Integration of Remote Sensing and Mexican Water Quality Monitoring System Using an Extreme Learning Machine.

Remote Sensing, as a driver for water management decisions, needs further integration with monitoring water quality programs, especially in developing countries. Moreover, usage of remote sensing approaches has not been broadly applied in monitoring routines. Therefore, it is necessary to assess the efficacy of available sensors to complement the often limited field measurements from such programs and build models that support monitoring tasks. Here, we integrate field measurements (2013-2019) from the Mexican national water quality monitoring system (RNMCA) with data from Landsat-8 OLI, Sentinel-3 OLCI, and Sentinel-2 MSI to train an extreme learning machine (ELM), a support vector regression (SVR) and a linear regression (LR) for estimating Chlorophyll-a (Chl-a), Turbidity, Total Suspended Matter (TSM) and Secchi Disk Depth (SDD). Additionally, OLCI Level-2 Products for Chl-a and TSM are compared against the RNMCA data. We observed that OLCI Level-2 Products are poorly correlated with the RNMCA data and it is not feasible to rely only on them to support monitoring operations. However, OLCI atmospherically corrected data is useful to develop accurate models using an ELM, particularly for Turbidity (R2 = 0.7). We conclude that remote sensing is useful to support monitoring systems tasks, and its progressive integration will improve the quality of water quality monitoring programs.

Mapping regional surface water volume variation in reservoirs in northeastern Brazil during 2009-2017 using high-resolution satellite images.

The multiple-year drought that started in 2011 and reached climax in 2015 was the most severe and prolonged one in the semiarid northeastern (NE) Brazil in recent decades. This study aimed to investigate the reservoir surface water volume (SWV) variation in NE Brazil from 2009 to 2017 in four representative regions covering a total area of approximately 10,000 km2 there and encompassing 2,140 reservoirs (areas range from 0.003 to 21 km2). High-resolution (10 m) digital elevation models (DEMs) were generated from the TanDEM-X data acquired during October-December 2015 to represent the reservoirs' bathymetric maps. The water extents in the reservoirs were delineated from high-resolution (6.5 m) RapidEye images acquired during 2009-2017. The combination of the aforementioned two variables yielded reservoir SWV with an accuracy of 0.64 × 106-1.06 × 106 m3, corresponding to 3.1%-5.6% of the maximum SWV in the reservoirs. The results showed that: 1) 81%-99% of the reservoirs in the four regions were from the groups with maximum water extent <50 ha and contributed 2%-59% of the regional reservoir SWV. In contrast, 0.6%-20% of the reservoirs were from the group of >50 ha and contributed 40%-98% to the regional SWV; 2) From 2009 to 2017, reservoir SWV in the four regions decreased at the rates of 2.3 × 106-17.8 × 106 m3/year; and 3) The SWV in the reservoirs responded differently to the regional terrestrial water budget, i.e. the differences between precipitation and evapotranspiration (P-ET). This study filled the data gap of bathymetric maps for the 2140 reservoirs, regardless of their sizes and macrophyte coverage. The SWV variations derived in those reservoirs over a period covering the recent drought can support better preparedness for drought in NE Brazil and better understanding of the regional hydrology in semi-arid regions.