A New Clustering Method to Generate Training Samples for Supervised Monitoring of Long-Term Water Surface Dynamics Using Landsat Data through Google Earth Engine

Abstract

Water resources are vital to the survival of living organisms and contribute substantially to the development of various sectors. Climatic diversity, topographic conditions, and uneven distribution of surface water flows have made reservoirs one of the primary water supply resources in Iran. This study used Landsat 5, 7, and 8 data in Google Earth Engine (GEE) for supervised monitoring of surface water dynamics in the reservoir of eight Iranian dams (Karkheh, Karun-1, Karun-3, Karun-4, Dez, UpperGotvand, Zayanderud, and Golpayegan). A novel automated method was proposed for providing training samples based on an iterative K-means refinement procedure. The proposed method used the Function of the Mask (Fmask) initial water map to generate final training samples. Then, Support Vector Machines (SVM) and Random Forest (RF) models were trained with the generated samples and used for water mapping. Results demonstrated the satisfactory performance of the trained RF model with the samples of the proposed refinement procedure (with overall accuracies of 95.13%) in comparison to the trained RF with direct samples of Fmask initial water map (with overall accuracies of 78.91%), indicating the proposed approach’s success in producing training samples. The performance of three feature sets was also evaluated. Tasseled-Cap (TC) achieved higher overall accuracies than Spectral Indices (SI) and Principal Component Transformation of Image Bands (PCA). However, simultaneous use of all features (TC, SI, and PCA) boosted classification overall accuracy. Moreover, long-term surface water changes showed a downward trend in five study sites. Comparing the latest year’s water surface area (2021) with the maximum long-term extent showed that all study sites experienced a significant reduction (16–62%). Analysis of climate factors’ impacts also revealed that precipitation (0.51 ≤ R2 ≤ 0.79) was more correlated than the temperature (0.22 ≤ R2 ≤ 0.39) with water surface area changes.

Spatio-temporal water quality variability and regionalization of tropical reservoirs based on Landsat imagery

“…[18][19][20][21][22][23] Remote sensing has been applied in reservoirs worldwide to assess spatial and temporal variability of bio-optical parameters and temperature of the water to determine water quality. 8,22,[24][25][26][27][28][29][30] In Brazil, there has been a growing effort to study continental waters, always combining in situ measurements with the processing of satellite images. 4,[31][32][33][34][35][36][37][38][39] However, most of those explore very narrow temporal and/or spatial ranges.…”

Opportunities and Challenges of Spaceborne Sensors in Delineating Land Surface Temperature Trends: A Review

“…Techniques, such as data fusion and cloud masking as well as machine learning algorithms can also be performed through GEE, opening an opportunity for users to explore the available datasets and investigate spatiotemporal dynamics of LST, vegetation, land cover, etc. [122]. Thanks to its computing infrastructure, GEE can efficiently and rapidly handle big data and extensive computations [123].…”

Flood Inundation Mapping with Supervised Classifiers: 2021 Gediz Plain Flood

“…Kernel function type, gamma value, and cost value are the parameters of this algorithm. The linear kernel is used to project the input space to higher spaces when the data volume is high in order to separate the data (Taheri Dehkordi et al 2022;Amani et al 2019).…”

“…For each node, a random feature is selected from the feature set to split the tree properly. Generally, two parameters are used in RF: the number of trees and the number of nodes(Taheri Dehkordi et al 2022;Amani et al 2019).…”