Ewa Kmiecik

The Impact of Human Errors on the Estimation of Uncertainty of Measurements in Water Monitoring

The main purpose of a physical act of measurement is to enable decisions to be made. In case of an assessment of the chemical status of groundwater body, or assessment of suitability of water for drinking purposes, or possibility of discharges sewage into surface waters, the measurements of physicochemical parameters of water are an indispensable first step. The reliability of the mentioned above decisions heavily depends on knowing the uncertainty of the measurement results. If the uncertainty of measurements is underestimated, for example because the human errors are not taken into account, then erroneous decisions can be made that can have in some cases substantial financial consequences. In this work there are presented examples of human error identification and estimation in measurements made during water monitoring on the base of duplicate control samples (empirical approach) with the use of control charts method.

Use of neural networks for assessment of adverse impact duration of solid waste disposal facilities on the aquatic environment

Routine monitoring and long-term studies conducted in 19-years’ hydrologic cycle in the Upper Silesia Coal Basin (USCB), Poland, show extensive release to ground and surface waters of contaminant loads from mining waste. For simulation of the time-dependent changes of Acid Rock Drainage (ARD) generation expressed as sulfate formation due to oxidation of ferrous sulfides occurring in solid phase of mining waste, models of supervised neural networks were used. It was found that with use of such a model, the time span in which the concentration of a contaminant will reach the permissible level or the process of its release will terminate could be evaluated with a precision sufficient for practical purposes (the relative error did not exceed 1%). The results of simulation of temporal and spatial contaminant concentration changes will be utilized as a basis for assessment of an extent of the environmental deterioration dependent on the duration of a waste disposal in the site. These analyses enable to obtain reliable models describing time-dependent changes of water quality in the vicinity of long-term contamination sources, which seems to be their most essential merit The models allow also to evaluate the duration of the adverse impact of a facility on the aquatic environment and to reduce the expenses on the monitoring through the reduction of a number of samples and analyses.

Precision of selected trace elements in groundwater: examples from regional groundwater quality monitoring in the Upper Vistula River Basin (Poland)

Regional Groundwater Quality Monitoring (GQM) in the Upper Vistula River Basin (UVRB) is a pilot scale element of an Integrated Water Management System in SE Poland (about 50,000 km2). The system is currently in the implementation phase. The objectives of the regional GQM network are: (1) to provide data for quality control in regionally important groundwater basins, particularly Major Groundwater Basins (MGWB) in this area; (2) to establish the groundwater vulnerability to large scale diffuse sources of pollution; (3) to perform the prognosis and to identify trends of long time groundwater quality change; (4) to establish the extent of natural and anthropogenic processes impact on groundwater quality. During the sampling of the groundwater monitoring network the Quality Assurance/Quality Control (QA/QC) program for hydrogeochemical measurements was implemented in parallel. The QA/QC program included the collection (using the same equipment as for normal groundwater samples) and analysis (to the same extent as in the normal samples) of additional special samples. The duplicate samples taken from randomly selected GQM sites were used to assess the precision of hydrogeochemical measurements based upon the analysis of variance (ANOVA). The QA/QC program as applied to the GQM in the Upper Vistula River Basin is illustrated in this paper in the examples of two selected trace elements: fluoride (F) and zinc (Zn). Technical variance ((sigma) tech2) values calculated by means of the classical and robust ANOVA are lower than 20% for the discussed trace elements. This confirms the satisfactory precision of hydrogeochemical measurements.

Assessment of chemical quality of ground waters exemplified in Major Groundwater Basin 332

The chemical state of groundwater in the Major Groundwater Basin (MGWB) 332 was assessed on the grounds of hydrogeochemical monitoring conducted in 2001-2003 in the network that comprised 37 monitoring wells (points). On the basis of concentration data of organic and inorganic chemical constituents, the chemical quality of water in the monitoring wells was evaluated. Aggregated data were used for an assessment of the chemical state of the whole basin. The evaluation of the water quality in the monitoring wells was conducted in accordance with the Directive of the Minister of Environment (RMS, 2004) by the comparison of concentrations of analyzed chemical constituents with limit values in the quality classes I-V. In general, the water quality fulfilled the criteria of III class, occasionally of IV and V.class. Water chemical state in MGWB 332 was evaluated as good, following the criteria specified in the EU Water Framework Directive 2000/60/EC and the draft EU Directive on the protection of groundwater against pollution (COM 2003), based on the mean values of chemical indicators for the whole basin. The recommended method of the chemical state assessment on the basis of mean concentrations in all the monitoring points caused vanishing the zones of unsatisfactory quality (class IV) in the averaged backgrounds. In practice, this will result in desisting from any action aiming to improvement of water quality in these zones. The results of this study show the need of reporting chemical state of groundwater quality also in the specific monitoring points, and not just in the averaged hydrogeologic units or subunits.