[Seite 1]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
JRP EMRP-ENV57
- MetroERM -
Metrology for radiological early warning networks in Europe
1st June 2014 – 31st May 2017
JRP - Coordinator
Stefan Neumaier (PTB)
Deliverable Number: Deliverable D1.5.4
Deliverable Description: Site characterisation methods and corrective procedures developed
Type: Report
Lead Participant: BfS
Other Participants:
Delivery Due: May 2016
Actual Delivery Date: March2017
File name: MetroERM Deliverable D1.5.4_report.docx Page: 1/12
[Seite 2]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
1 Introduction
Monitoring of ambient dose rate is an important feature of radiological emergency preparedness and response systems. Automatic monitoring networks using dose rate probes exist in most European countries. More than 20 years ago, the European data exchange platform EURDEP was established sharing the data from the early warning networks.
Additional information is needed for the interpretation and comparison of observed ambient dose rate data from different monitoring networks – e.g. on European scale. Obviously, the physical properties of the detectors have to be known. In addition, site characterisation techniques are needed for the interpretation and comparison of measured data. This information is of special interest for the interpolation of dose rate data – e.g. in EURDEP – and for data assimilation techniques used in decision support systems like RODOS.
Table 1: Overview on site characterisation aspects derived from response of 14 countries to MetroERM WP1 questionnaires (see Deliverable 1.1.1).
| Country | Concept of ideal site / excluding criteria | Height a.g. | Documentation | Sites with 5 m flat grassland |
|---|---|---|---|---|
| Austria | no | 1 m | buildings | |
| Belgium | not really (5m grass) always flat natural ground | 1 m | 4 probe photos | 80% |
| Finland | lawn, no obstacles in 10 m (asphalt, walls, forests) | 2 m | no systematic documentation | 40 % |
| Germany | grassland in 20 m / no walls, roofs | 1 m | systematic incl. photos | about 80 % |
| Greece | grassland in 10 m | 1 m | no documentation | 100% |
| Ireland | grassy area (flooding < 30 cm) / no building in 20 m | 1 m | 360° photos record | 93% |
| Italy | not really (5m grass) | 1.5 m | some photo | 45% |
| Latvia | no | no | ||
| Lithuania | smooth grassland / no building in 20 m | 1 m | 42% | |
| Netherlands | no | 1 m | yes | 15% |
| Norway | No | 3 m | no | 64% |
| Poland | flat grassland / no roofs, walls, trees, tall buildings | 1,5 m | no documentation | 77% |
| Slovakia | meteorological gardens | 1 m | yes | 94% |
| Switzerland | grassland in 20 m / no walls, roofs | 1 m | digital documentation | 93% |
For this purpose, the definition of an ideal site and standard conditions for the mounting of the probes are helpful. The German national monitoring network (BfS network) approach uses the following definition: The probe is installed 1 m above extended flat and smooth grassland.
File name: MetroERM Deliverable D1.5.4_report.docx Page: 2/12
[Seite 3]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
Obviously, buildings and walls in the vicinity of a real probe site will influence measured dose rate data by shielding effects especially in case of freshly deposited activity after an accidental release. Other relevant disturbing objects are trees and sealed surfaces.
Other networks – especially in northern Europe - use elevated probe position above flat grassland. Other early warning networks use dose rate probe fixed on walls and on roofs.
- Impact of disturbing objects in the vicinity of a probe on measured dose rate
2.1 Natural background dose rate
Background data of ambient dose rate at a given location depend on secondary cosmic radiation (SCR) contribution and terrestrial radiation (TR). The SCR is a function of the altitude of a location and weakly depends on geographical latitude, air pressure and solar activity. On the one hand, the TR component is mainly given by natural radioactivity in soil and air. Secondly, artificial radioactivity from global fallout and accidental fallout (e.g. Chernobyl) as well as additional sources contribute to this component.
