英文摘要: |
An efficient recognition and extraction of the estuarine turbidity maximum zone (TMZ) is important for studying terrestrial hydrological processes. Although many studies relevant to the TMZ have been conducted worldwide, the extraction methods and criteria used to describe the TMZ vary significantly both spatially and temporally. To improve the applicability of the methods adopted in previous studies and to develop a novel model to accurately extract the TMZ in multiple estuaries and different seasons from remote-sensing imageries, this study estimated the total suspended solid (TSS) and chlorophyll a (Chl a) concentrations in three estuaries. These were the Pearl River estuary (PRE), the Hanjiang River estuary (HRE), and the Moyangjiang River estuary (MRE) of Guangdong Province, China. The spatial distribution characteristics of the TSS and Chl a concentrations were analyzed. A nearly opposite association was found between the TSS and Chl a concentrations in the three estuaries, particularly in the PRE. The regions with high (low) TSS concentrations had relatively low (high) Chl a concentrations and, therefore, a turbidity maximum zone index (TMZI), defined as the ratio of the difference and sum of the logarithmic transformation of the TSS and Chl a concentrations, was firstly proposed. By calculating the TMZI values in the PRE on 20 November 2004 (low-flow season), it was found that the criterion TMZI > 0.2 could be used to identify the TMZs of the PRE effectively. The TMZ extraction results were generally consistent with the visual-interpretation results. The area-based accuracy measures showed that the quality (Q) of the extraction reached 0.8429. The same criterion was applied in the PRE on 18 October 2015 (high-flow season), and high accuracy and consistency across seasons were observed (Q D 0.8171). The western shoal of the PRE was the main distribution area of TMZs. Extracting TMZs by the newly proposed index performed well in different estuaries and on different dates (HRE on 13 August 2008 in the high-flow season and MRE on 6 December 2013 in the low-flow season). Compared to the previous fixed threshold of TSS or turbidity methods, extracting the TMZ using the TMZI had higher accuracy and better applicability (Q: 0.1046-0.4770 vs. 0.8171-0.8429). Evidently, this unified TMZI is potentially an optimized method for the global monitoring and extraction of TMZs of estuaries from different satellite remotesensing imageries. It can be used to help the understanding of the spatial and temporal variation in TMZs and estuarine processes at regional and global scales as well as improve the management and sustainable development of regional society and the natural environment. |