基于热红外遥感的全球核电站温排水热污染研究

        数十年来，全球海岸带或大湖沿岸建立了数百个核电站，核电站运行产生的高温冷却水通常排放到邻近的水体中，对生态系统、水体物理过程、沿岸经济产业和居民健康生活等方面都可能造成严重的后果。在目前对核电站温排水监测技术中，热红外遥感以实时、大范围、高精度的优势得到广泛应用。目前对核电站温排水热污染的热红外遥感研究主要局限包括：环境基准温度的选取缺少时空稳健性，缺少在热噪声条件下精确的温排水热羽提取方法，全球尺度核电站热污染的全面监测分析仍是空白。

        本研究选用常用于长时序热污染监测的中分辨率成像光谱仪（Moderate resolution imaging spectroradiometer, MODIS）和陆地卫星（Landsat）热红外遥感数据分别进行温排水热羽提取。本研究采用MODIS数据反演海表层温度（Sea Surface Temperature, SST），并提出一种基于热污染潜在影响范围的环境基准温度提取方法用于稳健地计算SST升温增量。同时定量评估了MODIS升温影像信噪比，并进一步提出一种分簇深度优先搜索算法对升温影像热羽进行提取，评估提取的热羽。结果表明，MODIS SST数据不适用于精细的温排水热羽分析。基于Landsat数据反演SST，通过筛选并标注热羽样本，将位置先验信息融入卷积神经网络，提出集成位置信息的深度学习语义分割模型。通过与主流阈值法进行对比实验表明，基于深度学习的模型在热噪声条件下具有更好的表现。

       本研究采用构建的深度学习模型与Landsat数据，对全球核电站温排水热羽进行提取，分析了温排水热羽的SST升温增量、面积与形态、特征间的相关性，讨论了热羽潜在影响因素。分析结果表明，本研究中所有目标核电站均在运行期间内均监测到较高SST升温增值，全球目标核电站平均最大SST升温增值为4.80 K，表明全球核电站在40年中广泛存在热污染超限现象。浅型排水对环境水表温度的影响较深型排水大。北美五大湖区域经常出现面积大于3 km²的大型热羽，而河口区域的核电站则以1 km²小型热羽为主。本文生成的全球目标核电站热羽出现频率图可作为未来研究的基线，且本文提出的核电站热羽提取流程可作为今后业务化监测的基础。

   For decades, hundreds of nuclear power plants have been built along coastlines or great lakes around the world. The high-temperature cooling water generated by the operation of nuclear power plants is usually discharged into the adjacent water bodies, which may pose serious impacts on the ecological system, the physical process of water bodies, the industries along with the coastal areas and the health of residents. Thermal infrared remote sensing has been widely used in monitoring the thermal drainage of nuclear power plants with the advantages of real-time, large range and high precision. At present, the main limitations of thermal infrared remote sensing research on thermal pollution of cooling discharge from nuclear power plants include: the selection of environmental reference temperature lacks spatiotemporal robustness, and an accurate extraction method of thermal plume under noisy conditions is unavailable, the comprehensive monitoring and analysis of thermal pollution of nuclear power plants on a global scale is still absent. 

   In this study, Moderate Resolution Imaging Spectroradiometer (MODIS) and Landsat thermal infrared remote sensing data, which are commonly used for long time series thermal pollution monitoring. Sea surface temperature (SST) was retrieved from MODIS data, and a method of calculating environmental reference temperature based on the potential affected area was proposed to compute SST increment. At the same time, the signal-to-noise ratio (SNR) of MODIS SST increment images was quantitatively evaluated. Furthermore, a cluster depth-first search algorithm was proposed to extract thermal plumes from images and the extracted thermal plumes were evaluated. Based on Landsat data, SST was further retrieved and environmental reference temperature was calculated to obtain SST increment. By screening and labeling thermal plume samples and integrating location prior information into the convolutional neural network, a deep learning semantic segmentation model integrated with location information was proposed. Comparing with mainstream threshold methods, the experimental results show that the proposed model has better performance under noisy conditions.

   The deep learning model proposed in this study is used to extract thermal plumes of global nuclear power plants based on Landsat images. SST increment, area, morphology and correlation between characteristics of thermal plumes were analyzed, and potential influencing factors of thermal plumes were discussed. The analysis results show that high SST increments were observed in all the target nuclear power plants, and the maximal SST increment of global target nuclear power plants is 4.80 K on average, indicating that thermal pollution overrun phenomenon has been widespread near the global nuclear power plants in the past 40 years. Shallow drainage commonly sees higher SST increments than deep drainage. Plumes larger than 3 km² frequently occur in the Great Lakes, while small plumes of 1 km² are predominant in estuarine nuclear power plants. The global thermal plume occurrence map generated in this thesis can serve as the baseline for future research, and the thermal plume extraction process adopted in this thesis can be used as a framework for operational monitoring in the future.

