Surface Water Extent Research Articles

Satellite Detection of Surface Water Extent: A Review of Methodology

Water is an imperative part of the Earth and an essential resource in human life and production. Under the effects of climate change and human activities, the spatial and temporal distribution of water bodies has been changing, and the shortage of water resources is becoming increasingly serious worldwide. Therefore, the monitoring of water bodies is indispensable. Remote sensing has the advantages of real time, wide coverage, and rich information and has become a brand-new technical means to quickly obtain water information. This study summarizes the current common methods of water extraction based on optical and radar images, including the threshold method, support vector machine, decision tree, object-oriented extraction, and deep learning, as well as the advantages and disadvantages of each method. These methods were applied to the Huai River Basin in China and Nam Co on the Qinghai-Tibet Plateau. The extraction results show that all the aforementioned approaches can obtain reliable results. Among them, the threshold segmentation method based on normalized difference water index is more robust than others. In the water extraction process, there are still many problems that restrict the accuracy of the results. In the future, researchers will continue to search for more automatic, extensive, and high-precision water extraction methods.

Merging Landsat and airborne LiDAR observations for continuous monitoring of floodplain water extent, depth and volume

The Darling River system in Australia is under pressure from water extraction and climate change. Management interventions such as environmental flow releases require understanding of water storage dynamics and the connectivity of floodplains and wetlands. Such knowledge can be gleaned from the long observational record of the Landsat series of satellite sensors and high (<5 m) resolution digital elevation models derived from airborne light detection and ranging (LiDAR). Here, for the first time, we develop and demonstrate an approach to reconstruct 16-day floodplain water dynamics, including extent, depth, and volume for a long Landsat time series (1987 to present). Time series mapping of surface water extent at 5-m resolution was achieved by topographic downscaling of Landsat-derived surface water data. We propose a simple and effective algorithm to restore missing data in the images caused by, e.g., cloud and shadows, swath edges and the Landsat 7 Scan Line Corrector (SLC) failure, thereby increasing the number of useable images five-fold. The 5-m surface water extent maps clearly delineate the narrow river channel and the boundary of floodplain wetlands. They can capture the development, peak and retreat of flood events. By combining Landsat and airborne LiDAR observations, we produced time series of surface water depth mapping at 5-m resolution, accounting for the degree of hydraulic surface water connectivity. Based on these maps, we derived 16-day floodplain volume dynamics for 1987 to present. The correlation coefficient between upstream river flow records and floodplain volume time series was 0.88, indicating that the estimates were robust. The algorithms developed can be used for ongoing very high-resolution mapping to assist in managing human water use and environmental health in the Murray-Darling Basin.

Effects of Climate and Anthropogenic Drivers on Surface Water Area in the Southeastern United States

AbstractSurface water is the most readily accessible water resource and provides an array of ecosystem services, but its availability and access are stressed by changes in climate, land cover, and population size. Understanding drivers of surface water dynamics in space and time is key to better managing our water resources. However, few studies estimating changes in surface water account for climate and anthropogenic drivers both independently and together. We used 19 years (2000–2018) of the newly developed Dynamic Surface Water Extent Landsat Science Product in concert with time series of precipitation, temperature, land cover, and population size to statistically model maximum seasonal percent surface water area as a function of climate and anthropogenic drivers in the southeastern United States. We fitted three statistical models (linear mixed effects, random forests, and mixed effects random forests) and three groups of explanatory variables (climate, anthropogenic, and their combination) to assess the accuracy of estimating percent surface water area at the watershed scale with different drivers. We found that anthropogenic drivers accounted for approximately 37% more of the variance in the percent surface water area than the climate variables. The combination of variables in the mixed effects random forest model produced the smallest mean percent errors (mean −0.17%) and the highest explained variance (R2 0.99). Our results indicate that anthropogenic drivers have greater influence when estimating percent surface water area than climate drivers, suggesting that water management practices and land‐use policies can be highly effective tools in controlling surface water variations in the Southeast.