Thus, measured data of the TR component of ambient dose rate may strongly depend on the location of the probe within a few meter distance. Near the main entrance of the BfS building in Munich, this component varies between 0,035 and 0,09 µSv/h within a distance of 20 m.
Assuming a probe mounted 1 m above flat ground and a homogenously contaminated soil, the effective field of view of the detector is about 10 m. A circular area of 3 m accounts for more than 50 % of the total dose rate. Thus, locations well suited for measurement of the natural background could be chosen considering the following criteria:
natural ground within 5 m circle flat terrain within 5 m circle no buildings within 20 m circle
Ambient dose rate data from probes fixed above artificial ground or fixed on buildings may be strongly influenced by the radionuclide concentrations in used materials. For such locations, the impact of disturbing objects in the vicinity of the probe on background dose rate should be assessed. The representativeness of such locations can be derived from dose rate data measured by a handhold device at the location of the probe and at nearby locations on natural grassland.
2.2 Dose rate contributions from accidental release
In case of an accident radioactive material released into the atmosphere will be dispersed by atmospheric transport processes. Particles on aerosols and gaseous components will deposit due to diffusion (dry deposition conditions) and by rain-out and wash-out effects (wet deposition conditions). One important purpose of most monitoring networks is to detect and delineate regions
File name: MetroERM Deliverable D1.5.4_report.docx Page: 3/12
[Seite 4]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
where artificial radioactive contamination exceeds given levels of intervention. For an ideal location the additional contribution to the ambient dose rate can be assessed from radionuclide activity concentrations in air and activity concentration deposited on ground using nuclide specific conversion factors [3]. Similarly, activity concentration on ground can be assessed from observed net dose rate using these conversion factors, if the relative contributions of radionuclides are known [4].
For real locations, site location factors are introduced to the more complex relation between activity in air and on ground and ambient dose rate.
H*(10) = H* + H* + H* BG air depos
H* = ∑ f (i) g (i) A (i) (1) air loc,air air air
H* = ∑ f (i) g (i) A (i) depos loc depos depos
In this notation, H* characterise the background value for ambient dose rate at the given BG location, A (i) denotes the nuclide specific activity concentration in air and A (i) gives the air depos activity concentration deposited on ground.
Fig 1: Dose rate contribution of a circular contaminated area with a given radius. The calculated data are adequate for a Cs137 (662 keV) contamination and a location, where the probe is fixed 1 m above flat extended terrain. The calculations were made for four different relaxation mass (from [1]).
The nuclide specific dose rate conversion factors are denoted by g (i) and g (i). They are valid air depos to calculate ambient dose rate for homogeneously contaminated air and fresh deposited activity deposited on flat grassland.
File name: MetroERM Deliverable D1.5.4_report.docx Page: 4/12
[Seite 5]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
The factors f (i) and f (i) describe the impact of the real probe location and the disturbing loc loc air influence of buildings, vegetation, non-flat terrain on dose rate.
In principle, it is possible to assess nuclide dependent site calibration factors f (i) using Monte- loc Carlo simulation methods described by Zähringer and Sempau [1]. This approach was used to assess site location factors for locations in the NPP surveillance monitoring network of Baden- Württemberg and locations of the BfS monitoring network in the same state [2]. However, the assessment of site location factors would rely on a realistic three dimensional model within a circle of about 100 m around the probe. For example, this model has to consider topography, real dimensions of buildings as well as knowledge about used materials, size and height of single trees. Additionally, the deposition conditions and relative contributions of radionuclides have to be considered.
