Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 
Home Print this page Email this page Small font size Default font size Increase font size Users Online: 865


 
 Table of Contents 
ORIGINAL ARTICLE
Year : 2016  |  Volume : 39  |  Issue : 1  |  Page : 30-37  

Evaluation of tritium dispersion in the atmosphere by Risψ Mesoscale Puff modeling systems using on-site meteorological parameters for the nuclear site Tarapur, India


1 Health Physics Division, BARC, Trombay, Mumbai, Maharashtra, India
2 Radiation Safety Systems Division, BARC, Trombay, Mumbai, Maharashtra, India

Date of Web Publication1-Jul-2016

Correspondence Address:
Vedesh K Varakhedkar
Health Physics Division, Trombay, Mumbai - 400 085, Maharashtra
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/0972-0464.185172

Rights and Permissions
  Abstract 

Dispersion models are important predictive tools that are used to simulate the way the atmosphere transports and diffuses contaminants from industrial sources of pollution. Risψ Mesoscale Puff (RIMPUFF) modeling system is used to simulate the radioactive H-3 released into the atmosphere through stack height of 100 m from a Tarapur Atomic Power Station 3 and 4 (TAPS 3 and 4) to predict downwind tritium ambient air concentrations in the environment around nuclear power plants. The tritium air concentrations by field measurement (measured tritium air concentrations using bubbler setup in the areas adjacent to TAPS 3 and 4, a pressurized heavy water reactor [PHWR]) were compared with that by calculation to validate the modeling system RIMPUFF for the Tarapur site. The computed and measured atmospheric tritium concentrations were quite consistent in trend and magnitude and the value of fractional bias computed is − 0.2524 (i.e., model predictions are within a factor of 2). This RIMPUFF modeling system will be useful in reviewing and evaluating environmental radiological impacts for PHWRs, especially it will be of great help to predict the behavior of tritium in the atmospheric environment around nuclear power plants during emergencies. Tritium concentrations in ambient air computed and measured at various locations around TAPS 3 and 4 showed a best fit regression line passing through the origin as Y (computed concentration) = 0.6725 X (measured concentration) with correlation coefficient of 0.75.

Keywords: Risψ Mesoscale Puff Modeling system, tritium (H-3) air concentrations, tritium bubbler setup


How to cite this article:
Varakhedkar VK, Baburajan A, Vanave S V, Rao D D, Ravi P M, Tripathi R M. Evaluation of tritium dispersion in the atmosphere by Risψ Mesoscale Puff modeling systems using on-site meteorological parameters for the nuclear site Tarapur, India. Radiat Prot Environ 2016;39:30-7

How to cite this URL:
Varakhedkar VK, Baburajan A, Vanave S V, Rao D D, Ravi P M, Tripathi R M. Evaluation of tritium dispersion in the atmosphere by Risψ Mesoscale Puff modeling systems using on-site meteorological parameters for the nuclear site Tarapur, India. Radiat Prot Environ [serial online] 2016 [cited 2020 Oct 29];39:30-7. Available from: https://www.rpe.org.in/text.asp?2016/39/1/30/185172


  Introduction Top


Tritium from nuclear facilities is released in the environment through a high stack mainly as tritiated water (HTO). [1] Tritium is produced naturally by cosmic ray interactions with atmospheric nitrogen and oxygen nuclei and also present due to nuclear tests in the past. These sources have resulted tritium in natural water levels for the region of study, which varies from 30 to 50 T.U (1 T.U = 0.1184 Bq/L). [2] Tritium, the heaviest and only radioactive isotope of hydrogen, has been an ever-present contaminant produced in pressurized heavy water reactor (PHWR). Canada Deuterium Uranium (CANDU) reactors are both moderated and cooled by heavy water (D 2 O). Tritium is produced in CANDU reactors by neutron reactions with deuterium, boron, and lithium and by ternary fission. Activation of deuterium is by far the most important mechanism, which is responsible for the production of about 89 TBq of tritium per MW (e) per year compared to 0.7 TBq of tritium per MW (e) per year produced by ternary fission. Most of the tritium present in CANDU reactors is in the form of tritiated heavy water - DTO. [3] At Tarapur Atomic Power Station 3 and 4 (TAPS 3 and 4), the effluents from the off-gas system are led through a high-efficiency filter system and discharged to atmosphere through 100 m tall stack.

