Research data
收藏资源简介：
The present study focused in identifying and monitoring land use/land cover changes of the Bamenda mountain forest (North West region of Cameroon) using remote sensing data from 1978 to 2010 and GIS technology. To begin with, the massive change observed on the study site was largely attributed anthropogenic factors rather than natural disturbances. to keep going, it was deemed necessary that determining the magnitude of change of the Bamenda mountain chain forest could be effective and important to understand the rate of the changes. Moreover, Mapping, evaluating and predicting the trends of major land use/land cover changes of this site could be possible and useful. The methodological approach adopted for the research work was an interdisciplinary approach technique which combined the qualitative research technique from a social science perspective with a remote sensing and GIS technique from a geographical perspective. The remote sensing data-set used for the research work was cloud free and of the dry season, downloaded and classified in support of ground control points and google earth images. reconnaissance information gathered from the field of study was also analyzed in support of the issue being investigated. To better understand this, land sat MSS, TM and ETM+ for the years 1978, 1988 and 2010 respectively were used. A false color composition using bands 5, 4 and 3 was performed on the images to identify the different LULC types of the study area. These images were later on classified using the NDVI analysis for the identification of the changes in land cover and land use types. The NDVI results obtained were reclassified with the help of a supervised classification technique using maximum likelihood algorithm where the main land use and land cover categories were identified and mapped to understand the trends of the changes that occurred during the studied dates. The research revealed a significant loss of the vegetation cover of the Bamenda Mountain, which was the focus of interest of the study. According to the research findings, from 1978 to 1988, bare rocks, built-up, cultivated lands and savannah all increased at the detriment of the forest that dropped to 48.68% covering a surface area of -4520.49km2. Likewise from 1988 to 2010, both the forest and savannah dropped to 1.32% and 19.23% covering surface areas of -7046km2 and -1026.56km2 respectively. This thus showed the usefulness of remote sensing and GIS in pairing up less recent images to more recent ones in examining and demonstrating changes for a vast ecosystem and for a long period of time. This necessitated some proposed solutions such as employment of competent forest guards, reforestation and afforestation, and even improving technologies to combat the rate of forest degradation of this site which is under risk of invasion.
本研究聚焦于利用1978年至2010年的遥感（Remote Sensing）数据与地理信息系统（Geographic Information System，简称GIS）技术，对喀麦隆西北部地区的巴门达山地森林的土地利用/土地覆盖（Land Use/Land Cover，简称LULC）变化进行识别与监测。研究初期发现，该研究区域内出现的大规模变化主要归因于人为活动，而非自然干扰。后续研究表明，量化巴门达山地森林的变化幅度，对于理解该区域的变化速率具有关键意义。此外，对该区域主要土地利用/土地覆盖变化趋势开展制图、评估与预测，同样具备可行性与实用价值。本研究采用跨学科研究方法，融合了社会科学视角下的定性研究技术，以及地理学视角下的遥感与地理信息系统技术。本次研究使用的遥感数据集为无云的旱季影像，下载后结合地面控制点与谷歌地球（Google Earth）影像进行分类处理。同时，对研究区域实地调研获取的勘测信息也进行了分析，以辅助本次研究议题的探究。为更深入地开展分析，本研究分别采用了1978年、1988年与2010年的陆地卫星（Landsat）MSS、TM与ETM+影像。对影像进行了5、4、3波段的假彩色合成，以识别研究区域内不同的土地利用/土地覆盖类型。随后通过归一化植被指数（Normalized Difference Vegetation Index，简称NDVI）分析对影像进行处理，以识别土地利用与土地覆盖类型的变化。所得归一化植被指数结果借助最大似然算法的监督分类技术进行重分类，以此识别并制图展示主要土地利用与土地覆盖类别，进而明晰研究时段内发生的变化趋势。研究显示，作为本次研究核心关注对象的巴门达山地植被覆盖出现了显著流失。根据研究结果，1978年至1988年间，裸岩、建成区、耕地与稀树草原面积均呈增长趋势，而森林占比下降至48.68%，表面积减少4520.49平方千米。同样，1988年至2010年间，森林与稀树草原占比分别下降至1.32%与19.23%，对应表面积分别减少7046平方千米与1026.56平方千米。上述结果印证了遥感与地理信息系统技术在跨年代影像对比分析中的实用性，可有效用于大范围生态系统的长期变化监测与展示。针对该区域正面临入侵风险的现状，本研究提出了相应的解决方案，包括配备合格的森林护林员、开展造林与再造林工程，以及升级相关技术以遏制该区域的森林退化速率。
