Radiation Protection and Environment

: 2021  |  Volume : 44  |  Issue : 2  |  Page : 103--109

Estimation of surface layer scaling parameters using SODAR for the coastal site of Tarapur

Vedesh Krishnacharya Varakhedkar1, Sanjay Vasudev Vanave1, Aramana Baburajan1, IV Saradhi2,  
1 Environmental Survey Laboratory, EMAD, TMS Tarapur, Boisar, Palghar, India
2 Environmental Monitoring and Assessment Division, Bhabha Atomic Research Centre, Trombay, Mumbai, Maharashtra, India

Correspondence Address:
Vedesh Krishnacharya Varakhedkar
Environmental Survey Laboratory, EMAD, TMS Tarapur, Boisar, Palghar 401504, Maharashtra


This report presents the directional dependence of surface scaling parameters, namely, roughness length and corresponding friction velocity, for neutral category at the Tarapur coastal site. The average roughness length of the lowest value of 0.5 m in the west (W) direction and the highest value of 0.885 m in the N direction has been observed and average friction velocity of the lowest value 0.39 m/s in the W direction and the highest value 0.73 m/s in the NW direction was observed for 2019. Sector average turbulent kinetic energy was estimated to be 1.413 m2/s2 and its dissipation rate as 0.33 m2/s3 for the 10 m elevation from the surface. Surface drag coefficient for the 10 m height was 0.0229 for the smooth ocean surface and 0.0255 for the land surface was observed for the Tarapur coastal site.

How to cite this article:
Varakhedkar VK, Vanave SV, Baburajan A, Saradhi I V. Estimation of surface layer scaling parameters using SODAR for the coastal site of Tarapur.Radiat Prot Environ 2021;44:103-109

How to cite this URL:
Varakhedkar VK, Vanave SV, Baburajan A, Saradhi I V. Estimation of surface layer scaling parameters using SODAR for the coastal site of Tarapur. Radiat Prot Environ [serial online] 2021 [cited 2022 Jan 23 ];44:103-109
Available from: https://www.rpe.org.in/text.asp?2021/44/2/103/329140

Full Text


Detailed investigation of the atmospheric dispersion requires a comprehensive dispersion model, which, in turn, requires surface layer parameters, namely, roughness length “Z0,” friction velocity “U*,” characteristic temperature, and sensible heat flux for the estimation of diffusion parameters such as turbulent kinetic energy and its dissipation rate. For applying flat terrain formulations to a complex terrain site, a suggested approach is to assume that the site is divided into different sectors of uniform and similar conditions and to apply flat terrain formulations for each of these sectors separately. In this directional, dependence of surface scaling parameters for the site is considered instead of single site averaged value. In the present study, sector-dependent roughness length and corresponding friction velocity under neutral stability condition using wind speed, wind direction, and atmospheric stability class for 2019 were evaluated. The direction-dependent nature of these parameters in relation to the site topographic characteristics is also discussed.

Knowledge of directional variation of roughness length and friction velocity would be a handful parameter to extrapolate the wind speed at the stack height, using the log profile law, where wind experiences a different drag with direction to the wind measurements at the met tower or SODAR site. Using the wind profiles measured from SODAR and estimated directional surface scaling parameters, it will be possible to extrapolate the wind speeds for elevated heights, i.e., stack heights when the SODAR is nonfunctional or not available for the coastal site Tarapur.

For a coastal site, if the wind blows from a smooth to a rougher zone, the rougher surface retards motion close to the ground and gives rise to a sharp decrease of mean horizontal wind speed resulting in the change of the vertical profile. The roughness of the surface is one of the fundamental factors in determining the wind profile near the surface along with the atmospheric stability and topography. Roughness is normally measured by a parameter called roughness length (Z0). As the kinematic stress against the earth's surface is estimated by the square of the friction velocity (U*), the magnitude of the friction velocity reflects the intensity of turbulence of the wind. Faster winds over the rough surface cause greater kinematic stress.

Surface layer theory is usually used in models to make predictions of near-surface mean profiles as well as surface fluxes. Surface layer is effectively the lowest model grid layer adjacent to surface (typically lowest grid layer spacing is around 10 m). The lowest 10% of the boundary layer is the surface layer. This surface layer is about 10–100 times of roughness length.

