Search

The project OFIDIA2 (Operational FIre Danger preventIon plAtform 2), funded by the Interreg Greece-Italy 2014-2020 Programme, proposed a pragmatic approach to improve the operational capacity of the stakeholders to detect and fight forest wildfires. A data analytics system was designed and implemented within the project to manage, transform, and extract knowledge from heterogenous data sources, through forecasting models such as weather, fire danger, and fire behaviour models. The high-resolution weather forecasting network previously developed in OFIDIA1 was enhanced by using a mesoscale configuration of the WRF-ARW model over the Central Mediterranean Sea. A nested domain over the Southern Italy at ~2km horizontal resolution allows getting high-resolution weather forecasts (2x2km) and processing data into fire danger models. Fires, fuel, topography and weather data were collected from several sources and used to run and calibrate fire models (FlamMap and Wildfire Analyst) in Apulia region (Italy). Based on the analyses of recurrent weather conditions leading to large fires, fire metrics’ maps for prevention and fire-fighting activities were produced. Finally, a Decision Support System (DSS) was also developed to provide support for 1) the selection of fire behaviour scenarios by means of mathematical models; and 2) the prevention of emergencies thanks to weather forecast information with fire danger indices at high resolutions.

In the context of XF-ACTORS project, a IoT based monitoring system has been established and tested in the field - for an olive grove in Puglia (Italy) affected by Xf - to measure in near real-time some parameters proxies of trees’ conditions and vulnerability under both abiotic (climate) and biotic factors. The system is based on the TreeTalker (TT) technology, comprising multiparametric sensors to monitor water transport in trees, trunk humidity and diametrical oscillations, spectral characteristics of the leaves and microclimatic parameters (temperature, relative humidity). In particular, the sap flow density can be retrieved according to the Heat Balance Method (Granier 1985) after measuring the temperature of two 20 mm long probes inserted into the stem wood at 10 cm distance along the trunk vertical axis; the probe in the higher position is heated while the lower one provides the reference temperature. The TT system collects and transmits data at hourly time frequency, thanks to a LoRa based wireless connection, to a node managed by another microcontroller (TT-Cloud) serving a few tens of devices in a cluster. The TT-Cloud is in turn connected to the internet via the GPRS network and sends data to a computer server. Here, raw data are subject to ETL procedure that allows data Extraction from the TT-Cloud source, data Transformation by cleaning and converting them into variables with eco-physiological meaning, and finally data Loading to insert them into the target spatio-temporal database, adopting proper storage format/structure for querying and analysis purpose. From here, data can be further elaborated and visualized, e.g. into useful statistics, through a tailored Data Science environment. The preliminary results on sap flow density are here presented as they can give important information about the impact of Xf that is known obstructing xylem vessels, reducing hydraulic conductivity and thus affecting evapotranspiration.

In the context of XF-ACTORS project, a IoT based monitoring system has been established and tested in the field - for an olive grove in Puglia (Italy) affected by Xf - to measure in near real-time some parameters proxies of trees’ conditions and vulnerability under both abiotic (climate) and biotic factors. The system is based on the TreeTalker (TT) technology, comprising multiparametric sensors to monitor water transport in trees, trunk humidity and diametrical oscillations, spectral characteristics of the leaves and microclimatic parameters (temperature, relative humidity). In particular, the sap flow density can be retrieved according to the Heat Balance Method (Granier 1985) after measuring the temperature of two 20 mm long probes inserted into the stem wood at 10 cm distance along the trunk vertical axis; the probe in the higher position is heated while the lower one provides the reference temperature. The TT system collects and transmits data at hourly time frequency, thanks to a LoRa based wireless connection, to a node managed by another microcontroller (TT-Cloud) serving a few tens of devices in a cluster. The TT-Cloud is in turn connected to the internet via the GPRS network and sends data to a computer server. Here, raw data are subject to ETL procedure that allows data Extraction from the TT-Cloud source, data Transformation by cleaning and converting them into variables with eco-physiological meaning, and finally data Loading to insert them into the target spatio-temporal database, adopting proper storage format/structure for querying and analysis purpose. From here, data can be further elaborated and visualized, e.g. into useful statistics, through a tailored Data Science environment. The preliminary results on sap flow density are here presented as they can give important information about the impact of Xf that is known obstructing xylem vessels, reducing hydraulic conductivity and thus affecting evapotranspiration., IT; PPT; monia.santini@cmcc.it