Currently, water monitoring programs are mainly based on in situ measurements; however, this approach is time-consuming, expensive, and may not reflect the status of the whole water body. The availability of Multispectral Imager (MSI) and Ocean and Land Colour Instrument (OLCI) free data with high spectral, spatial, and temporal resolution has increased the potential of adding remote sensing techniques into monitoring programs, leading to improvement of the quality of monitoring water. This study introduced an optical water type guided approach for boreal regions inland and coastal waters to estimate optical water quality parameters, such as the concentration of chlorophyll-a (Chl-a) and total suspended matter (TSM), the absorption coefficient of coloured dissolved organic matter at a wavelength of 442 nm (aCDOM(442)), and the Secchi disk depth, from hyperspectral, OLCI, and MSI reflectance data. This study was based on data from 51 Estonian and Finnish lakes and from the Baltic Sea coastal area, which altogether were used in 415 in situ measurement stations and covered a wide range of optical water quality parameters (Chl-a: 0.5–215.2 mg·m−3; TSM: 0.6–46.0 mg·L−1; aCDOM(442): 0.4–43.7 m−1; and Secchi disk depth: 0.2–12.2 m). For retrieving optical water quality parameters from reflectance spectra, we tested 132 empirical algorithms. The study results describe the best algorithm for each optical water type for each spectral range and for each optical water quality parameter. The correlation was high, from 0.87 up to 0.93, between the in situ measured optical water quality parameters and the parameters predicted by the optical water type guided approach.
Retrieval of aquatic biogeochemical variables, such as the near-surface concentration of chlorophyll-a (Chla) in inland and coastal waters via remote observations, has long been regarded as a challenging task. This manuscript applies Mixture Density Networks (MDN) that use the visible spectral bands available by the Operational Land Imager (OLI) aboard Landsat-8 to estimate Chla. We utilize a database of co-located in situ radiometric and Chla measurements (N 4,354), referred to as Type A data, to train and test an MDN model (MDN A). This algorithm's performance, having been proven for other satellite missions, is further evaluated against other widely used machine learning models (e.g., support vector machines), as well as other domain-specific solutions (OC3), and shown to offer significant advancements in the field. Our performance assessment using a held-out test data set suggests that a 49% (median) accuracy with near-zero bias can be achieved via the MDN A model, offering improvements of 20 to 100% in retrievals with respect to other models. The sensitivity of the MDN A model and benchmarking methods to uncertainties from atmospheric correction (AC) methods, is further quantified through a semi-global matchup dataset (N 3,337), referred to as Type B data. To tackle the increased uncertainties, alternative MDN models (MDN B) are developed through various features of the Type B data (e.g., Rayleigh-corrected reflectance spectra ρ s). Using held-out data, along with spatial and temporal analyses, we demonstrate that these alternative models show promise in enhancing the retrieval accuracy adversely influenced by the AC process. Results lend support for the adoption of MDN B models for regional and potentially global processing of OLI imagery, until a more robust AC method is developed. Index Terms-Chlorophyll-a, coastal water, inland water, Landsat-8, machine learning, ocean color, aquatic remote sensing.
The Sentinel-3 mission launched its first satellite Sentinel-3A in 2016 to be followed by Sentinel-3B and Sentinel-3C to provide long-term operational measurements over Earth. Sentinel-3A and 3B are in full operational status, allowing global coverage in less than two days, usable to monitor optical water quality and provide data for environmental studies. However, due to limited ground truth data, the product quality has not yet been analyzed in detail with the fiducial reference measurement (FRM) dataset. Here, we use the fully characterized ground truth FRM dataset for validating Sentinel-3A Ocean and Land Colour Instrument (OLCI) radiometric products over optically complex Estonian inland waters and Baltic Sea coastal areas. As consistency between satellite and local data depends on uncertainty in field measurements, filtering of the in situ data has been made based on the uncertainty for the final comparison. We have compared various atmospheric correction methods and found POLYMER (POLYnomial-based algorithm applied to MERIS) to be most suitable for optically complex waters under study in terms of product accuracy, amount of usable data and also being least influenced by the adjacency effect.