Description of site

Tarapur nuclear site is situated, on the west coast of India, about 100 km north of Mumbai. A well-equipped MM lab with 30 m high met tower and SODAR along with other surface meteorological sensors, located at 1.5 km fetch distance from the east of coast line as shown in [Figure 1].{Figure 1}

The tower site is exposed to relatively open fetches consisting mostly of annual grasses in all directions. A range of 5–10 m high buildings are located over a fetch distance of about 500 m from the met tower. The 30 m high meteorological tower and SODAR installed at MM lab which is in the northeast direction from TAPS 1 and 2, and is 8 m from MSL as shown in the [Figure 1]. Meteorological tower is a self-supporting, square type, hot dip galvanized steel structure, with three platforms at 10 m, 20 m, and 30 m, with access ladder and safety guard.

Wind data measurements

A SODAR and meteorological tower at MM lab, Tarapur, is used in this study for the analysis. It has been ensured that a SODAR is sited properly as per the regulations and also quality checks and performance audits for proper functioning of the SODAR are ensured. The details of wind sensors as per the manufacturer are given in [Table 1].{Table 1}

Metek GmbH SODAR

The METEK PCS2000-64MF mono-static SODAR is operated on 5 configurable frequencies: 1415, 1605, 1795, 1985, and 2175 Hz. To get the mean wind data, a clustering method is applied in the SODAR software by the Metek, Germany, which uses radial components from instantaneous spectra.[1] [Figure 2] shows the SODAR installed at Tarapur.{Figure 2}

[Table 2] gives the detailed technical Specifications of the M/s Dynalab Weathertech Pune make Cup anemometer (CA) and wind vane system and Metek GmbH Germany make Multi Frequency SODAR.{Table 2}

Siting and installation of SODAR at meteorological lab, Tarapur

A proper care has been taken for the installation of SODAR at MM lab, Tarapur, by ruling out any obstructions such as big strands of trees, broad and high buildings, or power lines, nearby. Met tower nearby which can reflect, transmitted pulses from SODAR and can contaminate the data, therefore met tower at MM lab, Tarapur, is placed 45 m away from the SODAR toward north and also it is assured that acoustic beams from the SODAR are aimed away from the met tower.

Advantages of SODAR over meteorological tower

The cost of the SODAR is normally higher than the wind monitoring/measuring system installed on meteorological tower, but when we compare all the costs of the tower erection, installation of sensors on multiple heights along with its maintenance with that of the cost of SODAR installation and maintenance, a SODAR system will look cheaper. Meteorological towers can be erected practically to about 100 m–150 m, but most SODAR systems measure wind and turbulence in the atmosphere almost up to 1 km, and avoid many liability issues, and does not require any permits.

Drawbacks of SODAR systems

During the times of high intensity of rainfall SODAR's, do not measure actual winds. SODAR's do not give point measurements, but instead they sample over a volume and at multiple points in space and time therefore measures only the mean wind. Wind parameters, such as standard deviation of wind speed, wind direction, and wind gust, are either not available or not reliable.

SODAR is not a replacement of meteorological mast. Collocation of SODAR with micrometeorological laboratory provides several advantages. Wind and turbulence data measured from SODAR can be a complimentary to wind and turbulence data measured from sensors installed on meteorological tower. For the data validation purposes, both the SODAR and met tower can complement each other, i.e., for data reasonableness checks (like whether surface winds from meteorological tower are roughly agree with the lowest height near surface winds from SODAR).

SODAR calibration techniques

SODAR normally measures the winds from near surface and up to a height in atmosphere which is not accessible to other measurement systems, such as meteorological tower. Therefore, the calibration technique is different from normal mechanical anemometers and wind vane and is described as follows:

The SODAR array's response to known input frequencies. Wind speed/F Shift (m/s/Hz)The beam steering of all the transducers needs to be checked by measuring phase angles for both the tilted beam directionsThe physical condition and its performance of all the transducers should be ensured for its proper operationFor the specific wind speed and wind direction for a particular height, SODAR input sound pulses may be measured and checked for correctnessPower amplifiers associated with each transducer may be checked with oscilloscope for transmitted and backscattered signal.

