The timescale for operation and maintenance (O&M) activities is a critical parameter, and depends on the technology adopted and remediation end-point. The timescale contributes to the total lifecycle costs of a project and, when more than one technically feasible technology has been selected, it should be calculated as precisely as possible to help identify which one is the most commercially feasible. If the O&M time is cal-culated incorrectly, a suboptimal technology might be selected, while another technology with a different cost structure might have been more cost effective in the long run. Hence, in the feasibility study phase it is im-portant to investigate the costs structure (investment and re-investment versus O&M) of different technologies. The question is how to deal with uncertainties. Contingencies to cover a possible time overrun that are too high may be prohibitive on the market. Considering the possibility of a remediation prolongation in the costs estimate may lead the client to think that the bidder is not able to reliably calculate the time needed for remediation. On the other hand, many projects run longer than expected, giving the impression that the work was not carried out efficiently enough and not fulfilling the expectations of the client. Furthermore, correct calculation of the O&M time is crucial for fixed price projects.
Therefore, based on the experience of thousands of remediation projects globally, we have gathered all our expertise and tools in estimating remediation O&M times. The basis for all calculations is the determination of the contaminant inventory as precisely as possible. Usually the occurrence of residual product phases makes it impossible to determine to inventory reliably. However, recently several methods such as the impulse-neutron-neutron-tool (INN) for the determination of DNAPL and nuclear magnetic resonance measurement (NMR) for the quantification of light none aqueous phase liquid (LNAPL) have been developed allowing an improved inventory calculation. If the inventory and the removal rates are known it is possible to estimate the reme-diation time.
The removal rate can be determined in different ways, ranging from the oxygen mass transfer rates for air-sparging to the delivery rates of electron acceptors for biodegradation, to determine the current degradation rates in-situ. Knowing the electron acceptor delivery rate, the contaminant mass degradation rate can be calculated applying stoichiometric factors. When using the electron delivery rates there is a need to consider other adverse processes, besides non-productive processes, that are reducing the delivery rate. Furthermore, since natural processes in the aquifer may also use the electron acceptors in a non-productive way, it is im-portant to identify these processes and to characterize the subsurface in order to quantify the non-productive substrate consumption.
Full-scale degradation rates may be derived from pilot tests applying isotopic signature analysis. The principle underlying this investigation method is that only biodegradation causes an enrichment of heavier stable isotopes (e.g. 13C versus 12C) in the not yet degraded fraction. On the basis of the isotope enrichment factor and the biodegradation kinetics (frequently 1st order), the biodegradation rate may be calculated. Another option to determine the in-situ degradation, provided that the groundwater flow velocity is not too high, is push-pull-tests using the compound of concern (as non-conservative tracer) together with conservative non-degradable compounds. The tools will be discussed in detail.
Other tools involve the transport modelling including subsurface heterogeneity, a factor which can substantially increase the remediation time in comparison to the estimates.
The main objective of this project is to build a tool allowing the contaminated soil owner to evaluate the efficiency of the remediation scenarios. For this purpose, it is necessary to determine, for each scenario, and according to the pollutant, the critical parameters to successfully achieve the remediation, depending on the selected technique. The impact of each parameter is evaluated to weight its effect on the remediation efficiency.
The overall objective can be split in three phases. These phases require an important quantity of soil remediation data obtained from grey literature, remediation final reports and PhD thesis.
− First, a set of data is used to ascertain the different steps involving in each remediation technique and for each pollutant. This phase allows us to determine the critical points of each technique to achieve the remediation of the target area.
− The purpose of the second phase is to define the weight of different specific parameters (geology investigation, number of wells, determination of the radius of influence…) for each remediation technique. Like this, a different weight on the final efficiency of pollutants recovery is attributed to each parameter.
− The last phase is the validation of the tool, based on complementary reports. All the actions described in these reports are input in the tool and the estimated efficiency is compared to the real efficiency. This phase will allow us to confirm the influence of each parameter and the validation of the project.
The presentation will be focused on the major phase of the project: i.e. the weighting of the main factors influencing the overall efficiency. A detailed analysis of the project revealed that two factors are of major influence. The first is the heterogeneity of permeability and pollutant distribution in the source zone. For this purpose, we suggest some approaches for the analysis and ranking of the heterogeneity level at the sites. The second point concerns the adaptation of each step of the remediation to the specific pollution at the site. For the second point we suggest a catalog gathering the methods that were used in the field to better target the pollution. A statistical analysis is then used to weight these methods.
