Newsletter 3
June 2015

GLaSS newsletter banner
Dear reader,

Spectacular news from ESA: the Sentinel-2 satellite has been launched successfully! The first images are already available. With last weeks launch of Sentinel-2, the period of the operational optical Sentinel satellite series has started. GLaSS has been preparing algorithms and tools for monitoring water quality monitoring of lakes with this satellite, as well as with Sentinel-3 (scheduled for launch autumn 2015).

In this newsletter we will show you our first results of the global lakes use cases, in which we will demonstrate what can be done for water quality monitoring with the new Sentinel satellites and the GLaSS tools. This summer, we are busy with field validations. For this, we also created the first optical in situ protocol comparison for inland waters, which you will find further down in this newsletter.
Best regards,
The GLaSS team

Shallow eutrophic lakes - potentially toxic algae
Problems with eutrophication and algal blooms all around the world have generated the need for cost-effective ways of monitoring of the ecological status of water bodies and its changes. Presently, the routine field monitoring is both time- and resources-consuming, being able to cover only few field sites with a maximum frequency of once per month or even less for smaller lakes. Earth Observation methods have proved to give far better coverage both in spatial and temporal scale for evaluation of in-water constituents in the water bodies. GLaSS has selected several lakes with the presence of potentially toxic cyanobacteria blooms and studied them based on satellite data.
The images above show a cyanobacterial bloom development according to the maximum chlorophyll index (a) based on the MERIS satellite image of Lake Peipsi (11 July 2010), and the dominant cyanobacteria (b) Dolichospermum lemmermannii, (c) Microcystis wesenbergii, (d) Aphanizomenon flos-aquae). The complete results of this use case are expected at the end of August.

Deep clear lakes - study the effect of climate change
Deep clear lakes are less vulnerable to eutrophication than small shallow lakes. However, a continuous input of nutrients has led to increasing eutrophication in many of them. It is important to study potential eutrophication in these lakes, because due to the size of the lakes the eutrophication process is slow and easily goes unnoticed but is also hard to revers. GLaSS therefore evaluates the trends of chlorophyll-a concentrations as a proxy of the trophic status based on the complete MERIS satellite archive, using the GLaSS tools. The case studies focus on lakes in different climatic and cultural regions of the world. The initial results were presented at the ESA Sentinel-3 for Science workshop in Venice. The extended abstract will be published in its proceedings.
Tea-coloured lakes - can we still use optical measurements?

Every water optical water quality expert knows: water absorbs light strongly, especially in lakes with a high concentration of coloured dissolved organic matter (CDOM). These are for example peat lakes or lakes that are surrounded by large forests. These lakes are often left out in remote sensing studies, because of the low optical signal-to-noise level. GLaSS has especially taken these lakes in account for the atmospheric correction tests, algorithm comparisons and now also as a special use case, in which the possibilities and limitations for remote sensing for such lakes will be studied. The plots above shows the simulated Rrs in a typical humic lake in Finland (Pääjärvi) and in Lake Garda, Italy. Rrs was simulated with the Hydrolight model and shows that the effect the absorption of CDOM can have on the reflected light spectrum when TSM and Chl-a levels are roughly the same. The low light levels in humic lakes lead to problems especially in the atmospheric correction of satellite images.

Recovery of shallow lakes - European games in Baku
Lake Böyük ┼×hor is used to be one of the most polluted lakes in the world. However, currently the first European games are taking place on the lake shores. A large restoration project has been carried out and will be continued after the games. The effect of the dredging, damming off of parts of the lake to serve as sludge depot, and improvement of the water quality can be followed via satellite monitoring. The maps above show the lake with about 1 year of time interval, as observed with Landsat 8 (as proxy for Sentinel-2). Yellow, orange and red indicate a large(r) chance of oil present on the water surface, green and blue indicate a low change. 
Click here to download the preliminary results (pdf).
Mine tailing ponds - methods to locate them
In some areas in the world the location of mines and their (often highly polluted) tailing ponds are well known and frequently undergo risk analysis. However, that is not the case for all countries and continents. Especially in remote areas the precise location of ponds and their risks might not be clear. With its synoptic view, remote sensing can help to detect these ponds. The challenge is that ponds can have very distinct but also very different colours. As shown in the satellite image above (example from Finland), ponds might be red, blue, white..and are often small. This size requires high resolution satellite imagery, and therefore large amounts of data to be processed. We are working on an algorithm to automatically scan large amounts of data, discard land, clouds and open ocean and check the remaining coastal and inland water bodies for spectral properties that indicate that they might be mine tailing ponds.
Water Framework Directive reporting
The European Water Framework Directive (WFD) aims to establish a framework for the protection of inland surface waters, transitional waters, coastal waters and groundwater. The Member States have to achieve a good water status (the scale is High, Good, Moderate, Poor, and Bad status) or good potential (in heavily modified or artificial water bodies) in terms of chemical and ecological parameters. GLaSS explores how Earth Observation (EO) data (and especially that of the forthcoming Sentinel data) could be used to support WFD monitoring and reporting. Several examples of EO based data for WFD reporting will be provided. The time series above presents an example from Finland, where EO data is already used for the WFD. It presents a Chl-a time series for year 2011 with EO (black) and in situ data (red). The content of this figure was developed together with experts performing classification for the Water Framework directive reporting.

GLaSS in situ protocols for optical inland water monitoring
Ocean Optics protocols such as the SeaWifs protocols as not always suitable for inland waters. For example, with filtering standard amounts of water, filters might get clogged because of the higher turbidity, instruments can reach the bottom during depth-profiles, and surrounding mountain tops can accidentally be measured with Lsky measurements. GLaSS made a start to generate a general protocol document by collecting all partners inland waters protocols and comparing these to the standards for the open ocean.
Click here to download the protocols report (pdf).
During this summer, field campaigns will be carried out in Estonia, Italy, Finland and the Netherlands. A large campaign including an inter-comparison, in co-operation with the project GloboLakes, will take place in Sweden at the end of August.
The results of GLaSS will be made available to a larger audience. Training material based on the global use cases will be developed for students in ecology, environmental sciences, water management, remote sensing and GIS. If you are interested in these materials, please contact us! 
More information and contact
If you want to read more about the content of the project, check out our website, follow us on twitter (@glass_project) or simply reply to this email.