Meteorological tower

At MM lab, Tarapur, a 30 m high met tower is installed for the measurements of wind parameters with wind sensors (Dynalab Pune make CA and wind vane) which are installed at three levels of height 10 m, 20 m, and 30 m. The details of wind sensors installed on a meteorological tower along with SODAR at MM lab site are shown in [Figure 3].{Figure 3}

Cup anemometer

CA cups are of lightweight, and this cup assembly is attached to a vertical shaft to slotted disk at the bottom. A light photoregister assembly uses these slots to generate current pulses whose rate is proportional to wind speed and can be measured by a rate meter or otherwise electronically processed. The electronic data logger converts frequency into instantaneous wind speed. Wind speed measurement range is up to 65 m. s−1 with a resolution of 0.1 m. s−1.

Wind vane

Wind direction is measured by wind vane. It computes potential across a circular potentiometer, which is proportional to wind direction in degrees. Wind direction and wind speed give where and how far the air mass and airborne pollutant disperse. The wind vane sensor has a light weight vane assembly connecting the shaft of the wind vane to the mounting arm of 356° circular 5 KΩ. potentiometer. The shaft is supported by ball bearings. The vane aligns itself with wind direction and the measure of resistance of the potentiometer indicates the wind direction. Wind direction measurement range is 0°–359° with a resolution of 1°.

Meteorological tower data acquisition

The hourly CA wind speed and wind direction from wind vane for 10 m and 30 m height, and surface parameters ambient air temperature, relative humidity, solar radiation, delta T, atmospheric pressure and rainfall data averaged on hourly basis for the entire reporting period of 2019 was collected from a single Dynalab data logger at MM lab. The data acquisition program in visual basic has been written for storing hourly online data from Dynalab data logger in personal computer. Online data acquisition software output is shown in [Figure 4]. Diurnal variations of wind speed and wind direction profiles from SODAR for typical 24 h, i.e., 1 day on July 3, are shown in [Figure 5]. Concurrent hourly wind speed and wind direction from SODAR system, a different data acquisition program developed by Metek GmbH Germany, for 30 m, 40 m, 50 m,… 500 m is used and the data stored continuously in to PC. The data of wind speed for 10 m height, temperature difference called Delta T and solar radiation is taken from Dynalab data logger and used for the estimation of atmospheric stability. Wind speed and wind direction data from SODAR are used to study the surface scaling and wind profile studies in this report.{Figure 4}{Figure 5}

Atmospheric stability/air dispersion stability

Atmospheric stability category is the indicator of turbulence present in the atmosphere. Pasquill Gifford Stability Class (P G Class) denoted from A B C D-E F, with increasing stability. A Highly unstable, B Moderately unstable, C Slightly unstable, D Neutral, E–Slightly stable, and F Moderately stable. Stability describes the potential to vertical motion in to atmosphere. Atmospheric stability categories for the Tarapur site are estimated using the solar radiation/delta T method as outlined in.[2] During the day time, stability is estimated by wind speed and solar radiation and during night time wind speed and temperature difference “Delta T.”

Data analysis

In the present study, wind speed and wind direction for 30 m, 40 m, 50 m, 70 m, 90 m, and 110 m, i.e., for six levels of height, has been taken from SODAR for 2019 for the estimation of surface scaling parameters for neutral conditions.

[Table 3] gives the joint frequency distribution (JFD) of wind speed and wind direction in percentage occurrences for the predominant wind directions and for predominant wind speed classes for all the four quarters of the year and also for the whole year 2019. [Figure 6] gives the pictorial presentation of annual JFD called wind rose for wind speed and wind direction in percentage occurrences converted into centimeters for 2019.{Figure 6}{Table 3}

During different weather conditions, hourly average wind speeds for various heights are considered and the data have been extracted for the neutral atmospheric stability category. A total of 2517 hours of neutral category data has been observed and segregated direction wise in all the 16 sectors, i.e., 28.75% of the times in a year the stability category is neutral. Further data have been filtered or omitted, where variations in wind directions for 30 m and 50 m heights were observed >±22.5.


Log-law model is used for the estimation of roughness length and friction velocity and is defined as:


Where V(z) is mean wind speed at height “z,” “k” is Von-Karman constant (=0.41), U* is friction velocity = SQRT(τ/ρ), “Z0” Aerodynamic roughness length of the surface, τ Shear Stress at ground level, and ρ air density.

If the wind speed profile is available, under the steady state conditions, wind profile is given as:


The function Øm depends on the stability[3] as follows:


In the above expressions, L is the Monin-Obukhov scale length defined as:


Where Cp is the specific heat of air at constant pressure, “θ” is the potential temperature (K), H is the surface sensible heat flux in the vertical direction, and g is the acceleration due to gravity. In neutral conditions, L is infinite, it is positive under stable conditions and negative under unstable conditions.

