The Black Sea and Danube region is constituted of three parts – the sea basin, coastal areas and catchment areas. They must be understood as a whole for an accurate understanding of the region's dynamics in a systematic and holistic manner. The Danube catchment, by far the biggest tributary of Black Sea, constitutes the second largest river basin in continental Europe with its area of more than 800.000 km2. Shared by 19 countries (nine of which are EU Member States), more than 81 million people of different cultures and languages live here, making it also the world’s most international river basin. Although the countries are very different in terms of economic strength, the region is strongly interlinked, with potential for further integration and growth. Thus, the region is connected through both opportunity and challenge. The policies of the countries are interdependent, however, they all benefit greatly from improved cooperation in, for example, completing missing transport links, reducing pollution and danger from floods, sustainable management of the environment, lowering dependency on energy providers from outside the region, and addressing demographic change. This presentation is about two parallel activities in ESA's Black Sea and Danube Initiative - Applications Line:
1) The Danube Environmental Risk Assessment Platform (DEAP) provides a suite of applications based on Earth Observation data to support environmental risk management within the Danube catchment. The purpose of the platform is to provide regional stakeholders, who currently do not regularly use EO data, with access to dynamic environmental assessments using such datasets. The service comprise a suite of cloud-based applications (deployed on CREODIAS DIAS) which are able to detect, monitor, analyse and characterise the sources of environmental problems using available EO data in conjunction with in-situ inputs, numerical modeling outputs and other reference data to deliver meaningful (actionable) maps, statistics and other data across 19 countries. At an operational level, the service identified industrial waste discharge, transport waste discharge, agricultural run-off, and ecosystem degradation in near real time, and represents a unique tool for regional agencies. Stakeholders include environmental protection agencies, port authorities, fisheries management agencies and development agencies, etc.
2) Earth Observation services for Black Sea Protection (EO4BSP) overlap the entire area of the Black Sea and propose a holistic approach that covers different elements with potential environmental impact. The project is implementing six services that are being delivered to a number of 13 stakeholders from the Black Sea riparian countries and one International organization - The Black Sea Commission. The services comprise a suite of cloud-based applications which will detect, monitor, analyze and characterize the sources of environmental problems using available EO imagery in conjunction with in-situ inputs and other reference data. Service applications are developed for deployment in the cloud and shall employ advanced dispersion modeling techniques in conjunction with EO Data to deliver meaningful (actionable) maps, statistics, and other data. The services are: S1 - Land Use Land Cover coastal changes, where analysis and modelling of land-use change trends and urbanization allow us to evaluate the spatial development patterns providing a key for effective planning practices in the context of Marine Strategy Framework Directive (MSFD) and Maritime Spatial Planning (MSP) implementation; S2 - Eutrophication, which represents one of the most severe and widespread environmental problems for coastal zone managers; S3 - Marine front identification and mesoscale circulation, providing data fusion, satellite observations, numerical modelling and data assimilation, as well as skill assessment and metrics with a focus on sea state, temperature, turbidity, and SPM, identification of ocean fronts; S4 - Oil tankers path identification, making use of historical AIS data; S5 - Oil spills identification and monitoring, where Sentinel-1 Ground Range Detected Level 1 data is used to detect and characterize possible oil spills; S6 - High-resolution water quality monitoring in anchorage areas, focussing on chlorophyll a (chl_a), turbidity, and total suspended matter (TSM).

The Black Sea and Danube region integrate three basic components - sea basin, coastal areas and catchment areas. Such a collection and integration is essential for an accurate understanding of the region's dynamics in a systematic and holistic manner. With the increasing demands on natural resources arising from extensive human activities, the region faces several environmental problems including serious water, land, air pollution from agriculture, industry and cities.
Agriculture is an important part of the economic development in the Black Sea and Danube region and the basin is home to one of the main global grain production areas as well as high-value crops such as vineyards, orchards and vegetable farming as well as livestock production. In parallel, intensive agriculture and deforestation often contribute to overexploitation of the soil, increase of soil erosion, subsequent sediment input and also have an influence on bed-load transport not only in the immediate area but also far downstream until the river reaches the sea, the final sediment trap. As a consequence, soil degradation is accelerating, exacerbated by human activities such as the inappropriate management of arable land, grassland and forest land. Besides the agriculture priorities, ensuring sustainable use of forest resources is equally important for improving economic, social and environmental conditions in the region. Many countries are currently undergoing inventories of their forests and developing forest investment plans taking into account various public interests and support of forest reconstruction, afforestation, shelterbelt planting, as well as combating threats due to the increasing drought occurrence, forest fires and illegal deforestation is needed.
Therefore, the importance of harmonized monitoring and strategy for land management contributes to solutions for pressing environmental issues and works towards a sustainable future. Region-wide continuous monitoring fosters further innovation in agriculture, forestry and rural areas enhancing the viability and competitiveness of all types of agriculture, and promoting innovative farm technologies and sustainable forest management and among others help to mitigate the impacts of natural disasters and climate change consequences (especially droughts and crop failure).
Most of the Danube or Black Sea regional initiatives target ‘development of critical support systems’ or ‘smart, integrated observing systems’, so current Earth Observation (EO) capacity and derived spatial datasets might represent a major contribution here. A large number of satellite sensors, acquired in different spatial, spectral and temporal resolutions, are currently available for operational support, however, the use of satellite EO within these regional level activities remains still relatively low due to several factors, including requirements for tailored processing of the EO data in each region, requirements for the fusion of a range of diverse and heterogeneous datasets and more structured and integrated results.
The general approach of the project is based on the implementation of a regional web-based application platform providing demonstration and access to value-added services, processing tools and EO & non-EO data sources relevant for agricultural and forestry monitoring. The platform is designed to cover various aspects and with the aim to respond to a wide range of identified user requirements:

