Application of modeling for optimal localization of environmental monitoring sensors

The control of environmental phenomena is based on the existence of quantifiable and measurable parameters representative of the phenomena themselves. In the case of studies concerning contaminated aquifers, after an initial step of characterization it is necessary to implement groundwater hydrogeochemical monitoring phases. In the field of groundwater pollution long-term monitoring realized by means of samples withdrawal and subsequent analysis is labor-intensive, time-consuming and dependent on remote laboratories for sample analyses. Therefore, a high portion of the costs connected with long-term monitoring is associated with the sampling procedures, samples transportation and laboratory analyses. In this context, the utilization of sensors that provide real time measurements would significantly reduce site characterization times and costs and provide more complete and continuous data-sets. However, within the use of sensors in the environmental monitoring field, one of the most relevant problems is linked, in addition to the requisites of the sensor itself, to the correct individuation of the sensor's position in relation to the phenomenon to monitor. In the present paper are reported some case studies in which the precision of data measured by sensors may be invalidated by an incorrect localization of the monitoring points. In this field it proves to be fundamental to implement models able to simulate the real conditions of the system to be subjected to monitoring. The above mentioned models include both deterministic geostatistical techniques aimed at reconstructing the present situation, and also numerical simulations finalized at making prediction on spatial and temporal trends of the studied phenomena. On the basis of these simulations it is possible to optimize the positioning of the sensor in such a way as to guarantee the efficiency of data detection.