The sea surface wind stress has generally been estimated using the drag coefficient. A drag coefficient at 10 m height as reported in the literature[4] and is given by:


U-mean wind speed at Z = 10 m.


and its dissipation rate


The turbulence constants Cμ-empirical constant = 0.09 describes the ratio of the turbulent kinetic energy to the magnitude of the Reynolds stresses as suggested by Brown.[5]

 Results and Discussion

Under the neutral condition, roughness length and friction velocity have been estimated specifically, for each upwind sector considering hourly wind data from SODAR. Recognizing that the roughness length and friction velocity are indicators of terrain nonhomogeneity, they are examined with respect to the type of surface conditions existing in the different sectors of the site. This is particularly true for the site like Tarapur, with land use through the sectors N-NNE–E-SSE-S as shown in [Figure 1], and all the other upwind sectors S-WSW-WNW-N with Arabian Sea. This topography conforms well with the order of average roughness length values for the upwind sectors N-NNE–E-SSE-S (ranging from 0.753 m to 0.885 m) and are comparatively higher than that for the other upwind sectors in the Arabian Sea (ranging from 0.5 m to 0.865 m). High roughness length of 0.865 m for the upwind sector SSW and 0.835 for SW can be attributed to the 1.5 km fetch distance of MM lab from the coastline and due to the presence of tall and broad buildings, trees, etc. Directional dependence of average roughness length and average friction velocity is given in [Table 4]. It has been found that the lowest value of an average roughness length for 2019 was 0.5 m for the upwind sector W and the highest value was 0.885 m for the upwind sector N at TMS, Tarapur. The lowest value of an average friction velocity for 2019 was 0.39 m/s for upwind sector W and the highest value was 0.73 m/s for the upwind sector NW.{Table 4}

Estimation of average turbulent kinetic energy using surface scaling parameters is 1.41 m2/s2 and its dissipation rate is 0.33 m2/s3 at the 10 m height of measurement, which is again dependent on upwind buildings and obstacles. A related parameter which is often used in modelling is “surface drag coefficient” (CD) for the reference 10 m height averaged over the upwind sectors falling over the Arabian Sea is 0.0229 and averaged over the upwind sectors falling over land is 0.0255 as shown in [Table 4]. Considering the fact that Tarapur is multinuclear facility site and the 1.5 km fetch distance of MM lab from the coastline, the estimated values of surface drag coefficient CD are well in agreement with the result of 0.001 reported by Renderson,[6] for a mean 10 m level wind for smooth ocean surface, 0.01 for a bushy canopy and 0.1 for an urban environment.

As Tarapur site and its surroundings are characterized with the patches of trees of 5 m–20 m in height and Arabian Sea in the west, also considering that Tarapur is a multinuclear facility site, the roughness lengths have been found to be different for various upwind sectors. This indicates that no single value of the roughness length can be assigned for the site. Hence, average roughness lengths for each upwind sector would be more appropriate for use in dispersion studies.


The authors would like to thank Dr. A. Vinod Kumar, Head, EMAD, for his guidance and constant encouragement. Thanks are also due to TMS authorities for their continuous support in carrying out the studies. The authors gratefully acknowledge the contributions and help from NPCIL staff of MM lab, Tarapur.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1De Haij M. On the Use of SODAR Reflectivity Data during Low Visibility Conditions at Schiphol Airport. KNMI, R and D Information and Observation Technology De Bilt, (De Bilt 2010 Internal Report, IR 2010–02), Netherland; August, 2010.
2Meteorological Monitoring Guidance for Regulatory Modeling Applications. EPA-454/R-99-005, U.S. Environmental Protection Agency; 2000.
3Businger JA. Flux profile relationships in the atmospheric layer. J Atmos Sci 1971;28:181-9.
4Hicks BB. Some evaluations of drag and bulk transfer coefficients over water bodies of different sizes. Boundary Layer Meteorol 1972;3:201-13.
5Brown GJ. Prediction of Odour Dispersion and Plume Visibility. Melbourne, Australia: CSIRO; 2003.
6Renderson D. Atmospheric Science and Power Production. Ch. 5. USDOE: Published by Technical Information Centre, USDOE, DOC/TIC-27601; 1984. p. 160.