Task 5.1 Design and Development of the Back-end System (Leading: FRC, Participating: CYRIC) This task will initially set-up and test the infrastructure that will be used to host the platform’s back-end component. A distributed file system will be deployed to accommodate the Big Data generated by the various data sources (sensors deployed in the field, UAVs, satellite images), while NoSQL and distributed processing systems will be deployed to achieve high scalability and performance. Furthermore, a database system will be designed and developed based on the data specifications of WP3, which will be used to model, store, and maintain all different types of data required by Green-HIT. The task will also design and implement a web Content Management System (CMS), which will expose an intuitive and highly responsive interface for managing the data stored in the platform’s database. The CMS will provide all necessary functionalities and tools for viewing, inserting, updating, and deleting relevant data on demand. Finally, Task 5.1 will develop the required communication interface (API) between the back-end component and the front-end software components (Web dashboard and Smartphone applications), which will be developed in subsequent tasks (i.e., Tasks 5.2 and 5.3, respectively). The API will handle incoming data management requests (e.g., insertion and retrieval of data). To ensure a high degree of portability and interoperability between the various components of the platform, the API will be RESTFul and responses will be provided in JSON format. All design and development activities of this task will be undertaken by FRC, while feedback will be provided by CYRIC.

Task 5.2 is responsible for designing and developing the platform’s main front-end portal, namely a Web-based dashboard that offers facilities for forest management and monitoring. A tentative list of functionalities that will be supported by this dashboard (to be finalized upon commencing the project during requirements analysis in WP3) is listed below: ▪ Interactive visualizations of raw spatio-temporal data collected, for example, through in-field sensors, UAVs, satellites and/or the habitat mapping / inventory tool of the smartphone application. ▪ Interactive tools for better interpreting and acting upon the raw data collected, which will leverage the Intelligence Modules developed in WP6. ▪ Facilities for setting up the platform’s rules of engagement, e.g., periodic, event-based, and/or on-demand scheduling of UAV trips, two-way verification mechanism for incidents that potentially require human intervention, etc. ▪ Simulation-based estimations/predictions of the progression of a fire (trajectory, time-based safe zones, etc.), in case of fire emergencies, based on current conditions (e.g., weather and other environmental indicators, terrain attributes, etc.). The design and development activities of this task will follow the UI designs and UX principles defined in WP3. An iterative and incremental approach will be adopted, with rapid prototypes being delivered frequently to allow for early feedback from all stakeholders involved. Any feedback collected will likely trigger updates to the initial requirements and thus the final software outcomes. The design and development activities of this task will be handled by FRC, while all other partners of the Consortium will provide feedback.

Task 5.3 will design and develop the Green-HIT multi-purpose smartphone applications across all major mobile platforms, namely Android and iOS, which will allow for accessing Green-HIT’s supported functionalities while on the move. In addition to the functionalities discussed for the Web-based dashboard (in Task 5.2), the applications will also support in-the-field forest mapping and inventory management. In particular, the apps will offer highly intuitive, as well as easy and fast to operate map-based tools for habitat mapping of forests, forest areas, ecosystems, and biodiversity data on forest surroundings as well as inventory tracking and monitoring when on site. The design and development activities of this task will follow the UI designs and UX principles defined in WP3. An iterative and incremental approach will be adopted, with rapid prototypes being delivered frequently, thus allowing for early feedback from all stakeholders involved. Any feedback collected will likely trigger updates to the initial requirements and thus the final smartphone applications. The design and development activities of this task will be handled by FRC, while all other partners of the Consortium will provide feedback.

This task will be responsible to develop the data processing mechanisms by connecting to third-party sources and calculate important indices required by the intelligence modules of WP6. For example, Satellite image download using the Copernicus Open Access Hub of the European Space Agency ESA (https://scihub.copernicus.eu/). Copernicus and supporting missions satellite data provided from ESA and USGS /NASA images such as the legacy series of Landsat will be used. Therefore, it is expected that Sentinel 2A-2B, Landsat 7 ETM+, Landsat 8 OLI, and MODIS products will be used. SPOT and high resolution IKONOS / QuickBird and GeoEye images will be also used if necessary. It is expected that a high -temporal satellite datasets will be created. Considering Landsat-8, Sentinel-2A, and Sentinel-2B together will provide a global median average revisit interval of 2.9 days, and, over a year, a global median minimum revisit interval of 14 min (±1 min) and maximum revisit interval of 7.0 days. Meteorological data from Cyprus Department of Meteorology. Interpolation methods will be applied to meteorological data for the creation of temperature maps. Aspect: a key factor in igniting forest fires since it also plays a key role in the growth of vegetation types species. Elevation: DEM (Digital Elevation Model) used to create the relief of the area (altitude affects the climatic conditions and consequently the fire events.) Slope: Steep slopes lead to a greater rate of fire spread due to the convective ignition and pre-heating. It is also affected by water loss and more efficient convective preheating. Production of vegetation condition maps: Vegetation indices help the understanding of water retention in plants and soil used to predict the moisture content of vegetation, which plays an important role in fire ignition and propagation. ECoE team has already expertise on mapping and creating various vegetation indices using the Google Earth Engine platform. NBR is a useful index for the burned CODEVELOP-GT/0322 – PART B – TECHNICAL ANNEX 17 areas detection. NDVI is a widely used vegetation index computed from the combination of visible and near-infrared light reflected by vegetation, the utility of NDVI index data derived from high temporal resolution NOAA AVHRR imagery for dynamically assessing the potential fire risk. Other indices, such as the dNBR, the RdNBR, and RBR can be also used as satellite data to identify burned areas and burn severity mapping.

This task will experimentally verify the technical soundness of the Green-HIT software prototypes via carefully designed Unit tests. Appropriate testing will be carried out in the laboratory to verify that all recorded requirements are met successfully. Additionally, benchmarks will be used to assess the non-functional requirements of the prototypes, namely in terms of performance, fault-tolerance, and scalability. Any detected deviation from the software quality metrics defined WP3 will result in necessary actions to ensure near optimal conformance with the initial requirements. Overall, verification testing will pave the way for the successful trial of Green-HIT during the pilot operation described in WP7. Verification tests will be carried out primarily by FRC, with the involvement of CYRIC.