[1] LIN J, ZOU X Q, HUANG F M, et al. Quantitative estimation of sea surface temperature increases resulting from the thermal discharge of coastal power plants in China [J]. Marine Pollution Bulletin, 2021, 164.
[2] RAPTIS C E, VAN VLIET M T H, PFISTER S. Global thermal pollution of rivers from thermoelectric power plants [J]. Environmental Research Letters, 2016, 11(10).
[3] LIN J, ZOU X Q, HUANG F M. Effects of the thermal discharge from an offshore power plant on plankton and macrobenthic communities in subtropical China [J]. Marine Pollution Bulletin, 2018, 131: 106-114.
[4] CHEW L L, CHONG V C, WONG R C S, et al. Three decades of sea water abstraction by Kapar power plant (Malaysia): What impacts on tropical zooplankton community? [J]. Marine Pollution Bulletin, 2015, 101(1): 69-84.
[5] GRUBER N. Carbon at the coastal interface [J]. Nature, 2015, 517(7533): 148-149.
[6] MARTINEZ M L, INTRALAWAN A, VAZQUEZ G, et al. The coasts of our world: Ecological, economic and social importance [J]. Ecological Economics, 2007, 63(2-3): 254-272.
[7] STEINMAN A D, CARDINALE B J, MUNNS W R, et al. Ecosystem services in the Great Lakes [J]. Journal of Great Lakes Research, 2017, 43(3): 161-168.
[8] CHUANG Y L, YANG H H, LIN H J. Effects of a thermal discharge from a nuclear power plant on phytoplankton and periphyton in subtropical coastal waters [J]. Journal of Sea Research, 2009, 61(4): 197-205.
[9] 贺佳惠, 梁春利, 李名松. 核电站近岸温度场航空热红外遥感测量数据处理研究 [J]. 国土资源遥感, 2010: 51-53.
[10] 朱利, 赵利民, 王桥, 等. 核电站温排水分布卫星遥感监测及验证 [J]. 光谱学与光谱分析, 2014, 34: 3079-3084.
[11] RAYNER N A, BROHAN P, PARKER D E, et al. Improved analyses of changes and uncertainties in sea surface temperature measured in situ since the mid-nineteenth century: The HadSST2 dataset [J]. Journal of Climate, 2006, 19(3): 446-469.
[12] JIA H L, ZHENG S, XIE J, et al. Influence of geographic setting on thermal discharge from coastal power plants [J]. Marine Pollution Bulletin, 2016, 111(1-2): 106-114.
[13] 张贝贝, 周静, 纪平. 滨海电厂温排水数值模拟研究现状 [J]. 中国水利水电科学研究院学报, 2014, 12: 402-409.
[14] LIN J, ZOU X Q, HUANG F M. Quantitative analysis of the factors influencing the dispersion of thermal pollution caused by coastal power plants [J]. Water Research, 2021, 188.
[15] VINNA L R, WUEST A, BOUFFARD D. Physical effects of thermal pollution in lakes [J]. Water Resources Research, 2017, 53(5): 3968-3987.
[16] SUH S W. A Hybrid Near-Field/Far-Field Thermal Discharge Model for Coastal Areas [J]. Marine Pollution Bulletin, 2001, 43(7): 225-233.
[17] 许静. 基于HJ-1B红外相机监测宁德核电温排水研究 [D]： 中国地质大学（北京）, 2014.
[18] TANG D L, KESTER D R, WANG Z D, et al. AVHRR satellite remote sensing and shipboard measurements of the thermal plume from the Daya Bay, nuclear power station, China [J]. Remote Sensing of Environment, 2003, 84(4): 506-515.
[19] 栗小东. 基于遥感的滨海核电厂温排水污染监测研究 [D]： 华东师范大学, 2011.
[20] WANG X, WANG X X, FAN J C, et al. COMPARISON OF DIFFERENT SPATIAL RESOLUTION THERMAL INFRARED DATA IN MONITORING THERMAL PLUME FROM THE HONGYANHE NUCLEAR POWER PLANT [C]// Proceedings of the 36th IEEE International Geoscience and Remote Sensing Symposium (IGARSS). Beijing, PEOPLES R CHINA: IEEE, 2016: 4649-4652.
[21] HANAFIN J A, MINNETT P J. Measurements of the infrared emissivity of a wind-roughened sea surface [J]. Applied Optics, 2005, 44(3): 398-411.
[22] MINNETT P J, ALVERA-AZCARATE A, CHIN T M, et al. Half a century of satellite remote sensing of sea-surface temperature [J]. Remote Sensing of Environment, 2019, 233.
[23] BARTON I J. Transmission model and ground-truth investigation of satellite- derived sea surface temperatures [J]. Journal of Climate & Applied Meteorology, 1985, 24(6): 508-516.