Risψ Mesoscale Puff model (RIMPUFF) is a Lagrangian mesoscale atmospheric dispersion puff model used in this study for computing the concentration resulting from the dispersion of airborne materials. The model can cope well with the nonstationary and nonhomogeneous meteorological situations, which often are of interest in connection with calculations used to estimate the consequences of the short-term release of airborne materials into atmosphere. [4] This study deals with the computations of atmospheric 3 H concentration at different locations by RIMPUFF model and comparison with the measured concentration by appropriate instrumental methods.

Tritium in air is measured by collecting air moisture samples using condensation method as an environmental surveillance program on routine basis. Air moisture samples were collected and measured for concentration using Ultima Gold LLT scintillator from about 30 locations (population centers) in the off-site environment covering up to a distance of 30 km from the stack of TAPS 3 and 4. In this study, an attempt is made to diagnostically estimate concentrations using RIMPUFF by considering real-time on-site source and meteorological data (wind speed, wind direction, and temperature) and surface-based ambient air temperature, hourly rainfall, etc., for the year 2009.

Tarapur nuclear site is situated on the West Coast of India, about 100 km North of Mumbai. TAPS comprises two boiling water reactors TAPS 1 and 2 of 160 MW (e) capacity each and two PHWR type reactors TAPS 3 and 4 of 540 MW (e). Tarapur nuclear site is situated on a coastline of Arabian Sea and the terrain is fairly flat and even. A well-equipped micrometeorological laboratory (MM Lab) with 30 m high met tower is located at 1.5 km fetch distance from the East of coastal line as shown in [Figure 1].
Figure 1: Tritium sampling locations and Tarapur nuclear site

Click here to view



  Materials and methods Top


Meteorological measurements

The Met tower location is exposed to relatively open fetches consisting mostly of annual grasses in all directions. A range of 5-15 m high buildings are located over a fetch distance of about 500 m from the met tower. The MM Lab has a 30 m high meteorological tower which is a self-supporting, square type, hot dip galvanized steel, with four platforms at 7, 15, 22.5, and 30 m, with access ladder and safety guard.

MM Lab is operated near the nuclear site to generate the concurrent wind and turbulence data for assessing the dispersive capacity of the atmosphere and compute associated radiological impact to member of public at different distances in all the sixteen compass directions, due to the operation of nuclear facilities.

The ultrasonic anemometers are operated at heights of 7 and 30 m with 40 Hz frequency and provide wind and turbulence data on a continuous basis. Combined temperature and humidity sensor is operated at 1.2 and 30 m heights. The present investigation uses wind and turbulence data from sonic anemometer for the year 2009 on hourly averaged basis, and temperature humidity and rainfall data from surface met station. Sonic anemometers provide hourly averaged data for three wind components (u, v, and w) and their standard deviations (σu , σv , and σw ) in m/s, wind speed (m/s), wind direction (degrees), along with turbulence parameters such as, friction velocity (m/s) and characteristic temperature (degree K), drag coefficient, Monin-Obukhov stability parameter (m−1 ), vertical momentum flux (kg/ms 2 ), vertical heat flux (W/m 2 ), and atmospheric diffusion category. Pasquill atmospheric stability class is parameterized using values of heat flux provided by ultrasonic anemometers. [5]

Annual wind rose for the year 2009 for 30 m height is shown in [Figure 2].
Figure 2: Wind rose for the year 2009

Click here to view


Tritium measurement

A tritium bubbler setup [Figure 3] for sampling of HTO (tritiated water vapor) in air at various locations within plant site, fence boundary is used to run for 6 h of duration from local time 10 to 1600 h. Location of the tritium sampling using bubbler setup was chosen depending on the prevailing wind field. As the sampling was aimed at measuring ground level concentrations of tritium at the locations within 1.6 km fence boundary, i.e., nearer to a tall stack height of 100 m, daytime sampling was preferred overnight time as there will be a better probability of detecting ground level concentration during unstable atmospheric stability conditions than compared with night stable atmospheric stability condition.
Figure 3: Sketch of a bubbler setup

Click here to view


Twenty milliliters of distilled water are taken in a glass bubbler of capacity 125 ml under ice-cooled condition (to avoid the evaporation loss). Air is passed through the bubbler at a flow rate of 2 lpm. The 10 ml of the collected sample was mixed with 10 ml of Ultima Gold LLT scintillator solution and subjected to liquid scintillation counting. LSA system was used for counting the sample. The background of the system was 1.5 cpm and the minimum detectable activity is 0.04 Bq/m 3 for a counting time of 6000 s and a flow rate of 2 lpm for 6.0 h. The activity of tritium in the sample is computed using the following formula:





where, C1 = Gross count (sample + background) in time t 1 seconds, C2 = Background counts in time t2 seconds, E = Percentage efficiency of the system (LSA). V1 = Volume of water taken in the bubbler in liters. V2 = Volume of the sample taken for analysis in liters. M = Volume of air passed through the bubbler in m 3 . A = Tritium concentration in air (Bq/m 3 ).

