An automatic classification algorithm for submerged aquatic vegetation in shallow lakes using Landsat imagery

• Published in: Remote sensing of environment Vol. 260; p. 112459
• Main Authors: Dai, Yanhui, Feng, Lian, Hou, Xuejiao, Tang, Jing
• Format: Journal Article
• Language: English
• Published: New York Elsevier Inc 01.07.2021; Elsevier BV
• Subjects: Algae; Algorithms; Aquatic plants; Aquatic vegetation; Automatic classification; Biologi; Biological Sciences; Classification; data collection; detection limit; Distribution patterns; Dynamic threshold; Ecology (including Biodiversity Conservation); Ekologi; environment; FAI; humans; Hydrology; Lakes; Landsat; Landsat satellites; Natural Sciences; Naturvetenskap; Numerical simulations; Nutrients; Reflectance; Remote sensing; Satellite imagery; SAV; Short wave radiation; Spatial distribution; submerged aquatic plants; surveys; SWIR; temperature; Vegetation; Water circulation; Water column; Dynamic threshold; Classification; Landsat; SAV; SWIR; Aquatic vegetation; Remote sensing; FAI
• ISSN: 0034-4257; 1879-0704
• DOI: 10.1016/j.rse.2021.112459

Submerged aquatic vegetation (SAV) is one of the main producers in inland lakes. Tracking the temporal and spatial changes in SAV is crucial for the identification of state changes in lacustrine ecosystems, such as changes in light, nutrients, and temperature. However, the available SAV classification algorithms based on remote sensing are highly dependent on field survey data and/or human interventions, prohibiting the extraction of large-scale and/or long-term patterns. Here, we developed an automatic SAV classification algorithm using Landsat imagery, where the thresholds of two key parameters (the floating algae index (FAI) and reflectance in the shortwave-infrared (SWIR) band) are automatically determined. The algorithm was applied to eight Landsat images of four Yangtze Plain lakes and obtained a mean producer accuracy of 82.9% when gauged against field-surveyed datasets. The algorithm was further employed to obtain long-term SAV areal data from Changdang Lake on the Yangtze Plain from 1984 to 2018, and the result was highly consistent with lake transparency data. Numerical simulations indicated that our developed algorithm is insensitive to the Chl-a concentration of the water column. Yet, it has a detection limit of ~0.35 m below the water surface, and such a limit changes with different fractions of vegetation coverage within a pixel. The automatic classification algorithm proposed in this study has the potential to obtain the temporal and spatial distribution patterns of SAV in other shallow lakes where SAV grows in lakes sharing similar hydrological characteristics as the lakes in the Yangtze Plain. •An automatic classification algorithm is developed for detecting submerged aquatic vegetation in shallow lakes.•FAI and reflectance in SWIR band are two key variables in the algorithm.•Submerged aquatic vegetation at a depth of 0–0.35 m can be detected.•The algorithm is insensitive to the Chl-a concentration of water.•Application of the algorithm to eight Landsat images indicates the potential robustness.