| Land use and land cover change (LUCC) has been a hot spot issue and a frontier subject in the research area of global change research. Rubber plantation expansion is one of the most widespread land-use conversions in the north edge of the tropical regions of southern Yunnan. This land-use conversion has often important social, economic and ecological implications. Therefore, accurately mapping the spatio-temporal patterns of rubber plantation expansion, and their related physical environment driving factors, and thereby efficiently simulating the future change scenarios using LUCC model, are vital to better understanding and more thorough evalution of the regional eco-environment consequences of LUCC.This work takes Mengla County as a study case. Four Landsat TM/ETM+/OLI images acquired in2002and2014, together with field GPS measurements, DEM, forest resource inventory, medium-and long-term land use planning, and the statistics on social and economic development, are employed to map and analyze the local dynamic changes of land use/land cover in the13-year intervals. The land use map in2014was simulted using the CLUE-S model based on the initial land use map in2002. Thereby, the accuaracy of the resultant map was evaluated through comparing with the land use classification map retrieved from the Landsat OLI data of2014. Further to this, the model parameters were optimized.According to the land planning data of Mengla County from2010to2020, the range of change for each type of land is acquired. The demand of each type of land in2030is calculated according to that range and the land use conditions in Mengla County in2020and in2030are simulated.The results show that:(1) The overall accuracy and Kappa coefficient of classifying land use/land cover and the is92.61%and0.91in2002, respectively and91.27%and0.89in2014, respectively. Regarding the classification accuracy for each of land use/land cover types, waters had the highest, following by forests, rubber plantations, and built land, and garden land and arable land the lowest.(2) The dominant land use and land cover types are forests and garden land in2002and are forests and rubber plantations in2014. The most pronounced LUCC from2002to2014is the conversion to rubber plantations from the other land use/land cover types, which led to a increase of the rubber plantation areas by578.87km2.(3) The spatial distribution of land use/land cover is affected by terrain factors. Rubber plantations mainly occur in southward slope below1000m above sea level with a gradient of8°～35°, built land and waters in flat areas below600meters with a slope gradient below8°, and forests above1000m with a slope gradient steeper than15°. However, garden land and arable land exist at a wide range of altitudes with a gentle gradient, mostly at low altitudes.(4) Over the period from2002to2014, accompanied by a dramatic increase of the rubber plantation areas, the areas of built land slightly increased, whereas the areas of forests, garden land, arable land and waters decreased with different degrees. Spatially, the expansion of rubber plantations mainly occurred in the areas of an altitude of600-1000m in the southern and southwestern parts of the study area. A little rubber plantation gain appeared in the areas of an altitude of1000-1200m. The loss of forests, garden land, and arable area spread over the whole study area, and the gain of built land occurred around towns in patchy forms.(5) Based on the screened and optimized parameters, the spatial pattern of land use/land cover in2014was simulated by CLUE-S model with a Kappa coefficient of66.99%.(6) According to the local land use planning over the period of2014-2030, rubber plantations will forcibly be transformed to the other land use types. The simulated results showed that the planned loss of rubber plantation areas will mainly transform into forests, occurring in the towns of Menglun, Guanlei, Mengpeng and Mengla, whereas some of them around the towns of Mengla, Mengpeng and Menglun will mainly convert to arable land. |
PDF Full Text Request