Risψ Mesoscale Puff modeling system

The RIMPUFF model, Mikkelsen et al., [6] Thykier-Nielsen et al., [4] and Pδsler-Sauer [7] models the plume of released material as a number of individual puffs. At each time step, the model/code advects, diffuses and deposits the individual puffs according to local meteorological parameter values and calculates the concentrations from puffs.

Once the advection and size of all puffs have been calculated, updated grid concentrations are obtained at each grid point (x g, y g, and z g) summing up all the contributions from the puffs in the grid. Assuming Gaussian distributions and total ground and inversion lid reflection, the formula for the concentration at a grid point from puff number (i) is given by in Bq/m 3 .





where,

Q(i) - Puff inventory in puff "i" (activity), = (release rate [activity/time]) × (elapsed time between puff-releases [time] × correction factors). x c (i), y c (i), z c (i) - Center coordinates of puff number (i).

z inv - Height of the inversion lid, S xy (i) , S z (i) puff dispersion parameters in horizontal and vertical directions, respectively S xy(i) , S z (i) >0.

The local scale model chain is a tool used for real-time simulation of atmospheric dispersion of airborne materials (radioactive isotopes) in the vicinity of their release point from the source. It consists of a chain of models which may be grouped into two main subsystems, the local scale preprocessor for atmospheric dispersion (LSPAD) and the RIMPUFF dispersion model. LSPAD reads meteorological data for "now-cast" simulations, interpolates/extrapolates these data onto the points of the calculation grid covering area 5 km × 5 km with 250 m grid size with center as TAPS 3 and 4 stack and calculates the fields of derived parameters needed by RIMPUFF. Even though samples were collected within 1.6 km fence boundary, computational domain area is chosen as 5 km × 5 km so as to get a broader picture of concentration simulation. RIMPUFF reads the source data and the data fields of LSPAD and performs the calculations of dispersion, deposition, and concentrations for radioactive isotope H-3. In this study, RIMPUFF is simulated to find the ground level concentrations at various locations around TAPS 3 and 4 within the site boundary (i.e., radius of 1.6 km). The RIMPUFF model is simulated from 1000 to 1600 h diagnostically with changing met conditions and the average value of all the hourly output concentrations for the desired location is compared with the value of measured concentration.


  Results and discussion Top


Estimation of tritium activity in the atmosphere is rather a complex phenomenon, and strictly speaking, verification or validation of models is not possible. [8] Testing the accuracy of model predictions against independent data sets has given valuable information about the confidence that can be placed in model predictions and about the factors that contribute to the accuracy (or lack thereof) of the models.

Model performance was evaluated, from an operational and a scientific viewpoint, using statistical and graphical procedures. The measure of performance recommended by the USEPA [9] is the fractional bias (FB) screening test. The general expression for the FB is given by:



where,

OB: Observed/measured values of concentration, PR: Predicted/computed values of concentration. In this evaluation, FB was selected as a basic measure of performance because it has two desirable features. First, the FB is symmetrical and bounded; values for the FB range between − 2.0 (extreme underprediction) to + 2.0 (extreme overprediction). Second, the FB is a dimensionless number, which is convenient for comparing the results from studies involving different concentration levels. Values of the FB that are equal to − 0.67 are equivalent to under prediction by a factor of two while values of the FB that are equal to + 0.67 are equivalent to overprediction by a factor of two. Model predictions with an FB of 0 are relatively free from bias. [10]

The model computed concentrations and measured concentrations of 3 H in air are given in [Table 1] for different locations within 1.6 km from site. FB was evaluated for 25 highest values, for which measured concentration and computed concentration are considered.
Table 1: Ground level ambient air tritium concentrations computed and measured at various locations around Tarapur Atomic Power Station 3 and 4, Tarapur with center as Tarapur Atomic Power Station 3 and 4 stack

Click here to view


The FB test value has been estimated to be FB = −0.25241 and the value of FB obtained is within ± 0.67; hence, we can say that computed concentrations using RIMPUFF simulation have low bias and predictions are within a factor of 2.