European countries are obligated to monitor and estimate ecological status of lakes under European Union Water Framework Directive (2000/60/EC) for sustainable lakes’ ecosystems in the future. In large and shallow lakes, physical, chemical, and biological water quality parameters are influenced by the high natural variability of water level, exceeding anthropogenic variability, and causing large uncertainty to the assessment of ecological status. Correction of metric values used for the assessment of ecological status for the effect of natural water level fluctuation reduces the signal-to-noise ratio in data and decreases the uncertainty of the status estimate. Here we have explored the potential to create synergy between optical and altimetry data for more accurate estimation of ecological status class of lakes. We have combined data from Sentinel-3 Synthetic Aperture Radar Altimeter and Cryosat-2 SAR Interferometric Radar Altimeter to derive water level estimations in order to apply corrections for chlorophyll a, phytoplankton biomass, and Secchi disc depth estimations from Sentinel-3 Ocean and Land Color Instrument data. Long-term in situ data was used to develop the methodology for the correction of water quality data for the effects of water level applicable on the satellite data. The study shows suitability and potential to combine optical and altimetry data to support in situ measurements and thereby support lake monitoring and management. Combination of two different types of satellite data from the continuous Copernicus program will advance the monitoring of lakes and improves the estimation of ecological status under European Union Water Framework Directive.
A temporally and spatially detailed historical (1985–2018) analysis of cyanobacteria blooms was performed in the Curonian Lagoon (Lithuania, Russia), the largest coastal lagoon in the Baltic Sea. Satellite data allowed for the mapping of cyanobacteria surface accumulations, so-called “scums”, and of chlorophyll-a concentration. The 34-year time series shows a tendency towards later occurrence (October–November) of the cyanobacteria scum presence, whereas the period of its onset (June–July) remains relatively constant. The periods when scums are present, “hot moments”, have been consistently increasing in duration since 2008. The differences in the starting, ending and annual duration of cyanobacteria blooms have been significantly altered by hydro-meteorological conditions (river discharge, water temperature, wind conditions) and their year-round patterns. The most important environmental factors that determined the temporal changes of the scum presence and area were the standing stock of cyanobacteria and the ambient wind conditions. The “hotspots”, the areas where the blooms most likely occur, were distributed in the south-southwestern and central parts of the lagoon. The least affected areas were the northern part, which is connected to the coastal waters of the Baltic Sea, and the Nemunas River delta region. The longstanding, well-established spatial patterns of cyanobacteria blooms were linked to hydrodynamic features, namely renewal time and current patterns, and to potential nutrient sources that included muddy sediments and the locations of colonies of piscivorous birds. Our findings confirmed that the annual and seasonal variations of cyanobacteria blooms and their regulation are a complex issue due to interactions between multiple factors over spatially and temporally broad scales. Despite great progress in the prevention and control of eutrophication and cyanobacteria blooms, the lagoon is still considered to be in a poor ecological status. This work provides a new and missing understanding on the spatial and temporal extent of cyanobacteria blooms and the factors that govern them. Such an understanding can help in planning management strategies, forecasting the magnitude and severity of blooms under changing nutrient loads and potential climate scenarios.
Lake water quality monitoring has the potential to be improved through integrating detailed spatial information from new generation remote sensing satellites with high frequency observations from in situ optical sensors (WISPstation). We applied this approach for Lake Trasimeno with the aim of increasing knowledge of phytoplankton dynamics at different temporal and spatial scales. High frequency chlorophyll-a data from the WISPstation was modeled using non-parametric multiplicative regression. The 'day of year' was the most important factor, reflecting the seasonal progression of a phytoplankton bloom from July to September. In addition, weather factors such as the east-west wind component were also significant in predicting phytoplankton seasonal and diurnal patterns. Sentinel 3-OLCI and Sentinel 2-MSI satellites delivered 42 images in 2018 that successfully mapped the spatial and seasonal change in chlorophyll-a. The potential influence of localized inflows in contributing to increased chlorophyll-a in midsummer was visualized. The satellite data also allowed an estimation of quality status at a much finer scale than traditional manual methods. Good correspondence was found with manually collected field data but more significantly, the greatly increased spatial and temporal resolution provided by satellite and WISPstation sensors clearly offers an unprecedented resource in the research and management of aquatic resources.
