The LaMMa GeoPortal was designed and implemented in order to ingest, visualize, allow searched and dissemination for the wide variety of data sources that LaMMa acquires directly or indirectly. To name a few of them:
- MeteoSat 1st Generation Data
- MeteoSat 2st Generation Data
- MeteoSat 3st Generation Data
- NDVI Data
- Meteo Radar Data
- Real Time data coming from Meteorological Stations spread over Tuscany (rain, pressure, temperature and so on)
- Geophysical parameters coming from the ARW ECM model running twice a day at 3km and 12 km of resolution
- Geophysical parameters coming from the GFS model running twice a day 50 km of resolution
- Geophysical parameters coming from the MED model running twice a day 12 km of resolution
It is particularly important to remark that all this information is managed and visualized taking into account the TIME and ELEVATION dimension (e.g. sliding through the data as time changes is allowed).
Let us now briefly introduce the components used for building the infrastructure we deployed as well as the interactions between them.
Ingestion & Preprocessing
Data is produced or acquired at very short interval (e.g. MeteoSat data is downloaded from EUMETSAT every 15 minutes) and as such we implemented an ingestion system able to run in event-driven fashion ( as soon as new data arrives). We based our development on the GeoBatch Open Source application inside which we created a flow of action for each type of data described above. Each flow is responsible for preprocessing, indexing and publishing for later retrieval of specific type of data. It is worth to point out that some of the actions taken during these flow have been implemented in Python or Octave (e.g. going from raw MeteoSat data to products like Dust or AirMass) in order to allow LaMMa personnel to customize such processing further in the future.
Registry & Cataloguing
We have deployed and slightly customized the Open Source registry GeoNetwork in order to allow users to perform searches within the data ingested into the system for visualization and download purposes. It is worth to point out that the ingestion process is responsible for creating proper metadata (starting from templates that we agreed upon with LaMMa personnel) and ingesting them automatically into GeoNetwork.
We have deployed a cluster of instances of the GeoServer Open Source application in order to support visualization and dissemination through standard OGC protocols of the data acquired with support for time and elevation. As per the metadata, the ingestion flows has been configured in order to automatically ingest data into GeoServer without human intervention.
Front-end & Visualization
We have deployed and customized together with LaMMa personnel the MapStore Open Source WebGis (which is shown in Illustration 49) to act as the front-end for the end users of the infrastructure. It provides the standard capabilities of a WebGIS together with some advanced capabilities like advanced interaction with Mapping Services with time and elevation dimensions. It also provide and integrated, unique interface between search & view capabilities which allow users to perform searches for data and then load some of this data directly on the viewer itself.
It is eventually worth to point out that the entire infrastructure has been developed using Open Source components (the Operating Systems is Linux CentOS and the DBMS is PostGIS). We also tried to emphasize on the possibility to evolve the system independently by LaMMa staff both in terms of processing of new data as well as in terms of customizing the front-end further.