Development of a Multi-Index Method Based on Landsat Reflectance Data to Map Open Water in a Complex Environment

Mapping surface water extent is important for managing water supply for agriculture and the environment. Remote sensing technologies, such as Landsat, provide an affordable means of capturing surface water extent with reasonable spatial and temporal coverage suited to this purpose. Many methods are available for mapping surface water including the modified Normalised Difference Water Index (mNDWI), Fisher’s water index (FWI), Water Observations from Space (WOfS), and the Tasseled Cap Wetness index (TCW). While these methods can discriminate water, they have their strengths and weaknesses, and perform at their best in different environments, and with different threshold values. This study combines the strengths of these indices by developing rules that applies an index to the environment where they perform best. It compares these indices across the Murray-Darling Basin (MDB) in southeast Australia, to assess performance and compile a heuristic rule set for accurate application across the MDB. The results found that all single indices perform well with the Kappa statistic showing strong agreement, ranging from 0.78 for WOfS to 0.84 for TCW (with threshold −0.035), with improvement in the overall output when the index best suited for an environment was selected. mNDWI (using a threshold of −0.3) works well within river channels, while TCW (with threshold −0.035) is best for wetlands and flooded vegetation. FWI and mNDWI (with threshold 0.63 and 0, respectively) work well for remaining areas. Selecting the appropriate index for an environment increases the overall Kappa statistic to 0.88 with a water pixel accuracy of 90.5% and a dry pixel accuracy of 94.8%. An independent assessment illustrates the benefit of using the multi-index approach, making it suitable for regional-scale multi-temporal analysis.

Global seasonal dynamics of inland open water and ice

Seasonal changes of temperature and precipitation cause inland open surface water and ice cover extents to vary dramatically through the year from local to global scales. These dynamics of land, water, and ice have a significant impact on climate and often are critical to natural ecosystem functioning. However, global seasonal dynamics of both water and ice extent have not been well quantified. Here, we present the quantification of monthly surface water and ice areas for 2019 with associated uncertainties. Time-series reference data were created for a probability sample of 10 m grid cells by interpreting the entire 2019 time-series of 10 m Sentinel-2 data and a subset of 3 m PlanetScope data in selected places with a mix of land and water. From the probability sample reference data, we estimate that 4.86 ± 0.16 million km2 had inland water presence at some point during the year. Globally, only 23% of the total area with water was permanent water that remained open year-round (1.13 ± 0.19 million km2). Permanent water with seasonal ice cover extended 1.97 ± 0.21 million km2, comprising 41% of the total area with water. Seasonal water-land transitions (both with and without ice/snow cover) covered the remaining 36% of the total area with water (1.76 ± 0.19 million km2). February had the maximum extent of ice over areas of inland permanent and seasonal water, totaling 2.49 ± 0.25 million km2, and January – March had a larger global extent of ice cover than of open water. To investigate the spatiotemporal distribution of ice cover and the suitability of Landsat, prototype maps of surface water ice cover phenology were created by integrating the ice/snow and no data labels from the quality assurance layer of the GLAD ARD of Potapov et al. (2020) with the monthly surface water layers of Pickens et al. (2020), both of which are Landsat-based. While limited by data availability, these maps reveal the high spatiotemporal variability of ice phenology. The near-daily observations near the poles and the 10 m resolution bands of Sentinel-2 provide unprecedented potential to examine surface water and ice dynamics for 2016 forward and to investigate the drivers and impacts of this variability.