- Support the uptake of EO data and information use by the stakeholders involved in agriculture/forestry in the Danube Catchment and greater Black Sea basin
- Endeavour to tie on many other activities aiming at fully exploiting the potential of big data services for EO as well as at reaching towards other domains by promoting the cross-exploitation with non-EO datasets in an efficient computing environment
- To meet both technical and information content requirements and to ensure that BSADRI is correctly aligned and interoperable with them
- Respond to performance levels requirements (as long-term monitoring, country-wide extent - large area coverage, but still maintaining scalable solution addressing spatial and thematic variability)
- Stimulate both data-driven services and use of value-added - synthetic thematic information
- Provide an intuitive environment and descriptive demonstration of EO based services
- To fit needs and requirements of various types of users as international entities, regional organizations, national authorities but also the private sector, NGO organizations or even single farms

For this purpose, the regional platform is composed of interconnected elements covering various functions to target the needs of the above-mentioned bases such as EO data processing tools, value-added services, other services as Copernicus or cloud-based tools & applications or additional information services (EO, Spatial, Natural Resources, Agriculture and Forestry)
For better illustration selected use cases are presented covering various agricultural and forestry domains and involve implementation of value-added services addressed by the list of applications in the domain of Natural Resource Management in Agriculture and Forestry. The services will consume EO products generated by the pre-processing segment and apply tailored processing and analysis through dedicated processing chains. A number of processing routines including machine learning classification, temporal profile analysis, data mining, spatial-temporal feature analysis, data analytics and fusion will be integrated on the platform as scalable capabilities for Big Data processing. The use cases thematically focus on topics related to land balance (Georgia), precision farming, operational monitoring for CAP support and forest management to facilitate forest inventory.

Oil spills on the sea cause dark patches in C- and X-band Synthetic Aperture Radar (SAR) imagery, because they damping small-scale surface waves that are responsible for the radar backscattering. The wave damping, hence the radar contrast, is less pronounced at L-band, except for thick parts of the oil spills [1]. On global scales, the main source of marine oil pollution is operational ship traffic, though static sources such as oil seeps or oilrigs may also contribute. There was, however, no static oil-pollution source reported for the Bulgarian Black Sea coast.

On 1 July 1921, the US cargo steamship SS Mopang sank off Sozopol on the Bulgarian Black Sea coast, close to the city of Burgas, after hitting a sea mine that was left from World War I. After the ship wreck had been lying at a depth of about 30 m for almost one century, in August 2018, local newspapers reported an oil leakage that was observed after a period of strong winds and currents in that area. Heavy fuel, of which the SS Mopang had originally loaded 650 tons, was leaking out of the wreck’s tanks.

We used 16 years of spaceborne SAR imagery, acquired between 2006 and 2021 by Sentinel-1A/B SAR-C, ALOS-1/2 PALSAR-1/2, and Envisat ASAR, to investigate for how long, and under which conditions, heavy fuel was leaking out of the wreck. The oil spill of the SS Mopang was visible on more than 100 SAR images; however, we could not find it on every SAR image acquired in the area, which implied that either the leakage of heavy fuel occurred only sporadically, or a continuous leakage of oil could not be seen on every SAR image.

Further analyses revealed that oil spills, which could be attributed to the Mopang wreck, were detected only in the warmer season, from May to November, when the temperature of the bottom water layer exceeded 10 °C. Apparently, under these conditions the heavy fuel’s viscosity was low enough to allow for its leakage and subsequent rise to the sea surface. In contrast, during colder months, when the bottom temperature was below 10 °C, the viscosity of the heavy fuel was too high and hence, we did not find any manifestation of the Mopang spill on SAR imagery acquired from December to April.

Similar patterns of the environmental conditions were found for all cases, in which we observed a Mopang spill on SAR imagery: periods of high winds, resulting in high sea state, were always preceding each oil spill detection, which implies that some mechanical action on the wreck’s hull was needed to initiate further leakage. In addition, we always found a decrease in significant wave height, wind and current conditions shortly before the SAR image acquisition. We therefore hypothesize that a period of increased mechanical stress on the wreck’s hull was needed to force the leakage of heavy fuel, but that calmer conditions at the time of the SAR image acquisition were needed, likely to prevent the leaked heavy fuel from being mixed with sea water without forming a coherent spill at the sea surface that manifests on the SAR imagery.