2.3 Dose rate contributions after the passage of the contaminated cloud
After the passage of the contaminated cloud, dose rate contribution from activity in air is negligible. The proposed site characterisation method uses the following idea. In a first step, the real probe at a real location is replaced by a probe with ideal properties. In a second step, this probe is moved to an ideal location with the same deposition conditions. Thus, the observed ambient dose rate contribution from activities deposited on ground is related to the “true” value H* depos, ideal
H*(10) = H* + f H* (2) BG loc depos, ideal
f ≤ f ≤ f loc, lower loc loc, upper
The underlying assumption is, that the pure uncertainty of the measurement f is independent mu from the “additional” uncertainty f introduced by the deviation of the real location from an ideal loc one [5]. The basic idea is, to assess the “potential” range of effective location calibration factors with the help of a realistic but simple set of parameters describing a given location. In addition, a bias factor is calculated denoting the “best estimation” for the effective location calibration factor. This range of location calibration factors is characterised by their lower and upper limit. Thus, the concept of location calibration factor [1] is replaced by the approach of location evaluation factors f andf . loc, lower loc, upper
2.4 Disturbing objects
The impact of the height of the detector above ground was also discussed by Zähringer, Sempau [1]. Obviously, the dose rate from freshly deposited activity decreases slightly with increasing height of the detector.
Zähringer and Pfister [2] discuss the following disturbing objects and their main effects:
Sealed surfaces – lower deposition velocity and run-off effect Buildings, walls – shielding effect (attenuation, scattering of radiation)
File name: MetroERM Deliverable D1.5.4_report.docx Page: 5/12
[Seite 6]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
Trees, bushes – filtering effect due to dry deposition Uneven ground – shielding effect Sewage disposal plants
2.5 Proposed site characterisation procedure
Obviously, the impact of a single disturbing object on observed data strongly depends on its distance from the probe. For example, the impact of a near single tree is relevant. However, single trees are negligible, if their distance is beyond 20 m.
Thus, the proposed site characterisation procedure divides the vicinity of a given location into four zones:
Zone 1: circle of 3 m surrounding the probe Zone 2: circular ring between 3 and 7 m Zone 3: circular ring between 7 and 20 m Zone 4: circular ring between 20 and 100 m
For a fresh deposition of radionuclides on flat grassland, the corresponding dose rate contribution of each zone is about 25% (see fig. 1). For each zone, the total area is divided into sub-areas for four “surface types”:
Type 1: Grassland, agricultural used areas, areas with low vegetation Type 2: Sealed areas (e.g. streets, paved areas) Type 3: Shielded areas, buildings, areas covered with water Type 4: Trees, bushes, forest
The procedure could be separated in two main steps:
Step 1 – Documentation of disturbing objects
Obviously, the site characterisation procedure for a given location should rely on relevant information of disturbing objects in the vicinity of the probe. This information should be collected in a standardised manner. Within the BfS monitoring network, three different methods are used to supply this information: surveying techniques, photographic documentation and the usage of aerial views of the location.
In most cases, the surveying techniques are appropriate for zone 1, 2 and 3. Photographic documentation is helpful to give quick and objective impressions on the vicinity of the probe. However, these techniques are not appropriate for zone 4. Due to its large area, documentation have to rely on high resolution maps or aerial views. For most locations in central Europe this information is accessible via internet.
Step 2 – Assessment of site evaluation factors with the help of the site questionnaire
File name: MetroERM Deliverable D1.5.4_report.docx Page: 6/12
[Seite 7]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
With the help of the information collected in step 1, the relative contribution of each surface type to the total area of each zone has to be assessed. For example, the impact of a single tree is assessed by its distance from the probe location and the projected area of its crown onto the ground. For a tree in a distance of 12 m from the detector and a radius of the crown of 5 m the contribution to zone 3 is about 7 %.
From this input to the site questionnaire (see fig. 2) the site evaluation factors are assessed by a simple algorithm:
f = ∑ C(i,k) w(i) f (k) f = ∑ C(i,k) w(i) f (k) (3) loc, lower lower loc, dry dry
f = ∑ C(i,k) w(i) f (k) f = ∑ C(i,k) w(i) f (k) loc, upper upper loc, wet wet
In this equation set, C(i,k) denotes the contribution of surface type k to the area of zone i, w(i) denotes the weighting factor for zone i and f (k) and f (k) denote the lower and upper limit of lower upper the additional range of uncertainty for surface type k. These factors (see Table 2) are assessed by expert judgement based on the Monte-Carlo-simulations [1,2,6] and assumptions on the deposition process for freshly deposited activity. In a similar way “best estimates” for the bias can be assessed using bias factors for dry and wet deposition conditions.