PyGEE-SWToolbox: A Python Jupyter Notebook Toolbox for Interactive Surface Water Mapping and Analysis Using Google Earth Engine

Continuous monitoring of surface water resources is often challenging due to the lack of monitoring systems in remote areas and the high spatial distribution of water bodies. The Google Earth Engine (GEE) platform, which houses a large set of remote sensing datasets and geospatial processing power, has been applied in various aspects of surface water resources monitoring to solve some of the challenges. PyGEE-SWToolbox is a freely available Google Earth Engine-enabled open-source toolbox developed with Python to be run in Jupyter Notebooks that provides an easy-to-use graphical user interface (GUI) that enables the user to obtain time series of Landsat, Sentinel-1, and Sentinel-2 satellite imagery, pre-process them, and extract surface water using water indices, such as the Normalized Difference Water Index (NDWI), Modified NDWI (MNDWI), Automated Water Extraction Index (AWEI), and Dynamic Surface Water Extent (DSWE). The validation of the toolbox is carried out at four reservoir and lake locations: Elephant Butte Lake, Hubbard Creek Reservoir, Clearwater Lake, and Neversink Reservoir in the United States. A time series of the water surface area generated from PyGEE-SWToolbox compared to the observed surface areas yielded good results, with R2 ranging between 0.63 and 0.99 for Elephant Butte Lake, Hubbard Creek Reservoir, and Clearwater Lake except the Neversink Reservoir with a maximum R2 of 0.52. The purpose of PyGEE-SWToolbox is to provide water resource managers, engineers, researchers, and students a user-friendly environment to utilize the GEE platform for water resource monitoring and generation of datasets. The toolbox is accompanied by a step-by-step user manual and Readme documentation for installation and usage.

DSWEmod — The Production of High‐Frequency Surface Water Map Composites from Daily MODIS Images

AbstractOptical satellite imagery is commonly used for monitoring surface water dynamics, but clouds and cloud shadows present challenges in assembling complete water time series. To test whether the daily revisit rate of Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery can reduce cloud obstruction and improve high‐frequency surface water mapping, we compared map results derived from Landsat (30‐m) and MODIS (250‐m) data across the state of California for 2003–2019. We adapted the Dynamic Surface Water Extent (DSWE) model in Google Earth Engine to generate surface water map composites from MODIS imagery every 5, 10, 15, and 30 days, and compared products to monthly Landsat‐based DSWE maps. Results for DSWEmod (DSWE MODIS) in California suggest that more than 5% data loss (cloud obstruction, etc.) was present in only 2% of the 15‐day time series, as compared to 32% of the monthly Landsat DSWE time series. The five‐day DSWEmod composites averaged 8.4% obscuration in the winter months. Area estimates derived from cloud‐filtered MODIS and Landsat monthly products have the highest linear correlations compared to streamgage discharge records, suggesting that monthly scale analyses best explain the relationship between surface water area and general streamflow dynamics. Shorter‐interval DSWEmod products have lower correlations but utility for understanding the timing of surface water peaks and past flood events.

Lagrangian Coherent Structures and hypoxia in the Baltic Sea

The spatial and temporal variations of Lagrangian Coherent Structures (LCSs) in the Baltic Sea are extracted from finite-time Lyapunov exponent (FTLE) fields. A validated 3-D hydrodynamic model of the Baltic Sea coupled with a water quality model is applied for the years 2000–2009. The novelty of the work is on the variation of LCSs with the sea depth, the state of hypoxia and the possible relationship with the blooming patterns in the Baltic. The study reveals a variety of LCSs with a typical core diameter of 10–40 km and a duration of 2–7 days that are formed offshore in all the basins of the Baltic Sea. They occur throughout the year even during winter times when the sea at the northern basins is partially covered with ice. The LCSs are more abundant in the southern basins where extensive algae blooms occur. The dominant structures are large vortex dipoles and anti-rotating vortex pairs that are not limited to the surface water layer but spread to a depth of 143.5 m. Likely mechanisms for the formation and the spread of LCSs are Kelvin-Helmholtz type instabilities, Proudman-Taylor column and Ekman Spiral. In the vicinity of the shorelines, the LCSs are smaller in diameter scaling with the mean Rossby radius to around 5 km. In the southern basins of the Baltic Sea; the dissolved oxygen (DO) content is permanently below 2 mg/l at depths below 80 m. DO contents vary seasonally with high values during winter and early spring times as opposed to lower values during summer and autumn periods. During late summers, the decline in DO content appears related to the extensive algae blooming consuming oxygen through the decomposition process. The LCSs map the patterns of Algae blooms detected by satellites. The duration of Algae blooming agrees with the persisting time of LCSs as well as the spatial surface water extents. The major 2003 inflow (MBI) increased the seabed DO content only in the lower part of the Arkona Basin that lasted just for about 3 weeks. The inputs of DO from the rivers and the 2003 MBI were not sufficient to counteract the seabed hypoxia in the Baltic Proper for the years 2000–2009.