A preliminary analysis on Subalpine lakes The major lakes in the Italian subalpine ecoregion are part of a Use Case to be developed in the Lakes CCI in synergy with LTER (Long Term Ecological Research Network). Water quality of these lakes is in danger due to the increasing demand for freshwater, the effect of climate change and the anthropogenic pressure. Lake Surface Water Temperature (LSWT) for subalpine lakes were derived by Pareeth et al. (2017); while in this panel results based on Lake Water Leaving Reflectance (LWLR), and the derived Chlorophyll-a (Chl-a) concentrations products, are presented. Perspectives and aims Lakes are sentinels, regulators and integrators of climate change: • many lake hydrological and biophysical variables are potential indicators of current climate change • lakes regulate GhG but potentially may be important source of re-emission to the atmosphere • lake sediments are archives of past climate • the parameters needed to understand their role in climate change are numerous: level, extent, temperatures, ice, water colour • remote sensing techniques are very suitable to examine the role of lakes in climate change at the global scale Conclusions • The multi-temporal analysis on LSWT and LWLR and related Chl-a products highlighted the contribute of satellite data for a synoptic long-term and continuous monitoring of lakes at regional scale. • Further activities are planned need to combine EO with LTER data for analyzing patterns and trends in a climate change scenario. The Lakes CCI project is part of the Climate Change Initiative (CCI) European Space Agency (ESA) and run for three years. Main objective is to exploit satellite data to create the largest and longest possible consistent, global record of the five lake variables: lake water level (LWL), extent (LWE), temperature (LSWT), surface-leaving reflectance (LWLR), and ice cover (LIC). The project extends the state-of-the-art in altimetry, thermal, optical and SAR, drawing from over 30 satellite sensors. Research priorities include extending current methodologies to both older sensors and the latest Sentinels, improved uncertainty characterization and product inter-consistency checks. The first release of the Lakes ECV data set is expected in January 2020, including > 200 lakes Are these variables measurable from remote sensing with sufficient confidence to address climate questions? Please visit our website and provide your inputs to us! Methods-Earth Observation (EO) data from different satellite sensors (MERIS, OLI, MSI, OLCI) were used for acquiring timely, frequent synoptic information of Garda, Iseo, Como and Maggiore lakes with a focus on phytoplankton abundance (Chl-a concentration) in the period 2003-2018. Imagery were processed for radiometric and atmospheric correction, and bio-optical models (C2R, BOMBER) were applied to retrieve Chl-a. The products were validated against field data as in Odermatt et al., 2010, Giardino et al., 2014, and Bresciani et al., 2018. Trend analysis was conducted with Seasonal Kendall test, and Sen Slope estimator was calculated as estimation of median annual slope. Trend was considered meaningful if p < 0.05 and if the Sen's Slope was at least 1% of the median value. Results-Trends show a slight tendency to increase of Chl-a concentrations during the period considered, particularly for Iseo and Maggiore lakes. From these time-series products was then possible to extrapolate phenology data such as the start of season of phytoplankton. Summer mean LSWT from 1986 to 2015; data were smoothed using the LOESS interpolation (blue line). The grey area depicts the 95% confidence interval. (Pareeth et al., 2017) Chl-a concentrations from multiple satellites observations from 2003 to 2018 Maps: Chl-a concentrations for the time window of algal bloom events in 2016 in: Garda (a-c), Como (d-f), Maggiore (g-i). Histograms: Chl-a concentration on x-axis and fraction of total pixel on y-axis. Vertical lines indicate WFD boundaries for water quality status classification based on Chl-a concentration (Bresciani et al., 2018)