Table 2: Excel sheet approach site characterisation procedure for a given location: parameters for weights w(i) and factors in equation (3) (blue figures), input values (blue cells) and results for location factors (green cells).
| Weight | 0.4 | 0.3 | 0.3 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contribution to zone area | Uncertainty | Bias | ||||||||||||
| Type | Zone 1 | Zone 2 | Zone 3 | Zone 4 | Lower | Upper | Dry | Wet | ||||||
| Lawn | 1.00 | 0.60 | 0.80 | - | 0.8 | 1.2 | 1.0 | 1.0 | ||||||
| Sealed | 0.00 | 0.20 | 0.00 | - | 0.1 | 1.0 | 0.3 | 0.6 | ||||||
| Shielded | 0.00 | 0.20 | 0.20 | - | 0.0 | 0.2 | 0.1 | 0.1 | ||||||
| Trees | 0.00 | 0.00 | 0.00 | - | 0.50 | 10.0 | 5.0 | 1.0 | ||||||
| Trees Zone 4 | 0.00 | 0.15 | 3.00 | 1.50 | 0.30 | |||||||||
| 0.66 | 1.07 | 0.85 | 0.87 |
The proposed simple site characterisation procedure can be illustrated table 2. The operator of the network have to assess ten input parameter which indicate the contribution of each surface type to the total area of each zone. For zone 4 only the contribution of trees (or forest) has to be estimated. It is adequate to restrict the precision of the input parameters to 10 % - this precision was suggested by previous experiences with the BfS approach. Table 2 shows the parameters used in equation 3 (blue figures), the input parameters for a given location (blue cells) and the
File name: MetroERM Deliverable D1.5.4_report.docx Page: 7/12
[Seite 8]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
results for assessed location factor (green cells). It uses a simple excel sheet application of the procedure for a site, where zone 1 is fully covered by lawn, while 20 % of zone 2 is paved and about 20 % of the zone 2 and 3 is covered with a small building.
Table 3: Results of site characterisation procedure for different locations
| Probe is fixed | Lower | Upper | Dry | Wet |
|---|---|---|---|---|
| 1 m above flat grassland | 0.8 | 1.20 | 1.00 | 1.00 |
| 1 m above flat grassland (50%) and sealed area | 0.45 | 1.10 | 0.65 | 0.80 |
| on a wall 1 m above flat grassland | 0.40 | 0.70 | 0.55 | 0.55 |
| in 3 m distance from a building | 0.56 | 0.90 | 0.73 | 0.73 |
| Example for probe location on a roof | 0.34 | 1.17 | 0.57 | 0.80 |
| Above flat grassland in a 10 m clearance | 0.91 | 5.52 | 3.10 | 1.30 |
2.6 Extension to non-standard height of the probe
The impact of the height of the detector above ground was also discussed by Zähringer, Sempau [1]. Obviously, the dose rate from freshly deposited activity decreases slightly with increasing height of the detector. On the other hand, dose rate contribution of zone 1 and 2 strongly decreases due to geometric effects. Initially, the proposed site characterisation procedure assumes that the probe is mounted 1 m above grassland. However, the procedure could be expanded to locations with non-standard probe height above grassland with the help of information obtained from Fig. 2.