Integration of vegetation classification with land cover mapping: lessons from regional mapping efforts in the Americas

Aims: Natural resource management and biodiversity conservation rely on inventories of vegetation that span multiple management or political jurisdictions. However, while remote sensing data and analytical tools have enabled production of maps at increasing spatial resolution and reliability, there are limited examples where national or continental-scaled maps are produced to represent vegetation at high thematic detail. We illustrate two examples that have bridged the gap between traditional land cover mapping and modern vegetation classification. Study area: Our two case studies include national (USA) and continental (North and South America) vegetation and land cover mapping. These studies span conditions from subpolar to tropical latitudes of the Americas. Methods: Both case studies used a supervised modeling approach with the International Vegetation Classification (IVC) to produce maps that provide for greater thematic detail. Georeferenced locations for these vegetation types are used by machine learning algorithms to train a predictive model and generate a distribution map. Results: The USALANDFIRE (Landscape Fire and Resource Management Planning Tools Project) case study illustrates how a history of vegetation-based classification and availability of key inputs can come together to generate standard map products covering more than 9.8 million km2 that are unsurpassed anywhere in the world in terms of spatial and thematic resolution. That being said, it also remains clear that mapping at the thematic resolution of the IVC Group and finer resolution require very large and spatially balanced inputs of georeferenced samples. Even with extensive prior data collection efforts, these remain a key limitation. The NatureServe effort for the Americas - encompassing 22% of the global land surface - demonstrates methods and outputs suitable for worldwide application at continental scales. Conclusions: Continued collection of input data used in the case studies could enable mapping at these spatial and thematic resolutions around the globe. Abbreviations: CART = Classification and Regression Tree; CONUS = Conterminous United States; DSWE = Dynamic Surface Water Extent; EPA = United States Environmental Protection Agency; FGDC = Federal Geographic Data Committee; IVC = International Vegetation Classification; LANDFIRE = Landscape Fire and Resource Management Planning Tools Project; LFRDB = LANDFIRE Reference Database; LiDAR = Light Detection and Ranging; NDVI = Normalized Difference Vegetation Index; NLCD = National Land Cover Database; USNVC = United States National Vegetation Classification; USA = United States of America; WWF = World Wildlife Fund or Worldwide Fund for Nature.

Assessing Surface Water Losses and Gains under Rapid Urbanization for SDG 6.6.1 Using Long-Term Landsat Imagery in the Guangdong-Hong Kong-Macao Greater Bay Area, China

