Monitoring tools

Monitoring tools Erftverband
> (.pdf)

Monitoring tools Tudor:

 Introduction

 Passive samplers

 Autosamplers

 River metabolism parameters

 Suspended sediment nets

Introduction


The M3 project has analyzed current monitoring efforts in the 3 partner regions in the light of their pertinence towards pressure identification and quantification in view of the evaluation of successful Programs of Measures (POM). This analysis can be read in a separate report. The main outcomes were that regulators and river basin managers are rather sticking to a threshold motivated monitoring than a monitoring aiming at improving process understanding and source allocation/quantification. The latter are however the prerequisite to initiate successful POM. Monitoring campaigns are needed to confirm the magnitude of a certain pressure, i.e. via the calculation of river loads or the estimation of exposure. Irrespective of their calculation by extra- or interpolation, these estimations carry a bias (accuracy &precision) that has to be quantified in order to assess significant changes. The same applies to modeling approaches: here, the calibration/validation data sets should be used beyond simple goodness of fit measures (Nash-Sutcliffe coefficient and alike) to calculate loads and their bias. Depending on the processes described by the models short-term changes in the river immission situation need to be monitored (f. ex. diel oxygen profiles). These data are rarely available from routine measurements. Since shortcomings in monitoring have essentially been identified in the field of operational and investigative monitoring the demonstration campaigns by the M3 project will focus on the following aspects:

  • Ability to sample exposure (immission situation) in an integrated way
  • Quantify ecosystem metrics (metabolism, transient storage)
  • Serve as water quality model validation dataset
  • Quantify loads of dynamic components
  • Circumvent problems related to limits of detection and spatial representativeness

Every region applies monitoring campaigns according to its own needs although some parallels are planned to compare issues of local pressure differentiation and issues of scale.

To top

Passive samplers



In Luxembourg the assessment of the immission situation for pesticides is limited to those belonging to the priority substances list on 7 locations. The sampling frequency is not adapted to the actual application periods. Knowing the potentially pulsed dynamics of pesticides, the current sampling scheme is hardly fit for a consistent exposure monitoring strategy. Pesticides show a seasonal and intermittent immission dynamic. Their concentrations are likely to vary broadly depending on sources and intensity of use in a catchment. Traditional grab sampling, as performed in the 3 regions is likely to underestimate pesticide exposure in terms of exposure time, level and composition (number and ranking of compounds).

There are two main sources for pesticides:

  • Emission by WWTPs from left-over spillage and cleaning of spraying equipment (if performed at the farm). This is a rather continuous base-flow phenomenon during application periods.
  • Emission by surface-runoff during rain-fall events from treated agricultural fields (or impervious surfaces). The levels of contamination are strongly linked to the infiltration capacity of the soil and the lag between application date and rainfall event. Highest levels are expected in main application periods (March-April, May, October). Sampling should be concentrated in these periods and cover exposure in a representative way. Ideally sampling should be continuous. Passive samplers have been applied to monitor pesticide exposure over longer times.

Monitoring set-up
CRTE has got experience with POCIS passive samplers (EST). They have been used in the DomesticPest project on four campaigns featuring WWTP outlets and rivers. Laboratory tests showed the applicability to a large array of pesticides. POCIS membranes are deployed in canisters with holders for 3 membranes as replicates/or different exposure times). The recovery of the sorbent from the membranes for lab extractions can induce some error. It is therefore preferable to run at least duplicates. A 14 days exposure time is a reasonable practice with regards to POCIS costs, time period covered and membrane permeability. Average concentrations in the river water can be calculated if uptake rates (Rs) for the specific compound are known. These can be determined in the lab and vary according to the polarity and diffusion coefficient of the individual compound. Flow conditions and turbulence in the field influence the uptake rate although the exposure cage mitigates these effects to some extent.. CRTE has investigated field determined Rs on several rivers with different flow velocities and turbulence by taking composite samples with autosamplers in the same period. The determined Rs vary by 20-30% between sites for most substances. Longer exposure induces other problems like the clogging of the membranes which results in slower uptake rates over time. This can be extrapolated by comparing short exposures to longer exposures with substances that occur at stable concentrations under base flow like most pharmaceuticals. The behavior of POCIS under high flow conditions is not known very well but is expected to depend on the ratio between uptake and loss rates. The latter are often 2 magnitudes lower and hence the pesticide mass collected in the high exposure situations is quantitatively conserved during low-concentration periods.

