From Geostandards

Contents 1 Application of SWE in environmental monitoring and disaster management 1.1 OSIRIS 1.1.1 Forest fire fighting 1.1.2 Air pollution 1.1.3 Water pollution 1.1.4 Fire detection in industrial buildings 1.1.5 Summary

Application of SWE in environmental monitoring and disaster management

The integration of sensor data into applications for environmental monitoring and for disaster management promises significant benefits. Below the OSIRIS project will be introduced as an example for an activity focusing on the application of SWE for buildong environmental monitoring and for disaster management systems.

OSIRIS

The acronym OSIRIS stands for “Open Architecture for Smart and Interoperable Networks in Risk Management Based on In-situ Sensors”. OSIRIS was an EC funded integrated project that was performed within the Sixth Framework Programme (FP6) between September 2006 and April 2009. The OSIRIS consortium comprised 13 partners from different European countries which brought expertise from end user, research and commercial perspectives.

OSIRIS focused on the integration of in-situ sensors into systems for risk monitoring as well as crisis management. An important criterion for the system was its applicability independent of specific domains so that the resulting solution can be applied also to use cases that have not been addressed by OSIRIS. The technological foundation for the sensor integration was based on the OGC SWE architecture. However, OSIRIS has at the same time also contributed to the enhancement of the OGC SWE standards baseline and to the development of open source SWE implementations.

A comprehensive demonstration of the aims and especially the OSIRIS use cases can be found here: [1]

The OSIRIS use cases addressed for different domains:

• forest fire fighting
• air pollution
• water pollution
• fire detection in industrial buildings

Subsequently these four different use cases will be explained further.

Forest fire fighting

Within this scenario the usage of SWE components for improving the detection and fighting of forest fires was tested. It was realized in cooperation with the according authorities in the southern part of France, an area that is often seriously affected by forest fires.

In detail the following sensors were deployed:

• an airborne sensor platform for collecting overview image data; this data is used for detecting forest fires in an early stage and in case of a forest fire for closely monitoring its development
• an ultra-wideband system for localizing fire men as well as emergency vehicles within the forest fire site
• surveillance cameras for monitoring remote areas (e.g. for observing evacuated regions)
• a weather station for measuring meteorological parameters that help predicting the spread of forest fires

SWE components were used in order to achieve the following aims:

• SPS instances for controlling the airborne platform (setting its flight path) as well as the surveillance cameras (setting pan, tilt and zoom)
• SOS instances for providing weather data, positions of fire men and video data coming from the surveillance cameras

In addition, within this scenario WCS/WMS instances were used for accessing the overview image data delivered by the aerial platform.

Air pollution

Within this scenario sensors were applied in the domain of air quality monitoring. The practical realization took place in the city of Valladolid in Spain. On the one hand air quality measurements were used for a continuous monitoring that made it possible to calculate the air quality distribution within the city. On the other hand a crisis scenario of a chemical accident was treated. In this case sensors were deployed for assessing the exact situation and for calculating predictions in order to be able to evacuate endangered parts of the city.

In detail the following sensors were used:

• mobile sensors (mounted on public busses) that measure a variety of air quality data
• air quality measurement stations
• an unmanned aerial vehicle (UAV) that can be deployed in emergency scenarios for measuring the three dimensional pollutant distribution within pollutant clouds
• a weather station for delivering meteorological data needed by simulation models

Like in the other scenarios also in this case, SWE components were used:

• SOS instances for accessing air pollution and weather data
• SAS instance for dispatching alerts in case of critical air quality values

Water pollution

The water pollution scenario was realized in the Tuscany region in Italy. This scenario comprised two sub scenarios:

• Monitoring of arsenic concentration: Due to geological reasons a high natural concentration of arsenic within the ground water of the test area can be observed. However due to legal restrictions, the arsenic concentration in the ground water may not exceed a certain threshold. Thus SWE components are used for realising an according monitoring application.
• Hydrocarbon spills: Ground water which is used for delivering drinking water is very sensitive to any kind of pollution. In this case the risk of fuel truck accidents causing hydrocarbon spills is addressed. SWE services are used for exchanging the relevant measurement data and for providing input to according simulation models.

In detail sensors measuring the following phenomena have been deployed:

• arsenic concentration
• speed of water flow
• water temperature
• hydrocarbon concentration
• weather

The sensor integration was also in this case achieved by deploying according SWE components. The developed system relied on the following types of SWE services:

• SOS instances for accessing the concentration of arsenic and hydrocarbons, weather data and hydrological measurements
• SAS instances for dispatching alerts in case of critical concentrations of arsenic or hydrocarbons
• SPS instances for changing the sampling rate of arsenic and hydrocarbon sensors

Fire detection in industrial buildings

This scenario aimed at more reliable fire detection within industrial buildings with a special focus on the avoidance of false alarms. It was implemented within a fire fighting training area operated by the fire department of the city of Aachen. Basically this scenario was build upon three different types of sensors:

• smoke detectors
• temperature sensors
• surveillance cameras

In order to increase the reliability of fire detection these sensors were combined with each other so that more complex alert criteria could be applied. A simple example for such criteria concerns the combination of smoke and temperature sensors: Smokers usually do not cause any significant temperature rise within rooms. However, they can cause alerts generated by smoke detectors. On hot summer days certain rooms within a building may become very warm. In this case temperature sensors could cause an alert whereas smoke detectors will remain quiet. Thus, a criterion for a “real alarm could be that smoke detectors and temperature sensors have to deliver individual alerts.

To realise this scenario the following SWE services have been deployed:

• SAS instance for dispatching alerts if a fire is detected; this includes the definition of alert criteria and the according filtering of incoming data
• SOS instances for accessing the measurement data of smoke detectors and temperature sensors
• SPS instances for adapting sensor profiles: the realisation of this scenario relies on two different operational modes: normally sensors perform their measurements in longer time intervals; however, in case one sensor detects critical values, the sampling rate of the sensors is increased in order to detect earlier if also other sensors measure critical values

Summary

In summary OSIRIS has shown that SWE can be applied to a large variety of use cases. It offers a well defined and flexible architecture that allows the easy integration of sensors and their data into client applications without having to deal with specific sensor characteristics.

A topic that will be relevant for future research will be the integration of sensors into the different SWE components which could be further facilitated. However, concepts like the transactional profile of the SOS provide already now first solutions addressing this issue.

← previous

5 Sensor Web Enablement

next →