As one of the most open and dynamic regions in China, the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) has been urbanizing rapidly in recent decades. The surface water in the GBA also has been suffering from urbanization and intensified human activities. The study aimed to characterize the spatiotemporal patterns and assess the losses and gains of surface water caused by urbanization in the GBA via long time-series remote sensing data, which could support the progress towards sustainable development goals (SDGs) set by the United Nations, especially for measuring SDG 6.6.1 indicator. Firstly, utilizing 4750 continuous Landsat TM/ETM+/OLI images during 1986–2020 and the Google Earth Engine cloud platform, the multiple index water detection rule (MIWDR) was performed to extract surface water extent in the GBA. Secondly, we achieved surface water dynamic type classification based on annual water inundation frequency time-series in the GBA. Finally, the spatial distribution and temporal variation of urbanization-induced water losses and gains were analyzed through a land cover transfer matrix. Results showed that (1) the average minimal and maximal surface water extents of the GBA during 1986–2020 were 2017.62 km2 and 6129.55 km2, respectively. The maximal surface water extent fell rapidly from 7897.96 km2 in 2001 to 5087.46 km2 in 2020, with a loss speed of 155.41 km2 per year (R2 = 0.86). (2) The surface water areas of permanent and dynamic types were 1529.02 km2 and 2064.99 km2 during 2000–2020, accounting for 42.54% and 57.46% of all water-related areas, respectively. (3) The surface water extent occupied by impervious land surfaces showed a significant linear downward trend (R2 = 0.98, slope = 36.41 km2 per year), while the surface water restored from impervious land surfaces denoted a slight growing trend (R2 = 0.86, slope = 0.99 km2 per year). Our study monitored the long-term changes in the surface water of the GBA, which can provide valuable information for the sustainable development of the GBA urban agglomeration. In addition, the proposed framework can easily be implemented in other similar regions worldwide.

Influence of Urban Areas on Surface Water Loss in the Contiguous United States

AbstractUrbanization is one of the main drivers of surface water loss, which implies a transition from water to land. However, it is still unclear how urban areas affect the spatial pattern of surface water loss. Here, we use remotely sensed data to analyze and model the decrease of surface water extent and, in particular, the frequency of surface water loss as a function of distance from urban areas across the contiguous United States (CONUS). We employ an exponential distance‐decay model that confirms the presence of a higher frequency of surface water loss in the proximity of human settlements and provides innovative insights on surface water loss patterns at different spatial scales (i.e., river basins, water resource regions, and the CONUS). These spatial patterns are found to be influenced by climatic conditions, with more widely distributed losses in arid regions with respect to temperate and continental climates. Our results provide a new and deeper understanding of the spatial influence of urban areas on surface water loss, which could be effectively integrated in the definition of sustainable strategies for urbanization, water management, and surface water restoration, focused on both human and environmental water needs.

Impact of water resource development on connectivity and primary productivity across a tropical river floodplain

Abstract Floodplain wetlands provide an important subsidy for riverine food webs as sites of high algal production. However, this subsidy depends on the degree of landscape connectivity during flood pulses, which provides the opportunity for movement of higher order consumers between rivers and floodplains to access these productive habitats. Changes in floodplain inundation extent and duration, due to variable wet season flows or water resource development (WRD), can impact landscape connectivity and ultimately the magnitude of the food web subsidy. We explored landscape connectivity using graph theory and derived four new metrics to measure how different flow scenarios can affect connectivity and algal production. We considered a historic scenario with the present level of water resource development in our study area, the Mitchell River, and a WRD scenario with the inclusion of three new dams in the catchment. We generated 240 unique daily spatial graphs, using surface water inundation maps across 40‐day flood events to compare a dry year (2006), an average flow year (2001) and a wet year (2009) with and without a WRD scenario. Drier years and WRD scenario resulted in floodplain fragmentation, potentially constraining the movement of higher order consumers. Changes in connectivity due to WRD resulted in predicted reductions of up to 26% of algal production on the floodplain that was otherwise connected to the main river channels. Synthesis and applications. The approach developed in this study provides new metrics to identify how changes in floodplain surface water extent due to water resource development and climatic variation may impact ecosystem function such as connectivity that facilitates access of higher order consumers to primary production in floodplain wetlands. With a direct link to river flow alteration, these metrics can inform catchment planning and management to ensure that the conservation of floodplain ecosystem functions is adequately considered in water resource management decisions.

