Contribution to the assessment of European River Basin Management Plans
Yüklə 225,62 Kb.
|
- Bu səhifədəki naviqasiya:
- 3.2.2 Spatial and temporal resolution
- 3.2.3 Input
- 3.2.4 Preliminary assessment
- 3.2.5 Previous applications
- 3.2.8 State of play
3.2 Chemicals3.2.1 MethodologyChemicals under the WFD include: - priority substances, of EU level interest; - river basin-specific pollutants. The CIS Guidance document n. 28 “Technical Guidance on the Preparation of an Inventory of Emissions, Discharges and Losses of Priority and Priority Hazardous Substances”8 provides indications for member states on the preparation of emission inventories. The JRC Pressure Indicators focus on priority substances as discussed below, and consist of maps of Predicted environmental concentrations (PECs) of priority substances (one per substance). The proposed approach follows the “riverine load” concept of Guidance document n. 28, and consists of two modelling steps:
The first step requires comparing observed loads at a number of sample points with appropriate indicators of human activity in the catchment of each sample point (“emission pattern”). We approximate the relationship between the emission pattern and the observed loads as linear, and the slope of the relationship represents the “emission factor”. In the simple case of a conservative chemical emitted homogeneously by households, for instance, catchment population could be used to represent potential emissions. In the example of population, this would be read as the emission per capita in the catchment (Figure ). Human activity in the catchment Observed load
The model can be adjusted to account for the spatial distribution of emissions within the catchment and the time of travel from each location in the catchment, to the monitoring point. We take into account chemical dissipation through an exponential decay depending on the travel time of contaminants in the catchment, and we allow emission patterns to follow different sources (e.g. agriculture, livestock, etc.) or a combination of sources. Details on the method are provided in Pistocchi and Loos, 2009, for conservative chemicals (specifically, PFOA and PFOS), and Pistocchi et al., 2012, and chapter 19 of Pistocchi, 2014, for chemicals subject to decay. The approach is conceptually similar to GREEN as discussed above. The main difference is that, while GREEN assumes an inventory of nutrient emissions as an input to the model, for chemicals we assume emission to be proportional to a pattern in space (e.g. population) and we infer the proportionality constant (or emission factor) based on monitored concentrations. Examples of typical emission patterns considered in practice are listed in Table .
Table – typical emission patterns used for chemical modelling. Once the emission factor is known, it is possible to use the emission pattern to map emissions and loads explicitly, and concentration follows from dividing by a representative discharge as pointed out for GREEN. For instance, of emissions per capita are known, we can map loads based on population, and concentrations dividing loads by water discharge. The calculation of loads is obtained through a simple steady state, plug-flow transport model (Pistocchi, 2014, ch. 19) enabling the routing of emissions along the stream network, including the effect of lakes and reservoirs. Typically, emission factors are characterized from a number of measurements, and the effect of point emissions is often hidden in the variability of the data. Therefore, it can be assumed that the emission factors reflect the “diffuse” component only of these emissions (see e.g. Pistocchi and Loos, 2009). When significant point emissions are known, they can be included in the calculation of concentrations in addition to diffuse emissions. This approach must be regarded as an exploratory tool allowing an independent evaluation of a set of available chemical concentration/load measurements. The calculation of concentrations does not reflect a number of chemical fate and transport processes (e.g. adsorption to particulate matter or sediment. Therefore, it by no means aims at emulating the assessment of individual member states based on more detailed investigations. At the same time, the approach may be useful to identify areas potentially affected by a certain concentration of a chemical, not necessarily taken into account during the monitoring. The calculation will be extended to priority substances identified by Directive 2013/39/EU, including the list of 45 substances in Table . Moreover, the Commission Implementing Decision (EU) 2015/495 of 20 March 2015 establishes a “watch list” of additional substances (Table ). These represent chemicals of potential concern at the European scale and will be included in the calculation as far as possible (depending on monitoring data availability).
Table – Priority substances (Dir. 2013/39/EU) EU-number: European Inventory of Existing Commercial Substances (EINECS) or European List of Notified Chemical Substances (ELINCS).
Table – chemical substances in the “watch list”. 3.2.2 Spatial and temporal resolutionIndicators can be computed with a resolution limited by the available maps of emission patterns. However, the results should be presented with a minimum level of aggregation, as the analysis does not capture local details (contmainated sites, individual emissions) that may significantly affect the concentrations. In practice, the indicators will be computed for sub-catchments as in the GREEN model. While in the case of nutrients we refer to a specific year in which emissions (point and diffuse) are known, in the case of chemicals the model reflects the load at the time of monitoring. Monitoring data can be quite scattered in time, and in order to assemble a representative set of sample points it may be necessary to consider together points sampled in different years and in different days of the year. This can be done insofar as we can assume that emissions are varying gradually from year to year, there are no drastic discontinuities in emissions, and emissions can be approximated to constant in time along the year. For instance, pesticides with a strong seasonality in application are a clear case of deviation from these assumptions, just like emissions from industrial activities that may be discontinued at a certain time. These considerations suggest the inherent uncertainty in the assessment of chemicals compared to other variables. Whenever sample points used to estimate emission factors are not referred to the same year, the results must be interpreted as a representative yearly average concentration for the period covered by the samples. Under the assumption that loads are constant throughout the year, a concentration duration curve can be computed based on the flow duration curve.
3.2.6 StrengthsSimplicity, transparency; by construction, good correspondence with observed concentrations. 3.2.7 WeaknessesDepends on data available; difficult to obtain robust estimates if data base is not sufficient. The method only addressed loads as measured (e.g. dissolved phase only, if loads are dissolved concentration times water discharge). Generalization to sediment-borne contaminants possible, but suffering from lack of reliable sediment flow estimates. The method allows estimating loads when these are constant: strongly seasonal or unsteady emissions are not addressed correctly. Emission factors derived from observations do not reflect point emissions and specific cases (e.g. contaminated sites or accidents). In all those cases, the model serves as a screening tool valid only for orders-of-magnitude assessment. The accuracy of the model can be evaluated for each of the substances considered and the uncertainty of the indicators will be characterized when prepairng the respective maps, thus allowing to discard those chemicals for which the indicator cannot be considered reliable due to the above limitations. 3.2.8 State of playCalculations for the 45 priority substances and 10 watch-list substances are underway. The sample point data used on purpose are those reported in IPChem. For some substances, data are rather limited and/or spanning a long sampling period in which the assumption of constant loads cannot be accepted. Additional data, including data reported under the WFD by member states, will be tested insofar as available. Monitoring obligations for the 12 most recently added priority substances are weaker, hence data availability may be limiting. Yüklə 225,62 Kb. Dostları ilə paylaş:
|
