The Use of Landsat and Sentinel-2 for Detecting Gold Mining in Colombia Based on River Water Turbidity

The Chocó department suffers from extraordinary amounts of alluvial gold mining resulting in increased river turbidity. A gap has been identified where a replicable means of identifying such turbidity was absent, and a developed method shrinks the gap with Landsat and Sentinel imagery; a GIS is used to assess the temporal mining change of Chocó and utilise a Normalised Difference Turbidity Index to quantify turbidity. Mitigating effects clouds had on results was a challenge, and time constraints meant additional in-situ values could not be acquired. Nonetheless, turbidity quantified easily using the index and related well to identified mining sites highlighted from NDVI change. Classifications were accurate when verified with photographic evidence and revealed mining on the Quito profoundly increased turbidity for the Atrato while also identifying other sites. The method used simple tools and open-source data that can benefit decision-makers and communities of such regions, allowing them to monitor their own rivers. Correspondence details: [e-mail] Figures in this paper include OS Data © Crown copyright and database [2018], Digimap License. This research did not receive any funding from agencies or other sectors.乔科省（Chocó department）面临着规模空前的砂金开采活动，导致河流浊度显著升高。当前仍存在一项研究空白——尚无可复现的河流浊度识别方法，而本研究提出的方法借助陆地卫星（Landsat）与哨兵（Sentinel）影像填补了这一空白：研究采用地理信息系统（Geographic Information System，GIS）分析乔科省采砂活动的时序变化，并借助归一化浊度指数（Normalised Difference Turbidity Index，NDTI）对浊度进行量化。 消除云层对实验结果的干扰是本研究面临的一大挑战，而时间限制导致无法获取更多原位（in-situ）实测数据。尽管如此，借助该指数可便捷地完成浊度量化，且量化结果与通过归一化植被指数（Normalized Difference Vegetation Index，NDVI）变化识别出的采砂点位高度相关。经影像证据验证，本分类结果准确度较高，结果显示在基多（Quito）区域开展的采砂活动显著加剧了阿特拉托河（Atrato）的浊度，同时还识别出了其他采砂点位。本方法采用简易工具与开源数据，可为此类区域的决策者与当地社区提供支持，使其能够自主开展河流监测工作。 通讯邮箱：[e-mail] 本文所用图片包含英国国家测绘局（Ordnance Survey, OS）数据，© 皇家版权及数据库许可[2018]，依据Digimap许可协议使用。 本研究未获得任何机构或部门的资助。