Mapping Surface Water Extents Using High-Rate Coherent Spaceborne GNSS-R Measurements

Coherent GNSS reflections over land predominantly occur over surface water bodies. This study presents a method to jointly use carrier phases and signal strengths of reflected signals to identify coherent reflections and applies it to the 50-Hz GNSS-R measurements from Spire Global Cubesats and CYGNSS microsatellites to map inland water bodies. A coherence detector was first developed using the circular statistics of carrier phase noises, identifying the input samples as coherent, semi-coherent, or incoherent. For any given track of data, we used this coherence detector to iteratively assess the coherency levels of the samples by a moving time window, then derived the coherency levels with the highest confidence. The circular statistics-based semi-coherent reflections with signal strengths above the prescribed threshold were regarded as coherent. The specular reflection points of the coherent reflections represent the locations of surface water. This method was applied to the Spire data to obtain the surface water extents for 1951 lakes and the CYGNSS data for 113 lakes. Compared to Global Surface Water Explorer observations, around 90% of the disagreements of the Spire data-based surface water boundaries are less than 0.73 km with a mean of 0.28 km and a standard deviation of 0.24 km. As for CYGNSS, ~90% of the disagreements are less than 0.43 km with a mean value of 0.18 km and a standard deviation of 0.16 km. The possible error sources are mainly fractional surface water, nearly flat and saturated ground surface, background land cover, and GNSS-R geometry.

Comparison of SAR and CYGNSS Surface Water Extent Metrics

Many sensors are suitable for accurate delineation of open water extent, but in vegetated environments, the vegetation canopy can obscure the presence of standing water from detection. Detecting inundation extent in these vegetated environments is especially critical for identifying flooding extent where surface water may exceed flood boundaries and extend into forests surrounding nearby lakes and streams. Regular and timely observations of water surfaces by optical sensors can be impeded by both cloud cover and by vegetation. Here, two microwave techniques for identifying inundation extent will be investigated and compared: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</i> -band global navigation satellite systems reflectometry (GNSS-R) and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</i> - and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</i> -band synthetic aperture radar (SAR); and will confirm that there are correspondences between metrics derived from GNSS reflected signals and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</i> -band SAR to inundated area, including wetlands covered by vegetation.

Automated mapping of the seasonal evolution of surface meltwater and its links to climate on the Amery Ice Shelf, Antarctica

Abstract. Surface meltwater is widespread around the Antarctic Ice Sheet margin and has the potential to influence ice shelf stability, ice flow and ice–albedo feedbacks. Our understanding of the seasonal and multi-year evolution of Antarctic surface meltwater is limited. Attempts to generate robust meltwater cover time series have largely been constrained by computational expense or limited ice surface visibility associated with mapping from optical satellite imagery. Here, we add a novel method for calculating visibility metrics to an existing meltwater detection method within Google Earth Engine. This enables us to quantify uncertainty induced by cloud cover and variable image data coverage, allowing time series of surface meltwater area to be automatically generated over large spatial and temporal scales. We demonstrate our method on the Amery Ice Shelf region of East Antarctica, analysing 4164 Landsat 7 and 8 optical images between 2005 and 2020. Results show high interannual variability in surface meltwater cover, with mapped cumulative lake area totals ranging from 384 to 3898 km2 per melt season. By incorporating image visibility assessments, however, we estimate that cumulative total lake areas are on average 42 % higher than minimum mapped values. We show that modelled melt predictions from a regional climate model provide a good indication of lake cover in the Amery region and that annual lake coverage is typically highest in years with a negative austral summer SAM index. Our results demonstrate that our method could be scaled up to generate a multi-year time series record of surface water extent from optical imagery at a continent-wide scale.

Temporal and spatial patterns of common hippopotamus populations in the Okavango Delta, Botswana