[24] ANDING D, KAUTH R. Estimation of sea surface temperature from space [J]. Remote Sensing of Environment, 1970, 1(4): 217-220.
[25] DUDHIA A. NOISE CHARACTERISTICS OF THE AVHRR INFRARED CHANNELS [J]. International Journal of Remote Sensing, 1989, 10(4-5): 637-644.
[26] KARLSSON K-G, HÅKANSSON N, MITTAZ J P D, et al. Impact of AVHRR Channel 3b Noise on Climate Data Records: Filtering Method Applied to the CM SAF CLARA-A2 Data Record [J]. Remote Sensing, 2017, 9(6): 568.
[27] SIMPSON J J, YHANN S R. Reduction of noise in AVHRR channel 3 data with minimum distortion [J]. IEEE Transactions on Geoscience and Remote Sensing, 1994, 32(2): 315-328.
[28] ZAVODY A M, MUTLOW C T, LLEWELLYN-JONES D T. A radiative transfer model for sea surface temperature retrieval for the along-track scanning radiometer [J]. Journal of Geophysical Research, 1995, 100(C1): 937-952.
[29] ROTHMAN L S, GAMACHE R R, GOLDMAN A, et al. The HITRAN database: 1986 edition [J]. Applied Optics, 1987, 26(19): 4058-4097.
[30] LLEWELLYN‐JONES D T, MINNETT P J, SAUNDERS R W, et al. Satellite multichannel infrared measurements of sea surface temperature of the N.E. Atlantic Ocean using AVHRR/2 [J]. Quarterly Journal of the Royal Meteorological Society, 1984, 110(465): 613-631.
[31] MONTANARO M, LUNSFORD A, TESFAYE Z, et al. Radiometric Calibration Methodology of the Landsat 8 Thermal Infrared Sensor [J]. Remote Sensing, 2014, 6(9): 8803-8821.
[32] WANG Y, IENTILUCCI E J, RAQUENO N G, et al. Landsat 8 TIRS Calibration with External Sensors [C]// Proceedings of the Conference on Earth Observing Systems XXII. San Diego, USA: SPIE, 2017.
[33] WANG Y, IENTILUCCI E. A Practical Approach to Landsat 8 TIRS Stray Light Correction Using Multi-Sensor Measurements [J]. Remote Sensing, 2018, 10(4).
[34] 石海岗, 梁春利, 张建永, 等. 基于CBERS-04星田湾核电温排水遥感监测研究 [J]. 地理空间信息, 2019, 17: 75-79+10.
[35] 张春雷, 梁春利, 张建永, 等. 基于Landsat-8 TRIS数据的田湾核电站温排水遥感监测 [J]. 地理空间信息, 2019, 17: 70-72+87+11.
[36] 王任飞, 杨红艳, 朱利, 等. 温升包络线在核电站温排水监测中的应用 [J]. 环境监测管理与技术, 2020, 32: 49-52.
[37] 成丰, 朱利, 吴传庆, 等. 基于遥感数据的核电站温排水季节性分布监测分析——以阳江核电站为例 [J]. 环境保护, 2017, 45: 44-48.
[38] LIU M D, YIN X B, XU Q, et al. Monitoring of Fine-Scale Warm Drain-Off Water from Nuclear Power Stations in the Daya Bay Based on Landsat 8 Data [J]. Remote Sensing, 2020, 12(4).
[39] AHN Y H, SHANMUGAM P, LEE J H, et al. Application of satellite infrared data for mapping of thermal plume contamination in coastal ecosystem of Korea [J]. Marine Environmental Research, 2006, 61(2): 186-201.
[40] AOYAGI M. The impact of the Fukushima accident on nuclear power policy in Japan [J]. Nature Energy, 2021, 6(4): 326-328.
[41] KILPATRICK K A, PODESTA G, WALSH S, et al. A decade of sea surface temperature from MODIS [J]. Remote Sensing of Environment, 2015, 165: 27-41.
[42] FENG L, HOU X J, LI J S, et al. Exploring the potential of Rayleigh-corrected reflectance in coastal and inland water applications: A simple aerosol correction method and its merits [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 146: 52-64.
[43] WULDER M A, LOVELAND T R, ROY D P, et al. Current status of Landsat program, science, and applications [J]. Remote Sensing of Environment, 2019, 225: 127-147.
[44] VERMOTE E, ROGER J C, FRANCH B, et al. LaSRC (Land Surface Reflectance Code): Overview, application and validation using MODIS, VIIRS, LANDSAT and Sentinel 2 data's [C]// Proceedings of the 38th IEEE International Geoscience and Remote Sensing Symposium (IGARSS). Valencia, SPAIN: IEEE, 2018: 8173-8176.
[45] MASEK J G, VERMOTE E F, SALEOUS N E, et al. A Landsat surface reflectance dataset for North America, 1990-2000 [J]. IEEE Geoscience and Remote Sensing Letters, 2006, 3(1): 68-72.
