Monitoring Good Environmental Status (GES) Indicators and Characteristics
The main goal of the Marine Directive is to achieve Good Environmental Status of EU marine waters by 2020. The Directive defines Good Environmental Status (GES) as:
“The environmental status of marine waters where these provide ecologically diverse and dynamic oceans and seas which are clean, healthy and productive”
GES means that the different uses made of the marine resources are conducted at a sustainable level, ensuring their continuity for future generations.
Monitoring achievement of the GES goals using Earth Observation
The roadmap to achieving any goal includes procuring tools and capabilities to be able to measure it. Measuring will eventually determine whether the goals were achieved, but most importantly – especially for such long term goals– measuring provides the continuous monitoring and assessment necessary to establish whether the current trend is moving in the right direction. And when it comes to environmental monitoring, this means: maps and graphs.
The production of ocean quality maps has significantly increased in the last decades thanks to exploitation of Earth Observation data. This has allowed us to continuously capture snapshots of physical variables over regional areas at consistent repeating time intervals (daily, weekly monthly, depending on the satellites used).
Many large projects have been launched along this path, among which it is worth mentioning MyOcean (http://www.myocean.eu/), EMODNET (http://www.emodnet-mediterranean.eu/) and ERA INTERIM (http://www.ecmwf.int/en/research/climate-reanalysis/ocean-reanalysis). They have facilitated the establishment of robust and continuously operational processing chains which provide updated oceans maps worldwide.
Melodies challenge and added value
So, why do we need yet another tool for computing Ocean state variables? The answer is simple.
All the existing large projects have created global databases, each with with standardised characteristics. Marine Directive monitoring for each Member State, however, requires customised analyses which depend on local charactesristics. For example, the physical status of the ocean at a given moment results from mechanisms ranging from decadal (such as the North Atlantic Oscillation) to hourly (such as wind or river forcings). Thus computing a single parameter such as the Sea Surface Temperature (SST) at global scales for a predetermined time period and range of depth values will not suit the case of all Member States (probably none of them!). Analyses must be tailor-made in terms of timescales, geographical regions and water depths.
A feature of ocean databases is that they are are huge - in 35 years' worth of daily data covering only Mediterranean Sea at 72 water depths, the temperature data alone amounts to >100 GBs of input data - and re-processing is computationally intensive. The challenge here lies in providing a tool which allows Member States to interactively (thus in a matter of minutes, at maximum) re-compute marine indicators with customised settings.
This is where the strong parallel processing power set-up of the MELODIES infrastructure comes into play!
Processing marine data is resource demanding but computationally simple as it mainly consists of performing basic algebra, such as averaging, integrating, etc. Therefore it is well suited to being split into simple smaller tasks that are each processed separately, and whose results are finally merged into one single result. The most widespread implementation of such approach is the well-known Hadoop Map Reduce paradigm, which is fully supported by the MELODIES processing infrastructure (see following figure).
Initial tests have shown that processing scales nearly linearly with the number of processing nodes adopted, thus demonstrating that the technical objective can be achieved by properly balancing needed HW infrastructure.
The MELODIES GES indicators analysis tool and portal
The MELODIES GES Indicators portal will therefore allow end users (depending on their privileges) to:
- Access available re-processed data
- Perform customised re-processing of data (only for privileged users)
Users will be able to define the GES indicator / variable to re-compute, and to provide ad-hoc parameters, such as :
- type of value: mean, max, min , trend, etc.
- geographical area / extension
- time frame of interest
- depth levels
After the re-computation, data will be made available to the user who submitted the processing and can be displayed on the Web Portal itself, like in the following figure.