Razionale: Il riscaldamento globale e l'eutrofizzazione causano un aumento della presenza di ciano-batteri nelle acque interne. Un'intensa fioritura di questi organismi fitoplanctonici emerge spesso in superficie come uno strato di schiuma (scum) contenente alte concentrazioni di tossine e me-taboliti secondari. Il contatto con queste tossine pone un rischio diretto per la salute. Pertanto, il monitoraggio della concentrazione di cianobatteri e la presenza di scum nei laghi sono attual-mente argomento di grande interesse . Problema monitoraggio: Il telerilevamento ottico è uno strumento validato per il rilevamento, il monitoraggio e lo sviluppo di una migliore comprensione dello stato dei laghi. Tuttavia, non può essere utilizzato in caso di coper-tura nuvolosa rendendo così difficile la caratterizzazione spaziale e temporale delle fioriture algali per un'analisi ecologica completa. Proponiamo quindi un approccio sinergico che coinvolge immagini ottiche e immagini SAR per monitorare la proliferazione di cianobatteri algali. Le immagini satellitari sono for-nite dai satelliti Sentinel-1,-2 e-3. Laguna dei Curi: La Laguna dei Curi è un ecosistema altamente dinamico e complesso con molti processi interagenti. È un bacino semi chiuso, influenzato dallo scambio delle acque dolci del fiume Nemunas e delle acque salmastre del Mar Baltico. A partire da fine giugno e inizio luglio e fino alla fine di otto-bre, i cianobatteri diventano le specie prevalenti durante e le loro fioriture sono spesso intense. È stata selezionata come area studio in quanto: Sono disponibili i prodotti di secondo livello del Sentinel-1 su cui si basa il presente approccio. Sebbene tali prodotti non sono generalmente di-stribuiti per acque interne, la stretta vicinanza della Laguna al mar Baltico, ne rende un'eccezione. È stata ampiamente studiata con RS ottico  È spesso soggetta a copertura nuvolosa. Dall'analisi effettuata da Mercury  sulle immagini MODIS-Terra 2001-2011 si stima infatti che annualmente la Laguna dei Curi è mediamente coperta da nuvole per il 76 % dei giorni. Tale percentuale si riduce al 64% $nel periodo estivo. RS SAR: Si propone un approccio "semi-automatico" basato sui prodotti L2 di Sentinel-1, in parti-colare i prodotti OCN-OWI (Ocean Wind Fields). Essi contengono sia i valori di vento forniti dal modello meteorologico dell'ECMWF (European Centre for Medium-Range Weather Forecasts), sia quelli stimati a partire dall'immagine SAR mediante l'inversione di un modello geofisico che dipende dalla rugo-sità della superficie dell'acqua . Differen-ze significative fra i due campi di vento indicano la presenza di sostanze galleg-gianti sulla superficie dell'acqua (es. ghiaccio, olio, scum). Dall'analisi di 80 imma-gini S1 abbiamo empiricamante definito un "allarme di scum" quando il vento SAR è in-feriore al 10% del vento ECMWF. Definiamo il rapporto dove è la velocità del vento stimata da Sen-tinel 1 e indica la velocità del vento ECMWF a priori. La condizione WR<0.1, nel perio-do dell'anno che va da fine giugno a fine ottobre, indica con buona probabilità la presenza di scum. RS Ottico: Data la bassissima trasparenza tipica delle ac-que della Laguna dei Curi e la tendenza delle alghe disperse nella colonna d'acqua ad accu-mularsi vicino alla superficie, le stime satelli-tari delle concentrazioni di clorofilla (Chl-a) sono rilevanti solo per la parte superiore della colonna d'acqua e quindi possono essere con-siderati un segnale della presenza di schiuma. La concentrazione di clorofilla si può determi-nare dal rapporto delle riflettanze nelle bande NIR e RED sfruttando l'elevata diffe-renza di retrodiffusione e assorbimento tra acqua con e senza scum. In particolare, la concentrazione di Chl-a si determina per Sen-tinel-2 con: e per Sentinel-3 con: dove con si intende la riflettanza della banda la cui lunghezza d'onda centrale è [2,3]. Validazione qualitativa: In questo intervallo temporale, in 28 giorni sono disponibili immagini sia ottiche e sia SAR. L'approccio basato sull'analisi delle immagini SAR restituisce una classificazione della laguna che è coerente con quella ottenuta dall'analisi delle immagini ottiche e con i dati meteorolo-gici (alta velocità del vento) per 22 casi. Si osserva quindi un tasso di accordo del 79%. La diversa classificazione può essere spiegata dal divario temporale di acquisizione per il 14% dei casi e solo il restante 7% mostra un possibile disaccordo (non investigabile per mancanza di verità a terra). Andamento temporale di WR (rettangoli rossi), delle concentrazioni di Chl-a (rettangoli neri) e delle precipitazioni totali (sfondo grigio) da luglio ad ottobre 2018. Immagini con Chl-a <72 [ mg/m 3 ] o WR> 0,1 sono classificate come "No BAD" . Lo schema tem-porale è spiegato nell'inserto a sinistra. La classificazione basata sull'indice WR non sembra essere mai falsata dalla presenza di precipitazioni. La precipitazione totale pro-viene dai dati orari ERA5 sul database di rianalisi a livello singolo. Validazione qualitativa: Le mappa ottenute tramite SAR sono consistenti con quanto osservato da dati ot-tici e con lo stato del vento nell'intervallo di tempo fra le acquisizioni. Tale approc-cio può quindi essere utilizzato per infittire le serie storiche di fioriture algali nella Laguna dei Curi e quindi migliorarne il monitoraggio. Mappe di concentra-zione di Chl-a derivate da S3 e da S2; mappa dell'indice WR; evoluzione spazio-temporale del vettore del vento nell'interval-lo di acquisizione. Il vettore del vento proviene dal database di rianalisi per dati orari a singolo livello di pressione ERA5.  