Background
The consequences of vapour transport in the unsaturated zone above groundwater aquifers is often overlooked and the advantages of the investigation and mapping of contaminant transport in confined and unsaturated media is underestimated.
At a site in Zealand, Denmark, a long standing contamination of the overlying moraine sands and clays with trichloroethylene (TCE) has been investigated and delineated to a relatively limited area of about 100 m². At a depth of 12 – 25 meter´s under the moraine clay, an unsaturated sandy horizon of about 10 - 15 meters overlies the regional groundwater reservoir.
Investigation of the soil air content in the unsaturated sand horizon using Geoprobe soundings indicated that the volatisation of trichloroethylene (TCE) from the TCE hotspot area in the moraine clay has led to diffusion to the underlying unsaturated zone and contamination of an area of about 5.000 m² with soil air concentrations of 500 – 1.000 µg TCE/m³. This contamination in the confined unsaturated zone under moraine clays constitutes a TCE contaminant mass of about 20 – 30 g which can continue to burden the groundwater reservoir for decades.
The contaminant situation with TCE in the unsaturated zone was further complicated by the presence of a severe contamination with perchloroethylene (PCE-hotspot) at an adjacent site which causes an addition source of PCE and TCE in the unsaturated and confined sandy horizon providing a one directional and increasing contribution with both PCE and TCE during the vacuum tests.
Objective
To estimate the amount of contaminant that can be removed by remedial activities involving ventilation of the unsaturated zone, vacuum ventilation and tracer tests have been performed.
Method
The investigation involved 5 vacuum ventilation tests from two wells each with two 5 m screens in both the upper and lower part of the unsaturated zone, i.e. just above the aquifer, and one well with a 5 m screen in just the top of the unsaturated zone. The three wells were situated at distances of 16, 60 and 62 m respectively. Vacuum was applied individually to the screens while measuring the differential pressure in the other screens.
The vacuum (max. capacity 103 m³/h at a vacuum of 150 mbar) was applied for a period of between 4 and 94 hours during the 5 ventilation tests. The lengthiest test was for the upper screen in the well closest to the TCE hotspot. Before and after each vacuum test, a soil air sample from the vacuum well was sampled and analysed.
At regular intervals during the vacuum tests, the content of volatile gases was screened with a PID (PhotoIonisation Detector), and the content of oxygen, carbon dioxide and methane was measured with a field instrument.
Furthermore, a tracer test with injection of 20 litres carbon monoxide gas at a well 16 m from the well closest to the TCE hotspot was carried out to evaluate gas transport time during vacuum extraction. The carbon monoxide content was measured continually at the vacuum well with an INNOVA gas meter.
Results
The ventilation tests demonstrate and compare the vacuum that can be achieved in the respective screens in the three wells. As expected, the influence due to the pumping from the vacuum well decreases with distance, so the well at a distance of 62 m to the vacuum well is less affected than the well at a distance of 16 m. The differential pressure measurements indicate a predominantly homogeneous unsaturated sand horizon in all directions.
The tracer test with carbon monoxide indicated a gas transport speed of 4 x 10-4 m/s with a vacuum of 54 mbar at the well closest to the TCE hotspot.
Data treatment of the results allowed calculation of the radius of influence at different depths and positions with different vacuum (pumping) strategies. If the unsaturated sand is sufficiently permeable, the vacuum test reveals a low differential pressure between wells which provides a large radius of influence, whereas a large differential pressure indicates a less permeable horizon in which case the radius of influence during remediation is less, indicating the need for additional vacuum wells.
Furthermore, the change in soil gas concentrations from start to stop of the vacuum ventilation test indicates the soil gas concentrations that can be achieved under steady state conditions. For the well close to the TCE hotspot, an increase in TCE soil gas concentrations by a factor of 5 was observed during the ventilation test. Due to the soil gas contamination with PCE and the degradation product TCE at the PCE hotspot at the adjacent site, the development in soil gas concentration at the well nearest to the PCE hotspot demonstrated an increase in soil gas concentrations with PCE and TCE by a factor of 17 and 7 respectively.