Abstract Conservation requires reliable estimates of a species' population and their spatial distribution. Knowledge of large‐scale habitat use and population trends in common hippopotamus (Hippopotamus amphibius; hereafter hippo) in wetland ecosystems is limited, with no studies of hippo populations in the Okavango Delta (the Delta), Botswana. We examined the drivers of long‐term trends and spatial patterns in the Delta's hippo population, using 13 years of aerial surveys (1996–2018) informing on the potential impacts of changes in inflow, flooding patterns and rainfall on hippo populations. We estimated temporal changes in hippo populations and relationships with rainfall and inflow. We also examined how spatially explicit hippo counts related to flood and surface water extent and vegetation class at different scales. The Delta's hippo population has increased, probably due to increasing long‐term rainfall and inflow, following a period of severe drought/low flooding. Hippos were positively associated with areas with high variation in flooding and negatively associated with broad‐scale surface water extent, indicating hippos avoided permanent swamps and main channels, probably due to water depth, lack of nearby grazing, and thick riparian vegetation. At a fine scale, hippos relied on large lagoons for daytime refuge. The most recent population estimate indicates that Botswana has the third highest hippo population in Africa, reflecting the importance of the Delta. Reductions in inflow and rainfall from climate change and water resource development would threaten these hippo populations by reducing grazing availability, lagoon sizes, and seasonal swamps. Ongoing monitoring of hippos should continue as they represent a good indicator for the entire floodplain ecosystem.

Application of Spectral Index-Based Logistic Regression to Detect Inland Water in the South Caucasus

The Caucasus is a diverse region with many climate zones that range from subtropical lowlands to mountainous alpine areas. The region is marked by irrigated croplands fed by irrigation canals, heavily vegetated wetlands, lakes, and reservoirs. In this study, we demonstrate the development of an improved surface water map based on a global water dataset to get a better understanding of the spatial distribution of small water bodies. First, we used the global water product from the European Commission Joint Research Center (JRC) to generate training data points by stratified random sampling. Next, we applied the optimal probability cut-off logistic regression model to develop surface water datasets for the entire Caucasus region, covering 19 Landsat tiles from May to October 2019. Finally, we used 6745 manually classified points (3261 non-water, 3484 water) to validate both the newly developed water dataset and the JRC global surface water dataset using an estimated proportion of area error matrix to evaluate accuracy. Our approach produced surface water extent maps with higher accuracy (89.2%) and detected 392 km2 more water than the global product (86.7% accuracy). We demonstrate that the newly developed method enables surface water detection of small ponds and lakes, flooded agricultural fields, and narrow irrigation channels, which are particularly important for mosquito-borne diseases.

An Adaptive Thresholding Approach toward Rapid Flood Coverage Extraction from Sentinel-1 SAR Imagery

Flood disasters have a huge effect on human life, the economy, and the ecosystem. Quickly extracting the spatial extent of flooding is necessary for disaster analysis and rescue planning. Thus, extensive studies have utilized optical or radar data for the extraction of water distribution and monitoring of flood events. As the quality of detected flood inundation coverage by optical images is degraded by cloud cover, the current data products derived from optical sensors cannot meet the needs of rapid flood-range monitoring. The presented study proposes an adaptive thresholding method for extracting water coverage (AT-EWC) regarding rapid flooding from Sentinel-1 synthetic aperture radar (SAR) data with the assistance of prior information from Landsat data. Our method follows three major steps. First, applying the dynamic surface water extent (DSWE) algorithm to Landsat data acquired from the year 2000 to 2016, the distribution probability of water and non-water is calculated through the Google Earth Engine platform. Then, current water coverage is extracted from Sentinel-1 data. Specifically, the persistent water and non-water datasets are used to automatically determine the type of image histogram. Finally, the inundated areas are calculated by combining the persistent water and non-water datasets and the current water coverage as derived from the above two steps. This approach is fast and fully automated for flood detection. In the classification results from the WeiFang and Ji’An sites, the overall classification accuracy of water and land detection reached 95–97%. Our approach is fully automatic. In particular, the proposed algorithm outperforms the traditional method over small water bodies (inland watersheds with few lakes) and makes up for the low temporal resolution of existing water products.