A total of 54 numbers of 3 H bubbler samples have been collected around TAPS 3 and 4 within a fence boundary of 1.6 km, of which measurement of tritium concentration for 39 bubbler sample was above MDL. It has been observed that for the samples measured, MDL, i.e., ≤0.04 Bq/m 3 , corresponding to which computed concentration also showed no or 0 concentration. The release levels of tritium from 100 m stack are low therefore order of concentration within 1.6 km are very low as can be seen from [Table 1], hence out of 39 positively measured samples computations were possible only for 28 samples. As shown in [Table 1], for 10 samples, computed concentrations are higher than measured concentration and for 1 sample where computed concentration is equal to measured concentration.

The RIMPUFF dispersion model exhibited the low bias between the computed and observed data for tritium in air measurements although the RIMPUFF dispersion modeling system requires more data input and computer time (related to the simulation of changing wind conditions in complex terrain).

Of 28 samples having both computed concentrations and measured concentrations, 22 samples agree with the ratio of measured to computed concentrations, within a factor of 2, and the average value of the ratio of all the measured to computed concentrations is 1.32 with maximum value being 2.64 which is in agreement with the findings. [11],[12] Totally, for 6 samples, the ratio of measured to computed value is greater than a factor of 2.

A total of 21 samples have been collected from TAPS 3 and 4 guard house (0.53 km, Sector ESE from stack). Concentration estimations were possible only for 14 samples out of 21. In seven samples collected during 16 th February, 2 nd , 12 th , 19 th , and 23 rd March, 28 th May, and 5 th November, during the sampling time, wind flow was not toward ESE, therefore computed concentrations were 0, but measurement of these samples showed small values of concentrations <0.1 (Bq/m 3 ) which may be due to its location being near stack i.e., around 500 m.

A total of 14 samples have been collected from TAPS 3 and 4 Gate #1 (1.42 km, Sector NE from stack) during the monsoon period (June, July, August, and September). Only 8 samples showed positive concentration above MDL but computation of concentration was only for 6 samples with 3 samples' ratio of measured concentration to computed concentration (M/C) was ≥2.

A total of 7 samples have been collected from TAPS 1 and 2 main gate (0.925 km, Sector NNW from stack), of which 3 samples showed positive concentration above MDL and also computation for all the 3 samples was possible and ratio of M/C was 0.52, 2, and 2.14 [Table 1].

A total of 8 samples have been collected from MM Lab (1.48 km, Sector ENE from stack), of which 6 samples tritium measurements showed positive concentration above MDL and computation for the 5 samples was possible and ratio of M/C ranged between 0.51 and 1.74.

[Figure 4] shows the best fit regression line passing through the origin for tritium concentrations in ambient air computed using RIMPUFF simulation and that measured by bubbler setup as Y (computed) = 0.6725 X (measured), and with a correlation coefficient of 0.75.
Figure 4: Tritium concentrations in ambient air computed and measured at various locations around Tarapur Atomic Power Station 3 and 4

Click here to view


RIMPUFF simulation for the tritium bubbler sample collected on November 23, 2009, 10-1600 h, at the location TAPS 3 and 4 guard house is considered. Meteorological data such as wind speed, wind direction, atmospheric stability, ambient air temperature, and rainfall data which were used for simulation are shown in [Table 2]. Actual Tritium release is considered to be releasing continuously for 6 h duration from 100 m stack and measured value of tritium concentration is 1.2 ± 0.03 (Bq/m 3 ). Wind speed and wind directions at 7 and 30 m levels of height are considered and wind field at release height level, i.e., 100 m was computed by LSPAD. Plotting of ground level concentration distribution over a calculated domain of 5 km × 5 km with center as TAPS 3 and 4 Stack at TAPS 3 and 4 guard house is shown in [Figure 4 [Additional file 1] [Additional file 2] [Additional file 3] [Additional file 4] [Additional file 5] [Additional file 6]]a-f.
Table 2: Risø Mesoscale Puff simulation for November 23, 2009 10-1600 h at the location Tarapur Atomic Power Station 3 and 4 guard house

Click here to view



  Conclusions Top


Results of tritium monitoring in the ambient air environment within the fence boundary of TAPS 3 and 4 demonstrate the validity of RIMPUFF modeling system adopted at Tarapur Nuclear Site and sampling method followed for the determination of tritium. A best fit regression line passing through the origin for tritium concentrations in ambient air is computed using RIMPUFF simulation and that measured by bubbler setup by Y (computed concentration) = 0.6725 X (measured concentration) with correlation coefficient of 0.75.