Conclusion and relevance
The vacuum ventilation and tracer tests provided valuable information about the extent of pollution in the confined unsaturated zone and allowed estimation of the potential rates of removal by full scale vacuum extraction.
Based on the tests, a theoretical extraction rate of about 2.6 kg /year was calculated for the well closest to the TCE hotspot, which indicates that remediation by ventilation should provide an adequate solution to the clean-up of TCE in the unsaturated and confined sandy horizon.
This case study presents the use of strategic Dual Phase Extraction (DPE) remediation technology together with conceptual site model and 3D mapping to maximise the return on investment in remediation of a chlorinated solvent, primarily Trichloroethene (TCE), plume identified beneath a former manufacturing site in the UK.
The site, occupying an area of approximately 8,200 square meters, comprises a former manufacturing facility that had previously been used for production of electronic components. The site is bounded by a school to the western and north –western, residential estate to the northeast and a road with residential properties beyond to the south, resulting in the presence of sensitive human health receptors around the site .
The results of previous environmental works identified potentially unacceptable risks to both water resources and human health receptors, primarily associated with concentrations of chlorinated hydrocarbons including TCE. The results also indicated the potential for off-site migration of dissolved plume of chlorinated solvent towards the residential area.
Site geology, hydrogeology and contaminant distribution were extremely complex. The initial investigations focused on understanding the geological conditions which did not match what was expected from the regional map, the vertical and lateral distribution of contamination, and understanding why the lateral distribution of contaminants appeared contrary to the groundwater flow direction. Concentrations of TCE were been measured in some areas greater than 1 g/L. Although the measured concentration of dissolved phase TCE is indicative of the presence of free phase Dense Non Aqueous Phase Liquid (DNAPL), no free phase DNAPL was observed during site investigation.
The client requirements were to manage immediate risks economically and effectively, while providing the maximum return on their investments in terms of managing the associated environmental liabilities. A series of different types of investigation techniques to provide real time data such as Membrane Interphase Probe (MIP), dye testing for LNAPL and Cone Penetration Testing (CPT) and multi-level soil and groundwater sampling were carried to understand the geology, hydrogeology and contaminant distribution beneath the site. The findings of these investigations were used to develop the conceptual site model together with 3D mapping of geology and contaminant distribution.
Based on the conceptual site model and 3D mapping, strategically selective Dual Phase Extraction (DPE) was chosen to achieve the maximum return on the investment. The DPE system was operated at the Site for only six months and a cumulative TCE equivalent contamination mass of 4 tonnes was removed from beneath the site over a period of 6 months. . A significant amount of contamination mass was removed and the groundwater concentration was reduced by 70 to 80% across the area while minimising the potential for off-site migration of contamination which was achieved at relatively low cost and effort over a short time scale.
Clients often request environmental cost estimates at different stages of the investigation or remediation project. Different levels of accuracy are generally associated with each stage of the project. Therefor contingencies are applied to cost estimates to cover unknowns, unforeseen circumstances, or unanticipated conditions that cannot be evaluated from the knowledge available at the time of estimate preparation (i.e at the different cost estimate levels).
ARCADIS developed a methodology to give insight in different aspects that have a direct influence on remediation cost estimate and hence the contingency of the cost estimate. The methodology is used to evaluate decisions and determine the financial risks within the progress of the investigation or remediation and hence optimize the remediation costs and reduce the contingency.
The methodology can be used at different levels of the project:
- Level 1: Post-Investigation cost estimate
- Level 2: Pilot testing/Feasibility Study cost estimate
- Level 3: Engineering/design cost estimate
Changes in the cost elements are likely to occur as a result of new information and data collected during the investigation or the design of the remediation.
As the level of project definition increases from the investigation stage through final design, the contingency should decrease along with the level of uncertainty.
Within the different levels of the project, different criteria that determine the cost estimate are defined. For each level of the project the cost estimate will range between an optimistic cost estimate, a realistic cost estimate and conservative cost estimate.
To date, this method has been used to determine cost estimates at a number of sites in Flanders.
Two case studies will be presented to show how strategic choices during the progress of the project will result in more reliable cost estimates with higher accuracy. Insight will be given in the percentage of uncertainty of the cost estimates at different levels of the project.