Table 3: Excel sheet approach site characterisation procedure for a given location mounted on a roof 10 m above ground.
| 1m Weight | 0.4 | 0.3 | 0.3 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contribution to zone area | Uncertainty | Bias | ||||||||||||
| Type | Zone 1 | Zone 2 | Zone 3 | Zone 4 | Lower | Upper | Dry | Wet | ||||||
| Lawn | 0.00 | 0.00 | 0.00 | - | 0.8 | 1.2 | 1.0 | 1.0 | ||||||
| Sealed | 1.00 | 1.00 | 0.20 | - | 0.1 | 1.0 | 0.3 | 0.6 | ||||||
| Shielded | 0.00 | 0.00 | 0.30 | - | 0.0 | 0.2 | 0.1 | 0.1 | ||||||
| Trees | 0.00 | 0.00 | 0.00 | - | 0.50 | 10.0 | 5.0 | 1.0 | ||||||
| Trees Zone 4 | 0.00 | 0.15 | 3.00 | 1.50 | 0.30 | |||||||||
| 10 m Weight | 0.04 | 0.1 | 0.24 | 0.35 | ||||||||||
| Lawn | 0.00 | 0.00 | 0.50 | 0.60 | 0.8 | 1.2 | 1.0 | 1.0 | ||||||
| Sealed | 0.00 | 0.00 | 0.00 | 0.00 | 0.1 | 1.0 | 0.3 | 0.6 | ||||||
| 0.34 | 1.17 | 0.57 | 0.80 |
File name: MetroERM Deliverable D1.5.4_report.docx Page: 8/12
[Seite 9]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
In a similar way it can be expanded to locations, where the probe is fixed on the wall or on the top of a building or on a pole. A straight forward extension can be represented by Table 3. The zone area is subdivided into two levels of height: For example, level 1 represents the level of a flat roof of a building where the probe is mounted, while level 2 represents the ground level (about 10 m) below (around the building). In this case the roof area is considered as type 2 (sealed area).
On the other hand, dose rate contribution of zone 1 and 2 strongly decreases due to geometric effects.
Fig 2: Dose rate contribution of a circular contaminated area with a given radius. The calculated data are adequate for a Cs137 (662 keV) contamination. The parameter h denotes the height of the detector above flat extended grassland (from [1]).
2.7 Monte Carlo simulation results for real probe locations
In addition to the simplified site characterisation procedure discussed in 2.5, a Monte Carlo simulation approach following the methods discussed by Zähringer, Sempau [1] can be used taking into account standardized site characterisation input for each location. Such an approach is
File name: MetroERM Deliverable D1.5.4_report.docx Page: 9/12
[Seite 10]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
useful for networks using probes fixed on large buildings. However, the approach should be used to assess the uncertainty range for location factors as well.
Furthermore, Monte Carlo simulation methods can be used to assess the impact of artificial radionuclide deposition on probe casing. Same aspects of this problem can be discussed using data from the MetroERM Radon chamber experiments [9].
2.8 Experimental verification
Using enhanced dose rate data due to wash-out effects of radon progenies (rain events), some aspects of discussed site characterisation methods can be verified. BfS analysed observed data from dose rate probes at two reference sites: the INTERCAL [7] facility on mount Schauinsland near Freiburg and the Neuherberg reference site. At the INTERCAL facility, one additional probe was fixed on the wall of a building and one was mounted 1 m above ground in a small forest close to the INTERCAL facility. At Neuherberg reference site, one additional probe was fixed 1 m above ground on the wall of BfS building and one probe was fixed 1 m above ground at 3 m distance from the same building. At both reference sites, observed dose rate data from probes in non-standard position have been compared with data from probes fixed 1 m above grassland.
At INTERCAL facility net dose rate during rain events was reduced to a factor of 0.5 for both non- standard locations. At Neuherberg, the following typical reduction factors for net dose rate data were observed during rain events: 0.5 for the probe fixed on the wall and 0.75 and from the probe at 3 m distance from the building.
Fig 2: Observed net dose rate from probes fixed on non-standard positions compared to data from probes fixed on standard position [8].