To top

Autosamplers



Characterizing sources of pollutants in a catchment is an ambitious endeavor, which is impossible to realize with infrequent grab sampling. Moreover, loads that are calculated from grab samplings via extrapolation with discharge substantially underestimate the real pollutant load masses passing a balancing point. Autosamplers that are triggered by pressure sondes and turbidity probes are an ideal tool to monitor substance flows during flood waves at discrete locations in a catchment. The results from these campaigns open a whole set of opportunities to evaluate the data: flood waves can be compared with respect to their mean event concentrations and even discrete components within flood waves can be discerned via pulse fitting. The application of this monitoring technique is ideal for measuring the impact of combined sewer and storm overflows but also for phosphorus, nitrate and pesticide mobilizations in rural settings. CRTE has been running campaigns in the past with this setup for CSO impacts in the southern Alzette catchment in 2005. In the M3 project, the team will test the autosamplers in a smaller and hence more reactive catchment with urban impact as well as in a rural setting to test for the ability to catch pesticide surface runoff as well as phosphorus mobilization by erosion.

To top

River metabolism parameters



The metrics employed within the WFD monitoring schemes are suffering from poor coherence. This is especially true for nutrient status and ecological indicators. Ecosystem metabolism measures such as Gross Primary Production (GPP) and Ecosystem Respiration (ER) could bridge this gap as they reflect ecosystem properties on time scales that match better the biological metrics applied. Furthermore, the technical basis of metabolism determination, i.e. continuous measurements of oxygen and nutrients on river stretches will enable the validation of water quality models. These measurements do not need to cover large timeframes, but can be short characterization periods of 2-3 diel cycles in different seasons. Hence, the need mobile instrumentation to be able to switch sites regularly.

 


Background of the technique
River metabolism monitoring features essential ecosystem function metrics like gross Primary Production (GPP) and Ecosystem Respiration (ER). These parameters are derived from diel O2-concentration curves. The instrumental set-up consists of 2 continuously measuring oxygen probes that are exposed at a stream length distance of 100-300 m. The balance is made by difference under light and dark regime by taking into consideration re-aeration. The latter is estimated from equations related to flow velocity and mean depth. In addition newly developed continuous probes for photometric measurement allow for the observation ofNH4, NO3 and o-PO4 concentrations in the same time interval. The GPP/ER have a seasonal tendency and therefore need to be measured under comparable conditions. Observations on the Alzette river in Luxembourg during 2009 showed little variability during the vegetation period, flood waves have a short-lived effect but temperature has a stronger grip on metabolism. The characterization of the stretches has to be performed in summer (mid-April until mid-September). It is quite probable that macrophyte dominated sites will show a strong life-cycle dependence. Hence repeated measurements of some sites might be useful. The sites will be hydraulically characterized by a tracer test with salt and Rhodamine WT. This allows for discharge, flow velocity, mean depth and structure characterization (transient storage). The sonde exposures will be accompanied by sediment re-suspension sampling and epilitihic biofilm sampling. These samplings are performed to determine algal and microbial biomasses as well as respiration of sediments and biofilm. These serve to make assumptions on biomasses in the different compartments needed for ecological modeling.

To top

Sediment nets


Many pollutants of relevance like most metals and hydrophobic organic compounds accumulate in sediments and biofilms. Their mode of conveyance is intermittent following re-suspension or abrasion of their carrier material. Uptake into food webs of particle-bound compounds is mainly via diet. In that respect it is of relevance to characterize suspended matter at low-flow, as bed sediments and epilithic biofilms provide the material for this constituent. Pollutants with high solid phase affinity are very difficult to measure in (whole)-water samples because of limit of detection issues. In addition grab samples have the disadvantage of representing a very small interval in time. Plankton nets with defined mesh size can collect suspended matter during base-flow and are able to accumulate an integrated sample of a few grams in 1-2 days depending on the turbidity of the river. After 63 μm sieving in the laboratory enough material is available for further processing of digestion and solvent extraction with sufficient replicates. The setup is simple with spiral anchors securing a ring with the net in the river. There are minimal sizes of ring aperture and net size to be respected to collect enough material. With such a simple set-up it is easy to monitor longitudinal profiles of 6-12 sites in a day. This offers very interesting opportunities of assessing the immission situation. The set of parameters analyzed will cover carbon, nitrogen, chlorophyll, microbial biomass (PLP) as well as metals, pesticides and PAH.

To top