RIMPUFF performs the best in matching the observed tritium in air concentrations data where this is a result of the ability of this modeling system to better address complex terrain and shoreline situations through an algorithm that simulates changing wind conditions within fence boundary; hence, this model is suitable for short-term simulation, i.e., during accidental or emergency conditions.

RIMPUFF would be the model of choice to simulate impacts in the vicinity of on-site as seen from the results of this study and can be used to optimize locations of monitoring stations in the off-site of the industrial site.

Acknowledgements


We would like to thank Dr. K. S. Pradeepkumar, Associate Director, Health, Safety and Environment Group, BARC, for his keen interest and encouragement. Also, we would like to express sincere thanks to Dr. N. Karunakara, Mangalore University, for the discussion. We gratefully acknowledge the continuous support from station management, TMS, and all staff members of ESL, TAPS, for their help during the experiment.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

 
  References Top

1.
Blaylock BG, Hoffman FO, Frank ML. Tritium in the Aquatic Environment. Radiat Prot Dosimetry 1986;16:65.  Back to cited text no. 1
    
2.
Health Safety and environmental activities" by Health Safety and Environment Group BARC, Mumbai, (1999).  Back to cited text no. 2
    
3.
Murgoci S, Popescu I, Ibadula R. Health Physics Department, Cernavoda NPP "Overview of the Tritium-In-Air Monitoring System of Cernavoda Npp U1 Romania - Modernization and Improvement Project" International Conference Nuclear Energy in Central Europe 2001 Hoteli Bernardin, Portorož, Slovenia; 10-13 September, 2001.  Back to cited text no. 3
    
4.
Thykier-Nielsen S, Deme S, Mikkelsen T. RIMPUFF User Guide. Version 8, Revision 103. Roskilde, Denmark: Department of Wind Energy, RIS0 National Laboratory; 2004.  Back to cited text no. 4
    
5.
Peters G, Fischer B. Estimation of stability classes with a sonic anemometer thermometer. Hamburg, Germany: Meteorological Institute, University of Hamburg; 2002.  Back to cited text no. 5
    
6.
Mikkelsen T, Larsen SE, Thykier-Nielsen S. Description of the Risø puff diffusion model. Nucl Technol 1984;67:56-65.  Back to cited text no. 6
    
7.
Päsler-Sauer J. Description of the Atmospheric Dispersion Model ATSTEP. RODOS (WG2)-TN (97) 01. Draft Version; November, 1997.  Back to cited text no. 7
    
8.
Oreskes N, Shrader-Frechette K, Belitz K. Verification, validation, and confirmation of numerical models in the Earth sciences. Science 1994;263:641-6.  Back to cited text no. 8
    
9.
United States Environmental Protection Agency (U.S. EPA). Protocol for Determining the Best Performing Model. Washington D.C.: United States Environmental Protection Agency (U.S. EPA); 1992.  Back to cited text no. 9
    
10.
Dispersion Modelling, Comparison to Available Data and Model Inter-Comparison at Pickering Nuclear Generating Station (PNGS) Using ISCST3, ISC-Prime, AERMOD and CALPUFF. Draft. January, 2004.  Back to cited text no. 10
    
11.
NCRP. Radiological Assessment: Predicting the Transport, Bioaccumulation, and Uptake by Man of Radionuclides Released to the Environment, NCRP Report No. 76, Second Reprinting. National Council of Radiological Protection, Bethesda, MD.; 1992. p. 49.  Back to cited text no. 11
    
12.
Sharma LN, Debe B, Kumar A, Kumar M, Sundara Rajan P, Varakhedkar VK. Environmental Distribution of Tritium and Associated Dose to Public at Narora Atomic Power Station Proceeding 24 th IARP Conference; 1999. p. 263-7.  Back to cited text no. 12
    


    Figures

  [Figure 1], [Figure 2], [Figure 3], [Figure 4]
 
 
    Tables

  [Table 1], [Table 2]


This article has been cited by
1 Evaluation of public dose from FHR tritium release with consideration of meteorological uncertainties
Xiao Wu,Yang Liu,Kimberlee Kearfott,Xiaodong Sun
Science of The Total Environment. 2020; 709: 136085
[Pubmed] | [DOI]



 

Top
   
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Materials and me...
Results and disc...
Conclusions
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed636    
    Printed8    
    Emailed0    
    PDF Downloaded134    
    Comments [Add]    
    Cited by others 1    

Recommend this journal