Inland waters, such as lakes, are crucial in sustaining ecosystems, supporting human livelihoods, and preserving biodiversity. The scarcity of drinking water, soaring food crises, and damaged ecosystems in many parts of the world manifest the significance of lakes in water resources management. Climate variations and extreme weather events have jeopardized water availability nowadays. Hence, it is imperative for government authorities to vigilantly monitor spatial and temporal changes in lakes (Muzaffar et al., 2019). Monitoring water levels and volumetric variations in lakes or inland waters is the basic principle of integrated water resources management, necessitating the effective management of available freshwater resources. Nonetheless, ensuring the sustainable utilization of water resources while keeping budgets in check is a real challenge.
Countries like Pakistan lack sufficient monitoring approaches to manage their water bodies effectively. The devastating 2022 flood in the country, triggering havoc conditions in Manchar Lake, has unveiled the shambolic and inadequate management structure of the Lake. In such conditions, cost-effective techniques are required to ameliorate the monitoring policies in water resource management. In this context, remote sensing offers an economical solution and contributes to the broader goal of responsible and efficient water resource management (Zaidi et al., 2021).
Satellite radar altimetry was initially utilized to monitor ocean water levels. However, concerning time and necessity, its scope extended to continental surface waters, enabling satellite-based inland water observations. Different studies have utilized myriad altimetry technologies to compute lake and river water levels to manage water bodies efficiently worldwide (Chen et al., 2022a), (Ali, Rehman, and Zafar, 2022), (Verma et al., 2021). Satellite radar altimetry can reasonably accurately predict inland water levels without border restrictions owing to the global coverage of satellite orbit. The working principle of satellite radar altimetry (SRA) is that the radar pulses are emitted from a radar altimeter, and the time they take to to travel to the surface and reach back to the satellite is used to calculate altitude or range between the satellite and the Earth surface (Kittel et al., 2021). A laser altimeter works on the same principle, emitting laser pulses.
Jason-3, a radar altimetry technology, was used to monitor the water level time series of different lakes, and they were found to be in good agreement with in situ gauges (Schwatke et al., 2015). DAHITI (Database for Hydrological Time Series of Inland Waters) uses multi-mission satellite altimetry and optical remote sensing imagery for water level (W.L.) time series of lakes, rivers, reservoirs, and wetlands around the world (Schwatke et al., 2015). ICESat-2 (Ice, Cloud, and Land Elevation Satellite), with a laser detection instrument and a smaller footprint, is suitable for computing lake water levels in areas less than one square kilometer. ATLAS/ICESat-2 L3A Along Track Inland Surface Water (ATL13) dataset showed greater accuracies for inland water measurements (Xu et al., 2021). Zhang et al. (2022) validated ICESat-2 derived W.L.s of Lake Chaohu, Hongze, with in situ measurements. They found good correlation coefficients ranging from 0.957 to 0.995. Similarly, a correlation coefficient of 0.9 suggested a statistically significant association between in situ and sentinel 3A data at the downstream reaches of the Indus River (Zaidi et al., 2021).
Along with water levels, surface area calculation is also integral for the computation of volumetric variations in lakes. Different studies have utilized the interpretation of optical and SAR images to calculate water surface area (Kavats, Khramov, and Sergieieva, 2022). SAR data of Sentinel-1 were utilized to monitor Poyang Lake. They used the U-NET convolutional neural network technique for classification purposes (Shen et al., 2022). A novel approach of polarimetric SAR speckle filtering investigated the probability of choosing the optimal parameter for water masking through SAR technology. This approach was also assessed with the Otsu threshold method. The method was found to be as good as the Otsu threshold with a reference water mask (Kavats, Khramov, and Sergieieva, 2022). Another study employed an optical remote sensing approach to calculate surface area variations of the Manchar Lake (Ismail et al., 2022). They experimented with Landsat 8 various spectral bands/ indices Shortwave Infrared 1 (SW1R1), Normalized Difference Built-Up Index (NDBI), and Normalized Difference Water Index (NDWI) to propose a new index (WIBI) for mapping water extents. The WIBI was applied to seven-year data, which precisely extracted the surface area of the Manchar Lake.
Surface areas and water level data can further help estimate the volumetric variation of a particular water body. Liu et al. (2022) utilized ICESat-2 to derive water level and surface area variations from optical images using Google Earth Engine (GEE) to calculate volumetric changes in the Yellow River. In another study, multisource satellite techniques were utilized to monitor Lake Victoria in Africa (Sichangi and Makokha, 2017). They used data from Jason, MODIS, and GRACE satellites to measure water levels and surface areas. The volumetric variations were calculated through a mathematical formula. Nevertheless, despite the multiple advantages of optical remote sensing, some grave issues impede monitoring water bodies during cloud cover. In these cases, radar remote sensing proves valuable. Owing to the limitation of optical images with cloud cover, SAR images were preferred in our study to compute surface area variations.
Nevertheless, certain limitations in satellite data processing can not be disregarded, notably associated with their low temporal and spatial coverages. Altimetry satellites generally have limited spatial extent (only waterbodies with virtual stations can be observed) with low temporal resolutions (ICESat-2 = 90 and Sentinel 3 = 27 days), creating shorter time series. ICESat-2 often has bad data due to land contamination with its small footprints (Chen et al., 2022b). The speckle noise in SAR images sometimes distorts the data accuracy. Despite these limitations, remote sensing has a promising future in water observations with fast-growing technical and scientific advancements.
Our study's main objective was to integrate satellite altimetry and SAR images to compute the volumetric variation in Manchar Lake, Pakistan. In this study, different altimetry satellites such as Jason3, ICESat-2, and Sentinel 3 were utilized to study the temporal variations in the water level of Manchar Lake. SAR technology was also applied to compute surface area variation. Volumetric variations were estimated using water level and surface area data.