File name: MetroERM Deliverable D1.5.4_report.docx Page: 10/12
[Seite 11]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
3 Use of site characterisation data in decisions support systems
Measurements of ambient dose rate can be used as input for decision support systems (e.g. RODOS) in several ways: (a) the dose rate data can be converted to ground contamination data (either using measured radionuclide ratios or ratios predicted by atmospheric dispersion models) and with these results model predictions of human exposure from groundshine, resuspension and possibly also the ingestion pathway can be calculated; (b) the dose rate data is used in data assimilation approaches to correct model predictions for e.g. ground contamination. In both cases, site characterisation data is needed to transform the measured data into dose rate data for an ideal reference site, for which only dose-conversion factors [3] are applicable. For the data assimilation approach additionally uncertainty information is required for the measured data: as the measurement uncertainty is partly caused by the deviation of the real measurement site from an ideal one, the lower and upper uncertainty bounds as being derived by the proposed approach are essential for ensuring a correct data assimilation process.
In the European R+D project CONFIDENCE, methods to reduce uncertainties in early and intermediate phases of accidental releases of radionuclides will be investigated. In WP2, measurement uncertainties of stationary and mobile monitoring systems will be assessed following results from the MetroERM project. A monitoring strategy will be developed, where mobile systems could be deployed to complement the stationary networks and optimise the assessment of doses to the population. Using the uncertainty estimations from WP1 in atmospheric dispersion modelling and the characterized measurement uncertainties, data assimilation approaches will be tested for improvement of the DSS in the early phase. The final goal is to derive individual exposure histories in order to identify critically exposed groups for subsequent monitoring.
4 Discussion and outlook
Site characterisation techniques have been used for different purposes. On the one hand they could be used for classification of different sites with respect to representativeness of measured data or as one criteria for network optimisation analysis.
In this paper, the proposed site characterisation technique is clearly linked to network harmonisation aspects and to an uncertainty model for ambient dose rate measurements under environmental conditions including disturbing contributions from non-standard probe locations. The proposed technique is adequate for situations, where freshly deposited activity dominates the total dose rate.
File name: MetroERM Deliverable D1.5.4_report.docx Page: 11/12
[Seite 12]
Deliverable D1.5.4
Delivered: March 2017
Author: Martin Bleher, Ulrich Stöhlker (BfS) Metrology for radiological early warning networks in Europe
References
[1] M. Zähringer and J. Sempau: Calibration Factors for Dose Rate Probes in Environmental Monitoring Networks Obtained from Monte-Carlo-Simulations. Internal BfS report BfS-IAR-2/97 (1997).
[2] M. Zähringer and G. Pfister: Representativeness and comparability of dose rate measurements: Description of site-specific uncertainties and data bias. Kerntechnik 63 (1998), p. 178.
[3] ICRU Report 53: Gamma-Ray Spectrometry in the Environment. (1994).
[4] M. Bleher, P. Jacob 93: Real-time Assessment of Radionuclide Deposition by PARK. Rad. Prot. Dosim. 50, p. 343-348 (1993).
[5] M. Bleher, U. Stöhlker, S. Burbeck: Uncertainty model for observed ambient dose rate data. Appendix to INTAMAP report D5.4. Draft version (2008)
[6] Gering, F. (2005) Data assimilation methods for improving the prognoses of radionuclide deposition from radioecological models with measurements, Dissertation, Leopold-Franzens- Universität Innsbruck.
[7] Bleher M, Doll H, Harms W, Stöhlker U: INTERCAL: Long-term inter-comparison experiment for dose rate and spectrometric probes. Rad. Prot. Dosim. 160 (2014), 306
[8] M. Bleher: Site characterisation methods and corrective procedures. EURADOS /AM2016 Milano 2/2016 Joint Meeting WG3-S1 + WG1 of JRP-MetroERM
[9] P. Kessler: Results of irradiations of dosemeters in PTB's Rn-chamber. EURADOS /AM2017 Karlsruhe, Meeting WG3-S1
File name: MetroERM Deliverable D1.5.4_report.docx Page: 12/12