In cases when the Mopang spill also showed up on L-band SAR imagery, the leakage of heavy fuel must have been strong enough to form a thick oil spill that caused a detectable contrast on L-band SAR imagery. Estimates of the spill’s thickness, however, were not possible.

The understanding of the marine and coastal bio-optical processes has important benefits for society especially in the areas with high human marine activities, as well as for environmental protection. The Black Sea receives drainage from almost one-third of continental Europe (five times its own surface) that includes significant portions of 17 countries, 13 capital cities, and some 160 million people. However, biogeochemical variables are generally derived from data acquired in the visible range of the spectrum, and thus they are hampered by the presence of clouds, a significant issue in a basin that has a 40-50% of average cloud coverage.

At some especially strong river discharges (such as the Danube plume), there is a strong correlation between Sea Surface Salinity (SSS) and the Colored Detrital Organic Matter (CDM) to the point that one can be considered as a proxy for the other, at least over the area most influenced by the river discharge. This connection is caused by the capability of SSS to track the proportion of freshwater contributed by the river. As far as the proportion of sedimentary material contributed by the river is approximately constant, some biogeochemical variables associated with primary productivity will strongly correlate with the amount of river water being mixed in the sea basin and thus with SSS.
In the framework of the ESA EO4SIBS (Earth Observation data for Science and Innovation in the Black Sea) project, we used this connection to derive a new CDM product in the Black Sea generated from SMOS SSS measurements. Despite its limited resolution (we use L4 SMOS SSS with spatial resolution grid of 0.05x0.05º), this new SMOS-derived product provides CDM information over all-weather conditions. Here we present the methods and the assessment of the performance of this new experimental product in comparison with the available CDM products in the Black Sea.

The monitoring of the Sea Surface Salinity (SSS) in semi-enclosed seas has a significant impact in the study of climate change. In those basins the oceanographic processes occur at shorter temporal scales than in the open ocean, and therefore, trends and anomalies can be detected before. In the Black Sea, river run-off and precipitation exceed evaporation making the typical salinity values much lower than in the global ocean (17-18 psu versus 32-38 psu). Moreover, unlike other large estuarine basins, the Black Sea is a deep basin (maximum depth of ~2200 m) with a large North-Western shelf. A distinct vertical layering is created between the surface waters in the upper 100m and the deep graphic conditions maintained by strong stratification resulting from river runoffs flowing at the surface and the entrance of saline Bosporus waters at depths.

The geophysical characteristics of the basin hinder the satellite SSS acquisition: 1) the dielectric models have some limitations in this range of low salinity values; 2) the strong stratification limits the use of in situ data to calibrate or validate the satellite acquisitions; 3) the sensitivity to potential trends requires a robust stable time series of measurements. Besides there are some already known acquisition issues in the region: 1) Strong land sea contamination; 2) Strong contamination from very close Radio Frequency Interference sources. In the framework of the ESA regional initiative project An Earth Observation Data for Science and Innovation in the Black Sea (EO4SIBS), we have developed new algorithms to deal with these issues. Here we present the enhanced methods used in the generation of the first regional SSS products in the Black Sea and the quality assessment of their performance. The products consists of: a) Daily level 2 maps at 0.25ºx0.25º with an accuracy of 1.5 psu; b) 9-day level 3 maps at 0.25ºx0.25º with an accuracy of 0.5 psu, and; c) daily level 4 at 0.05ºx0.0505º with an accuracy of 0.4 psu.

Many studies have proven the value of machine learning and remote sensing for crop yield forecasting. However, in years of severe droughts, such forecasts get unreliable. The goal of this study was to improve the reliability of the crop yield forecasts, particularly in drought years. For this purpose, the Pannonian Basin in southeastern Europe was chosen as the study area, as droughts have heavily affected agricultural production there in the last decades.

Wheat and maize yields were forecasted using a random forest model on various explanatory datasets based on Earth Observation, in situ measurements, and meteorological reanalysis and forecast data. Two drought indices, Evaporative Stress Index and Standardized Precipitation-Evapotranspiration Index, were used to include information about the occurrence of droughts. The predictions were established monthly for four lead times to harvest. The first prediction was made three months before harvest and the last one at the moment of the harvest. The results were cross-validated using three-year intervals as testing sets. Years of severe droughts, 2003, 2007, 2012, and 2015, were additionally analysed.

The validation showed that good predictions could be made from around two months before harvest. This was reflected in correlations of predicted and measured crop yields around 0.6. Forecasts with longer lead times than that led to significantly worse predictions. In years of severe droughts, the results were ambiguous. The wheat yield forecast model underestimates the crop losses, which led to a bad performance even shortly before harvest (correlations lower than 0.4). The maize predictions, on the other hand, showed good performances in drought years. The model underestimated crop yield losses only slightly, and the validation showed a good performance with correlations around 0.5 in the month before the harvest. Overall, a slight underestimation of drought impacts on crop losses remained for both crops at all lead times. Despite that, the results, especially for maize yields, have a considerable potential to predict crop yields reliably before the harvest and thus contribute to reducing socio-economic impacts of crop yield losses in the Pannonian Basin.