Full
Outline
Conventional remote sensing platforms like Landsat and Sentinel face inherent limitations in achieving continuous dynamic monitoring of cyanobacterial blooms due to their inadequate temporal resolution. Similarly, existing FY satellite-based monitoring and prediction approaches are constrained by unidimensional analyses and oversimplified models dependent on single-variable inputs. To overcome these challenges, this study developed a comprehensive multidimensional monitoring and prediction system utilizing high-temporal-resolution FY3D/MERSI data for Taihu Lake cyanobacterial blooms. Our methodology integrated spectral indices with an optimized Otsu algorithm for accurate bloom detection, coupled with innovative dual-scale (pixel and sub-pixel) quantification techniques. Through this framework, we conducted systematic multidimensional analysis of Taihu Lake's cyanobacterial blooms (2019~2023), examining critical parameters including bloom intensity, frequency patterns, and outbreak severity dynamics. Concurrently, integrating meteorological and water quality factors, we constructed a BP neural network model to predict bloom areas. Validation analyses demonstrated the strong performance of FY3D/MERSI data in cyanobacterial bloom extraction, showing excellent agreement with in-situ measurements (R2 = 0.98, RMSE = 21.6km2). Cyanobacterial bloom areas in Lake Tai exhibited seasonal double-peak patterns, with peaks occurring in May and October. Bloom severity predominantly ranged from mild to moderate, while severe blooms remained at low levels throughout the year. Outbreak frequency was significantly higher in the western, southern coastal, and central lake areas than in the eastern region, and higher frequencies occurred during summer and autumn. Inter annual outbreak intensity showed a gradual decreasing trend, with the proportion of "no cyanobacterial blooms" rising to 89% in 2023. Correlation analysis indicated that factors such as temperature and relative humidity promote bloom outbreaks, while turbidity and fluorescent dissolved organic matter exerted inhibitory effects. The BP neural network model demonstrated robust predictive capability, achieving a root mean square error (RMSE) of 27.07km2 and a Pearson correlation coefficient (r) of 0.94 on the validation set. Notably, 83% of predictions fell within ±25km2 of observed values. These metrics collectively confirm the model's effectiveness in capturing the dynamic spatiotemporal trends of cyanobacterial bloom areas.
106
16
| 科 Family | 属数 Number of genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) | 属 Genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) |
|---|---|---|---|---|---|---|
| 鹅膏菌科Amanitaceae | 2 | 11 | 5.26 | 鹅膏菌属 Amanita | 10 | 4.78 |
| 小菇科 Mycenaceae | 2 | 12 | 5.74 | 丝盖伞属 Inocybe | 5 | 2.39 |
| 多孔菌科 Polyporaceae | 8 | 14 | 6.70 | 蜡蘑属 Laccaria | 5 | 2.39 |
| 红菇科 Russulaceae | 3 | 23 | 11.00 | 小皮伞属 Marasmius | 6 | 2.87 |
| 小菇属 Mycena | 11 | 5.26 | ||||
| 光柄菇属 Pluteus | 5 | 2.39 | ||||
| 红菇属 Russula | 17 | 8.13 | ||||
| 栓菌属 Trametes | 5 | 2.39 |
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62662699(Operation and Technical Support)
qkjq@cast.org.cn
Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House
