Mining in the Kizel coal basin (Perm Region, Russia) ceased 20 years ago; however, AMD with high levels of total iron (Fe_total) and trace elements still affects the rivers. In this study, we attempted to estimate inter-annual and seasonal variability of AMD-related surface water contamination using Sentinel-2 images from 2016 to 2019. The acid mine water index (AMWI), which is a normalized difference of spectral reflectance in the red and blue bands, was calculated from Sentinel-2 images. We compared the AMWI values with measured Fe_total concentrations in the surface water. A statistically significant (at a 0.05 significance level) Spearman’s rank correlation between AMWI and Fe_total concentration was found for 5 out of 7 surface water sampling points. We found that surface water contamination reaches a seasonal maximum in July, 1–1.5 months after the end of the snowmelt high water period. Excessive summer rainfalls also contributes to increased contamination, causing contamination to possibly spread more than 200 km from the AMD sources. In contrast, arid summer conditions were associated with a substantially decreased AMD discharge and Fe_total concentrations in the surface water. The main uncertainties in our results are associated with the effect of contaminated bottom and bank sediments and suspended sediments on the spectral characteristics of the water surface, and the relatively coarse (10 m) spatial resolution of Sentinel-2 images. However, despite the data and method limitations, our results show that Sentinel-2 images have substantial potential for monitoring AMD-related contamination of surface water.

Kizel煤盆地（俄罗斯彼尔姆地区）的采矿已于20年前停止；但是，总铁（Fe _total）和微量元素含量高的AMD仍会影响河流。在这项研究中，我们尝试使用Sentinel-2图像估算2016年至2019年与AMD相关的地表水污染的年际和季节变化。酸性矿泉水指数（AMWI）是光谱反射率的归一化差异。红色和蓝色带是从Sentinel-2图像计算得出的。我们将AMWI值与测得的地表水中Fe_总浓度进行了比较。AMWI和Fe_总量之间Spearman等级相关性具有统计意义（显着性水平为0.05）在7个地表水采样点中，有5个被发现浓度最高。我们发现，积雪融化的高水期结束后的1–1.5个月，地表水污染在7月达到了季节性最大值。夏季过多的降雨也加剧了污染，导致污染可能从AMD来源扩散200多公里。与此相反，干旱夏季条件与基本上降低AMD放电和Fe相关联_总地表水中的浓度。我们结果的主要不确定性与受污染的底部和河岸沉积物以及悬浮的沉积物对水面光谱特征的影响以及Sentinel-2图像相对较粗的（10 m）空间分辨率有关。然而，尽管数据和方法有局限性，我们的结果表明Sentinel-2图像具有监测AMD相关的地表水污染的巨大潜力。