Hunter, P., et al., 2017. INFORM D5.15  Kerbaol, V. (2007) Conclusioni L'approccio SAR "semi-automatico" qui discusso presenta: il vantaggio di essere un approccio semplice che non richiede l'auslilio di altri dati; l'inconveniente di avere una bassa risoluzione e di non essere utilizzabile per qualsiasi corpo idrico interno. I risultati presentati supportano fortemente l'uso dei prodotti L2 S1 per migliorare il rileva-mento spazio-temporale dei fenomeni di scum nella Laguna dei Curi. L'uso dell'approccio pro-posto può sicuramente essere un vantaggio sia per i gestori delle risorse idriche che per gli scienziati. Una campagna sul campo potrebbe tuttavia perfezionare la soglia del rapporto del vento determinata empiricamente in questo documento.
The European Union and the European Space Agency (EU/ESA) have promoted since 1998 (Baveno Manifesto*) the GMES Programme (Global Monitoring for Environment and Security), nowadays called Copernicus (www.copernicus.eu). In the water quality domain, the use of Copernicus Sentinel missions and services improvements occur studying chlorophyll-a (Chl-a) concentrations, phycocyanin (PC), total suspended matter (TSM), colored dissolved organic matter (CDOM) and water surface temperature, with attention to the Cost/Benefit analysis (environmental and economic). The fundamentals of Earth Observation (EO), Geographic Information (GI) and Geomatics techniques, are crucial for the development of innovative strategies for professional skills adequacy and capacity building , supporting Copernicus user uptake. One of the main goal is to help bridging gaps between supply and demand in terms of education and training for the geospatial sector, reinforcing the existing academic proposal, fostering the uptake and the integration of Copernicus geospatial data and services in end-user applications (www.eo4geo.eu). Referring to the water resources domain, topics to be included in Summer Schools training course are: atmospheric correction, optical properties, algorithms to retrieve water quality information. The outcomes of several concluded or ongoing international projects (H2020 EOMORES, Earth Observation based services for Monitoring and Reporting of Ecological Status, and ESA-SEOM SEN2CORAL) should be also taken into account, being aimed at developing commercial services for monitoring the quality of inland and coastal water bodies and to improve processing algorithms. Such activities are supported by innovative and strategical novelties like the complete free access to the imagery archives of the Copernicus Sentinel missions and the availability of the dedicated data processing software SNAP (SeNtinel Application Platform). Additionally, cloud-based solution embedding global archive catalogues (e.g. Google Earth Engine) enables planetary-scale geospatial analyses, allowing real world scenarios to be operationally tackled.
The European Parliament and The Council of the European Union have established the Water Framework Directive (2000/60/EC) for all European Union member states to achieve, at least, “good” ecological status of all water bodies larger than 50 hectares in Europe. The MultiSpectral Instrument onboard European Space Agency satellite Sentinel-2 has suitable 10, 20, 60 m spatial resolution to monitor most of the Estonian lakes as required by the Water Framework Directive. The study aims to analyze the suitability of Sentinel-2 MultiSpectral Instrument data to monitor water quality in inland waters. This consists of testing various atmospheric correction processors to remove the influence of atmosphere and comparing and developing chlorophyll a algorithms to estimate the ecological status of water in Estonian lakes. This study shows that the Sentinel-2 MultiSpectral Instrument is suitable for estimating chlorophyll a in water bodies and tracking the spatial and temporal dynamics in the lakes. However, atmospheric corrections are sensitive to surrounding land and often fail in narrow and small lakes. Due to that, deriving satellite-based chlorophyll a is not possible in every case, but initial results show the Sentinel-2 MultiSpectral Instrument could still provide complementary information to in situ data to support Water Framework Directive monitoring requirements.
