated that the Huli and Xinglin Districts were in the first group. Their high ranking was attributed to their favorable geographical location due to the presence of shipping, transportation, and industrial facilities. These districts became a base of “exchanges of mail, trade, air and shipping services” on both sides of the Taiwan Straits. Therefore, these two districts played an important role in capital investments from Taiwan. In 1989, Haicang of Xiamen also became an area of investment interest for Taiwanese businessmen, who progressively promoted local economic development .Fig. 3 Dynamic degree of land use for the seven administrative districts in Xiamen during the period 1988 to 2001 with fast change being from 25% to 40%, moderate change from 12% to 25% (including 25%), and slow change from 8% to 12% (including 12%).The Jimei, Tongan, and Kaiyuan Administrative Districts belonged to the second group. They experienced a moderate land use change (Fig. 3). Siming and Gulangyu districts belonged to the third group, which showed a slow land use change. Siming was an old urban district in which the urbanization level was high to start with and therefore a further increase would have been difficult to achieve. The slow change on Gulangyu Island could be related to regulations that were designed to preserve the characteristic architectural style and the greenery.The rates of land use change for Xiamen and its administrative districts were calculated for the period 1988 to 2001 (Table 111). Among the various land use types, the cropland annual conversion rate was the highest. This was indicative of the rapid land use change in this region. A comparison of cropland land use change rates among the various administrative districts showed that Siming District had the highest rate of cropland land use change. This change was related to the rapid urbanization during the period 1988 to 2001. The orchard land use change rate in Huli District was the largest and orchards were mainly converted into rural-urban industrial land (Table 11). Most of the water body changes involved converting coastal beach into rural-urban industrial land. In the Xiamen Region, the forest change rate of the Huli District was the highest (Table 111). Losses of forest in Table I1 were mainly converted into orchard and rural-urban industrial land. Changes with unused land occurred only in Xinglin and Tongan Districts (Table 111), where the losses were mainly converted into forest and orchard (Table 11). This trend may be related to the influence of the “Making Green with Trees” Policy.Regional differences in the land use degree changeUsing the model of Wang and Bao (1999), the land use degree change parameter AI that expressed the change in the land use intensity index was 6.91 (Table IV). Since this was greater than 0, it indicated that the rural-urban industrial land areas could be increasing, and land in the region was becoming more intensively used. Table IV also showed that land use intensity was gradually increasing over time in Xiamen. Comparison of
展开阅读全文