|
Download README.md from UNICORNproject/Lava-platform_example-data: direct link, hf CLI and curl.
- Browser
- Download file 4.19 kB
-
https://huggingface.co/datasets/UNICORNproject/Lava-platform_example-data/resolve/main/README.md
- Command line
-
hf download hf://datasets/UNICORNproject/Lava-platform_example-data/README.md
-
curl -L -o README.md https://huggingface.co/datasets/UNICORNproject/Lava-platform_example-data/resolve/main/README.md
4.19 kB
| # UNICORN Project - Lava emergency management platform example data | |
| Partner: MITIGA, INGV-OE | |
| Project: UNICORN | |
| ## Overview | |
| The lava tool developed in the Unicorn project represents a coherent and easy-to-access platform for emergency managers and other stakeholders that integrates: | |
| 1. lava flow detection and monitoring through satellite imagery | |
| 2. lava flow path forecasting and | |
| 3. operational loss forecasting. | |
| The platform is built upon information sourced directly from volcano observatories and has therefore been co-developed with INGV Osservatorio Etneo (INGV-OE), | |
| utilizing Etna volcano in Sicily as a use case. | |
| This dataset represents exemplary data produced by the lava platform for the June 2026 lava flow of Etna volcano. The individual files are explained below. | |
| Note: the outputs included in this dataset are illustrative samples intended to showcase the system's capabilities, and do not represent final production outputs. | |
| ## Explanation of individual files | |
| ### example-report_June2026-eruption.pdf | |
| This is the report that end-users signed up for the lava service receive upon triggering of the tool by INGV. It contains hazard maps, satellite imagery and impact | |
| forecasts for the ongoing eruption. | |
| ### example-hazardmap_medium-scenario_June2026-eruption.tif | |
| This is the data behind the hazard map for the medium scenario shown in the pdf report in a geotiff format. Note that end-users are not receiving these data directly; | |
| this is only for exemplary purposes. | |
| ### example-simulation_medium-scenario_June2026-eruption | |
| This is one example simulation file created by the system for the medium scenario. Note that end-users are not receiving these data directly; this is only for exemplary | |
| purposes. | |
| ## Methodology summary | |
| While the lava platform is built and maintained by Mitiga, its operational use depends on the local volcano observatory (the “operator”). The operator is distinguished from | |
| end-users, who only receive the final pdf report once the operator has launched the service. | |
| Once the observatory detects a new lava flow, it can trigger the lava service through an interactive website. The website asks for a number of parameters, some of which are | |
| mandatory (vent location and uncertainty, eruption start date and duration), and some are optional (effusion rate, final volume). The operator also has to indicate whether | |
| the eruption has ended or is still ongoing. | |
| The service then runs on Mitiga computing systems following the below process: | |
| 1. Create eruption scenarios consisting each of a vent location and a flow volume. Vent locations are placed along a grid considering the vent location uncertainty. The flow volume is calculated: | |
| - For finished eruptions: the flow volume is either provided or based on eruption duration and effusion rate. | |
| - For ongoing eruptions: three future scenarios are considered, assuming a further duration of 2 hours, 1 day or 1 month. These durations are then combined with the effusion rate to calculate the flow volume. | |
| - If the effusion rate is not provided by the user, default values for Etna volcano are stored on Mitiga systems. | |
| 2. For each scenario, lava flow simulations are run using the MrLavaLoba simulator, which was tuned to Etna eruptions. | |
| 3. The individual simulation footprints are combined into hazard maps. | |
| 4. The hazard maps are combined with building locations from Open Street Map to estimate possible impact of the eruption. | |
| 5. Pre- and posteruption satellite images are created using Sentinel-2 and MODIS data. | |
| 6. A pdf report is created and send by email to the operator and the end-user | |
| The overall time from triggering the service to receiving the email is usually less than 30 minutes. | |
| ### Application to the June 2026 eruption | |
| The system was run for the 26th June 2026 eruption to demonstrate capabilities. The flow was active for around 8 days. The following parameters were used: | |
| - Vent location 37.75452 / 15.005 with an uncertainty of 100m | |
| - Duration: set to 168 hours (=7 days), so that the “medium scenario” by the platform should correspond to the final real lava flow placement | |
| - Effusion rate: 0.05 106 m3/d | |