中文摘要：

福建省长汀县曾是我国南方红壤地区水土流失最严重的县份之一,经过20多年的艰辛努力,长汀已成为中国水土流失治理的典范。采用遥感技术和景观格局分析技术,基于1988、1998、2004、2009和2011年的遥感影像,对长汀县水土流失最为严重的河田盆地区进行土地利用动态变化检测与景观格局变化分析。结果表明,研究区在这23a间的土地利用发生了很大变化,其中最主要的特征就是以针叶林为主的林地面积的快速增长和地表裸土面积的大幅下降。景观分析表明,水土流失治理新增的小块林地正逐渐形成连片分布,而裸土面积在大幅减少的同时,其斑块也趋于破碎。总的看来,这23a间的水土流失治理已使得研究区的生态明显趋于好转。

英文摘要：

County Changting in Fujian province was one of the most severe reddish soil erosion regions in South China. After more than 20-year efforts, County Changting has achieved a great success in the treatment of the soil loss and thus become a model in China in this regard. Therefore, the study on the land-use dynamics responding to the county＇s soil-loss treatment is particularly meaningful. This study used remote sensing technology to analyze land use and landscape dynamics in the Hetian basinal area for a long period from 1988 to 2011 because this area is the most serious soil erosion area in the county. Five remote sensing images were used in the study, which represent the years of 1988, 1998, 2004, 2009, and 2011, respectively. The latter three images are SPOT 5 10m resolution muhispectral images, while the 1988 and 1998 images are Landsat TM 30m resolution muhispectral images. To keep consistent in the spatial resolution with SPOT 5 images, therefore, the 30m TM images were fused with near-date-coincident 10m SPOT Pan images, respectively, to achieve a 30m resolution. The supervised maximum likelihood classification and change detection technique were used to reveal the land use dynamics. In addition, the spatial pattern analysis was carried out to study accompanied landscape pattern variations. The results show that the land use in the Hetian basinal area has greatly changed during the 23 study years. The forest land and built-up land have increased rapidly, while the bare soil land dropped considerably. Spatialpattern analysis indicates that while decreased, the bare soil has also been fragmentized since 1988. This suggests that the local eco-environment has been substantial improved during the study period.