Automated Training Data Generation from Spectral Indexes for Mapping Surface Water Extent with Sentinel-2 Satellite Imagery at 10 m and 20 m Resolutions

This study presents an automated methodology to generate training data for surface water mapping from a single Sentinel-2 granule at 10 m (4 band, VIS/NIR) or 20 m (9 band, VIS/NIR/SWIR) resolution without the need for ancillary training data layers. The 20 m method incorporates an ensemble of three spectral indexes with optimal band thresholds, whereas the 10 m method achieves similar results using fewer bands and a single spectral index. A spectrally balanced and randomly generated set of training data based on the index values and optimal thresholds is used to fit machine learning classifiers. Statistical validation compares the 20 m ensemble-only method to the 20 m ensemble method with a random forest classifier. Results show the 20 m ensemble-only method had an overall accuracy of 89.5% (±1.7%), whereas the ensemble method combined with the random forest classifier performed better, with a ~4.8% higher overall accuracy: 20 m method (94.3% (±1.3%)) with optimal spectral index and SWIR thresholds of −0.03 and 800, respectively, and 10 m method (93.4% (±1.5%)) with optimal spectral index and NIR thresholds of −0.01 and 800, respectively. Comparison of other supervised classifiers trained automatically with the framework typically resulted in less than 1% accuracy improvement compared with the random forest, suggesting that training data quality is more important than classifier type. This straightforward framework enables accurate surface water classification across diverse geographies, making it ideal for development into a decision support tool for water resource managers.

The Boreal–Arctic Wetland and Lake Dataset (BAWLD)

Abstract. Methane emissions from boreal and arctic wetlands, lakes, and rivers are expected to increase in response to warming and associated permafrost thaw. However, the lack of appropriate land cover datasets for scaling field-measured methane emissions to circumpolar scales has contributed to a large uncertainty for our understanding of present-day and future methane emissions. Here we present the Boreal–Arctic Wetland and Lake Dataset (BAWLD), a land cover dataset based on an expert assessment, extrapolated using random forest modelling from available spatial datasets of climate, topography, soils, permafrost conditions, vegetation, wetlands, and surface water extents and dynamics. In BAWLD, we estimate the fractional coverage of five wetland, seven lake, and three river classes within 0.5 × 0.5∘ grid cells that cover the northern boreal and tundra biomes (17 % of the global land surface). Land cover classes were defined using criteria that ensured distinct methane emissions among classes, as indicated by a co-developed comprehensive dataset of methane flux observations. In BAWLD, wetlands occupied 3.2 × 106 km2 (14 % of domain) with a 95 % confidence interval between 2.8 and 3.8 × 106 km2. Bog, fen, and permafrost bog were the most abundant wetland classes, covering ∼ 28 % each of the total wetland area, while the highest-methane-emitting marsh and tundra wetland classes occupied 5 % and 12 %, respectively. Lakes, defined to include all lentic open-water ecosystems regardless of size, covered 1.4 × 106 km2 (6 % of domain). Low-methane-emitting large lakes (&gt;10 km2) and glacial lakes jointly represented 78 % of the total lake area, while high-emitting peatland and yedoma lakes covered 18 % and 4 %, respectively. Small (&lt;0.1 km2) glacial, peatland, and yedoma lakes combined covered 17 % of the total lake area but contributed disproportionally to the overall spatial uncertainty in lake area with a 95 % confidence interval between 0.15 and 0.38 × 106 km2. Rivers and streams were estimated to cover 0.12 × 106 km2 (0.5 % of domain), of which 8 % was associated with high-methane-emitting headwaters that drain organic-rich landscapes. Distinct combinations of spatially co-occurring wetland and lake classes were identified across the BAWLD domain, allowing for the mapping of “wetscapes” that have characteristic methane emission magnitudes and sensitivities to climate change at regional scales. With BAWLD, we provide a dataset which avoids double-accounting of wetland, lake, and river extents and which includes confidence intervals for each land cover class. As such, BAWLD will be suitable for many hydrological and biogeochemical modelling and upscaling efforts for the northern boreal and arctic region, in particular those aimed at improving assessments of current and future methane emissions. Data are freely available at https://doi.org/10.18739/A2C824F9X (Olefeldt et al., 2021).