首页> 外文会议>IEEE International Geoscience and Remote Sensing Symposium >Multitemporal monitoring of the forest cover in Côte d'Ivoire from the 1960s to the 2000s, using Landsat satellite images
【24h】

Multitemporal monitoring of the forest cover in Côte d'Ivoire from the 1960s to the 2000s, using Landsat satellite images

机译:利用Landsat卫星图像对科特迪瓦1960年代至2000年代的森林覆盖物进行多时相监测

获取原文

摘要

The phenomenon of deforestation is a reality in Côte d'Ivoire, which dates back to the colonial period [1]. This study aims to determine the spatial and temporal evolution of dense rainforests using remote sensing data. Many previous studies have attempted to determine the area of forest cover in Côte d'Ivoire ([2], [3], [4], [5], [6] and [7]). Most of these studies were conducted from either aerial photographs or satellite images of low spatial resolutions such as NOAA / AVHRR [8]. However, through these studies, methods of estimates of the area of forest cover generally vary from one author to another, thus causing an inconsistency in the results [9]. Remote sensing allows a better assessment of the scope and scale of the problem of deforestation. Using multi-temporal data, one can better detect and analyze changes in the forest cover. By comparing images of previous years with recent scenes, it becomes possible to effectively measure the differences in the extent of deforestation and loss of rainforests in Côte d'Ivoire ([10] and [11]). In a first phase, the country's dense forest cover of the 1960s was determined from the digitization of existing vegetation map. In a second phase, the dense forest cover of the 1980s and 2000s were obtained by processing of Landsat TM and ETM+ satellite images. First the geometric correction of the satellite images was performed. Then, series of supervised classifications based on the maximum likelihood method were done to generate maps of land cover, with overall accuracies of 90%. From these land cover maps, the class of dense forests was extracted and converted into a map layer in vector format. By crossing dense forests covers of the 1980s and 2000 with those of the 1960s, the specific characterizations of those forests were derived (Figure 1, 2, 3). The results show a dense humid forest cover of about 8.4 million ha in 1960, with the presence of large blocks. The 1980s and 2000s have dense forest cover of 2.6 million ha and 1.35 million ha, respectively (Tables 1, 2, 3). These latter periods are characterized by high loss and fragmentation of rainforests in Côte d'Ivoire. Thus, since its independence, the country has lost more than 80% of its forest cover. The results of this study show that clearing associated with industrial agriculture mainly cocoa, coffee and rubber farming as well as demographic pressure and abusive exploitation of valuable wood species for export are the main causes of deforestation in Cote d'Ivoire. This study advocates the restoration of forests through the implementation of an awareness policy and reforestation one hand, and an efficient monitoring and protection of national protected areas and parks.
机译:毁林现象在科特迪瓦是现实,可追溯到殖民时期[1]。这项研究旨在利用遥感数据确定茂密雨林的时空演变。先前的许多研究都试图确定科特迪瓦的森林覆盖面积([2],[3],[4],[5],[6]和[7])。这些研究大多数是通过航空摄影或低空间分辨率的卫星图像进行的,例如NOAA / AVHRR [8]。但是,通过这些研究,估计森林覆盖面积的方法通常因作者而异,从而导致结果不一致[9]。遥感可以更好地评估毁林问题的范围和规模。使用多时相数据,可以更好地检测和分析森林覆盖率的变化。通过将前几年的图像与最近的场景进行比较,可以有效地衡量科特迪瓦森林砍伐和雨林丧失程度的差异([10]和[11])。在第一阶段,该国1960年代茂密的森林覆盖是根据现有植被图的数字化确定的。在第二阶段,通过处理Landsat TM和ETM +卫星图像获得了1980年代和2000年代茂密的森林覆盖。首先,对卫星图像进行了几何校正。然后,基于最大似然法进行了一系列监督分类,以生成土地覆盖图,总精度为90%。从这些土地覆盖地图中,提取了茂密的森林类别,并将其转换为矢量格式的地图图层。通过使1980年代和2000年的茂密森林覆盖物与1960年代的森林覆盖物交叉,得出了这些森林的具体特征(图1、2、3)。结果表明,到1960年,茂密的湿润森林覆盖约840万公顷,并存在大块地。 1980年代和2000年代森林茂密,分别为260万公顷和135万公顷(表1、2、3)。后几个时期的特点是科特迪瓦雨林大量流失和破碎。因此,自独立以来,该国已经丧失了80%以上的森林覆盖率。这项研究的结果表明,与工业农业有关的砍伐活动主要是可可,咖啡和橡胶耕作,以及人口压力和对有价值的木材物种的滥用开采是科特迪瓦毁林的主要原因。这项研究主张通过实施一项认识政策和一方面进行植树造林,以及对国家保护区和公园进行有效的监测和保护,来倡导恢复森林。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号