A new autonomous above water radiometer system (WISPstation) was developed based on the experience with the handheld WISP-3 system. The instrument records radiance and irradiance with an extended wavelength range of 350 to 1100 nm in two viewing directions, which enables continuous and autonomous high-quality measurements for water quality monitoring and satellite validation. All channels are measured with a single spectrometer and an optical multiplexer. This design makes resulting remote sensing reflectances less sensitive to radiometric and spectral calibration errors and drifts. In various Copernicus projects (TAPAS, EOMORES and MONOCLE) the WISPstation is being tested in highly diverse water types and environmental conditions, ranging from case-1 in Mediterranean coastal waters to turbid waters with cyanobacteria proliferation in lakes and lagoons. In view of its initial scientific application, the system is designed to reliably produce high frequency observations to quantify variability in physical and biological water system parameters. The WISPstation results are stored in the online database WISPcloud allowing users to extract data for analysis. A web interface is being set up to visualise the measurements. We present spectral results and time series analysis for various locations. Introduction.
In this work a dataset of maps of Chl-a derived from different sensors (MERIS-OLI-MSI-OLCI) satellite images is used to evaluate water quality of the four most important Italian subalpine lakes (Garda, Maggiore, Iseo, Como) in the period 2003-2018. In order to produce Chl-a concentration maps, imagery needs to be processed along a processing chain, which includes radiometric correction (to convert digital numbers into at-satellite-atmospheric radiance), atmospheric correction to obtain Remote sensing reflectance (C2R for MERIS, 6SV for MSI and OLI, Polymer for OLCI), and finally the application of bio-optical models to retrieve chl-a concentration. The chain products (Rrs and Chl-a) were validated against field data as in Odermatt et al.,2010, Giardinoet al., 2014 and Brescianiet al., 2018.
Under programme Copernicus European Space Agency has been launched new satellite sensors, which provide us spectral, spatial and temporal resolution to monitor optically complex waters. However, the variation of lakes’ water colour is large and standard products often fail. Classification of optical water types helps to clarify relationships between different properties inside a certain class and quantify variation between the classes.
The H2020 project EOMORES will develop operational monitoring and reporting services for inland and coastal water quality based on a combination of the most up-to-date satellite data, innovative insitu instruments and ecological models.
The H2020 project EOMORES will develop operational monitoring and reporting services for inland and coastal water quality based on a combination of the most up-to-date satellite data, innovative in situ instruments and ecological models. Lakes, reservoirs and coastal water bodies constitute essential components of the hydrological and biogeochemical water cycles, and influence many aspects of ecology, economy, and human welfare, providing ecosystem services in multiple and sometimes conflicting ways. Knowledge about the state of inland and coastal water bodies is therefore of great interest.
Cyanobacteria form spectacular mass occurrences almost annually in the Baltic Sea. These harmful algal blooms are the most visible consequences of marine eutrophication, driven by a surplus of nutrients from anthropogenic sources and internal processes of the ecosystem. We present a novel Cyanobacterial Bloom Indicator (CyaBI) targeted for the ecosystem assessment of eutrophication in marine areas. The method measures the current cyanobacterial bloom situation (an average condition of recent 5 years) and compares this to the estimated target level for 'good environmental status' (GES). The current status is derived with an index combining indicative bloom event variables. As such we used seasonal information from the duration, volume and severity of algal blooms derived from earth observation (EO) data. The target level for GES was set by using a remote sensing based data set named Fraction with Cyanobacterial Accumulations (FCA; Kahru & Elmgren, 2014) covering years 1979-2014. Here a shift-detection algorithm for time series was applied to detect time-periods in the FCA data where the level of blooms remained low several consecutive years. The average conditions from these time periods were transformed into respective CyaBI target values to represent target level for GES. The indicator is shown to pass the three critical factors set for marine indicator development, namely it measures the current status accurately, the target setting can be scientifically proven and it can be connected to the ecosystem management goal. An advantage of the CyaBI method is that it's not restricted to the data used in the development work, but can be complemented, or fully applied, by using different types of data sources providing information on cyanobacterial accumulations.