[46] ZHAO Q, YU L, LI X C, et al. Progress and Trends in the Application of Google Earth and Google Earth Engine [J]. Remote Sensing, 2021, 13(18).
[47] ZHU Z, WOODCOCK C E. Object-based cloud and cloud shadow detection in Landsat imagery [J]. Remote Sensing of Environment, 2012, 118: 83-94.
[48] PEKEL J-F, COTTAM A, GORELICK N, et al. High-resolution mapping of global surface water and its long-term changes [J]. Nature, 2016, 540(7633): 418-422.
[49] XU H Q. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery [J]. International Journal of Remote Sensing, 2006, 27(14): 3025-3033.
[50] MALAKAR N K, HULLEY G C, HOOK S J, et al. An Operational Land Surface Temperature Product for Landsat Thermal Data: Methodology and Validation [J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(10): 5717-5735.
[51] KALNAY E, KANAMITSU M, KISTLER R, et al. The NCEP/NCAR 40-year reanalysis project [J]. Bulletin of the American Meteorological Society, 1996, 77(3): 437-471.
[52] 张彩, 朱利, 贾祥, 等. 不同空间分辨率热红外数据在近海核电厂温排水监测一致性研究 [J]. 遥感信息, 2015, 30: 71-76.
[53] 王祥, 苏岫, 张浩, 等. 不同空间分辨率遥感数据在核电站温排水监测中的应用研究 [J]. 海洋环境科学, 2020, 39: 646-651.
[54] VINCENT R F, MARSDEN R F, MINNETT P J, et al. Arctic waters and marginal ice zones: A composite Arctic sea surface temperature algorithm using satellite thermal data [J]. Journal of Geophysical Research: Oceans, 2008, 113(C4).
[55] 朱博, 王新鸿, 唐伶俐, 等. 光学遥感图像信噪比评估方法研究进展 [J]. 遥感技术与应用, 2010, 25: 303-309.
[56] REICHSTEIN M, CAMPS-VALLS G, STEVENS B, et al. Deep learning and process understanding for data-driven Earth system science [J]. Nature, 2019, 566(7743): 195-204.
[57] DUGUAY-TETZLAFF A, BENTO V A, GÖTTSCHE F M, et al. Meteosat Land Surface Temperature Climate Data Record: Achievable Accuracy and Potential Uncertainties [J]. Remote Sensing, 2015, 7(10): 13139-13156.
[58] ERMIDA S L, SOARES P, MANTAS V, et al. Google Earth Engine Open-Source Code for Land Surface Temperature Estimation from the Landsat Series [J]. Remote Sensing, 2020, 12(9): 1471.
[59] MARTINS J P A, TRIGO I F, BENTO V A, et al. A Physically Constrained Calibration Database for Land Surface Temperature Using Infrared Retrieval Algorithms [J]. Remote Sensing, 2016, 8(10): 808.
[60] RONNEBERGER O, FISCHER P, BROX T. U-Net: Convolutional Networks for Biomedical Image Segmentation [C]// Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI). Munich, GERMANY: Springer, 2015: 234-241.
[61] SHIN H C, ROTH H R, GAO M C, et al. Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning [J]. IEEE Transactions on Medical Imaging, 2016, 35(5): 1285-1298.
[62] ZHENG Z D, YANG X D, YU Z D, et al. Joint Discriminative and Generative Learning for Person Re-identification [C]// Proceedings of the 32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach, USA: IEEE, 2019: 2133-2142.
[63] SALEHI S S M, ERDOGMUS D, GHOLIPOUR A. Tversky Loss Function for Image Segmentation Using 3D Fully Convolutional Deep Networks [C]// Proceedings of the 8th International Workshop on Machine Learning in Medical Imaging (MLMI). Quebec City, CANADA: Springer, 2017: 379-387.
[64] VASWANI A, SHAZEER N, PARMAR N, et al. Attention Is All You Need [C]// Proceedings of the 31st Annual Conference on Neural Information Processing Systems (NIPS). Long Beach, USA: NeurIPS, 2017: 6000-6010.
[65] YIM J, JOO D, BAE J, et al. A Gift from Knowledge Distillation: Fast Optimization, Network Minimization and Transfer Learning [C]// Proceedings of the 30th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, USA: IEEE, 2017: 7130-7138.
[66] BELETSKY D, SAYLOR J H, SCHWAB D J. Mean circulation in the Great Lakes [J]. Journal of Great Lakes Research, 1999, 25(1): 78-93.