SEARCH

SEARCH BY CITATION

Keywords:

  • complex terrain

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Site, System Description and Data Used
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[1] NCEP/NCAR Reanalysis wind components at 1000 mb level in the grid of 12.5°–15°N and 77.5°–80°E has been investigated in comparison with National Atmospheric Research Laboratory (NARL) Doppler Sodar data in a complex terrain environment of an Indian Tropical Latitude station, Gadanki (13.45°N, 79.2°E) for a period of one year 2007. Influence of complex terrain effect on wind flow patterns derived by NCEP/NCAR reanalysis II in comparison with remote sensor (NARL Doppler Sodar) during stable and unstable conditions is discussed. Slope-valley circulations and mountain wave activity may be the possible reasons for the deviation observed between both the systems. During unstable conditions, correlation co-efficient of 0.78 in zonal and 0.66 in meridional wind components is reported high compared to 0.71 in zonal and 0.56 in meridional wind components for stable conditions. Zonal winds dominant over this station might be the cause for higher correlation in zonal case.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Site, System Description and Data Used
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[2] Wind is an important parameter in the Atmospheric Boundary Layer (ABL) physics. It is the basic medium which transports heat, momentum and pollutants in the atmosphere in the horizontal. Wind is involved in all the physical process of advection and dispersion. It is a highly localized phenomenon which is influenced by the earth structures such as hills, ridges and local orography [Stull, 1988]. Hence, it is very important to have accurate information about wind flows. In situ instruments such as cup anemometers, wind wanes give information on wind at a particular location. Remote sensors such as Radars, Lidars and Sodars are also used for wind measurements with very high vertical and temporal resolutions. Many models also give wind information at various pressure levels of interest, which are used for research applications.

[3] The earth surface is not uniform around the world. A geographic variation always modifies the boundary layer flow and causes wind circulations in mountainous regions. Wind circulations are also generated in conjunction with diurnal heating cycles. A hill also can modify or blocks the wind flow; Lee waves and mountain waves are generated by the interaction of wind and hills. Complex terrain has a high influence on the wind flow depending on the height parameters of the hills and ridges. Detailed explanation of how the wind flow is modulated by flow over canopy, hills, changing terrain is given by Kaimal and Finnigan [1994]. Pollutant dispersion in a complex terrain is very difficult to model and more research is needed in this area [Egan and Schiermeier, 1986]. A study has been done on the zonal and meridional components of the NCEP/NCAR Reanalysis II data at a pressure level of 1000 mb with National Atmospheric Research Laboratory (NARL) Doppler Sodar at an altitude of 120 m to investigate the influence of complex terrain on the model derived wind components. The site description describing the non-uniform terrain in a detailed view is given in section 2. It also involves the data description and the operational parameters. Influence of the complex terrain over the model derived winds in comparison with Doppler Sodar measurements is discussed in section 3. Results are summarized in section 4.

2. Site, System Description and Data Used

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Site, System Description and Data Used
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[4] Gadanki, a tropical rural station is located at 13.45°N, 79.2°E in the southern part of the India. It is situated 370 m above the mean sea level, surrounded by hillocks whose altitude ranges from 200–300 m in the radius of 1 km and 200–700 m in the radius of 10 km. The topographic variation of the National Atmospheric Research Laboratory located at Gadanki is shown in the Figure 1. It is located 80 km radial distance from the east coast of Indian peninsula. The temporal and spatial sampling of NCEP/NCAR Reanalysis II winds of 2.5° × 2.5° is a distance of about 250 km. The terrain in the grid of 12.5°–15°N and 77.5°–80°E taken for the current study is very un-evenly distributed. Height variations of the hills vary to about 1000 m; shorter ridges are also present in between the valley.

image

Figure 1. Topographical view of the National Atmospheric Research Laboratory, Gadanki [after Krishna Reddy et al., 2002].

Download figure to PowerPoint

[5] Sodar pumps an acoustic energy of 100 W for each pulse in to the atmosphere. The system consists of 8 × 8 antenna array of piezo-electric transducers which pump the energy on to a reflector board from where the signal is sent into the atmosphere. The same transducers are used for the reception of the back-scattered signal. This array produces zenith and two off-zenith (North and East, tilt angle 16°) beams. Sodar measures the radial velocities along these three beams by analyzing the observed Doppler shifted signals adopting Doppler beam swinging technique. Such type of configuration enables measurement of horizontal (zonal, meridional) and vertical (zenith) components of wind speed. The system has height coverage right from 30 m above ground to 1 km (in favorable conditions). Here we are operating with a frequency of 1800 Hz with a pulse length of 180 ms which results in a 30 m vertical resolution. The returned signal depends upon the turbulent inhomogeneties present in the sampling area of the volume. Each beam takes 9 seconds to derive the velocity with altitude ranging from 30 m–1500 m with an interval of 30 m. The radial velocity values for each beam obtained for a given range bin over the total observation period are done through a process of consensus averaging, i.e. difference between continuous measurements have been considered to remove spurious spikes, noises and outliers from the data. Only moments computed from intervals with at least 80% of well distributed good data points have been taken for analysis. The radial wind velocities after passing through the quality check have been used to calculate the zonal, meridional and vertical components of wind. The formulas that are used to calculate the radial wind velocities from sodar are as follows

  • equation image
  • equation image

where ϕ is the tilt angle from the zenith beams.

[6] This observational specification gives high temporal resolution wind profile data for every 27 seconds from the sodar system for different atmospheric studies. The technical specifications of the NARL sodar system details inclusive of signal processing, data quality control and preliminary validation of the system data products have been given in a recent publication [Anandan et al., 2008]. The main operational parameters of the NARL Doppler Sodar are given in Table 1.

Table 1. Main Operational Parameters of the NARL Doppler Sodar
ParameterValue
Operating frequency1800 Hz
Pulse width180 ms
Wavelength (λ)0.17 m
Range resolution (r)30 m
Pulse repletion interval9000 ms
Transmitted acoustic power (Pt)100 W

[7] In the present study, wind components-zonal and meridional measured by the Doppler sodar during January 2007 to December 2007 have been used to investigate and discuss the influence on the model derived winds in the lower ABL in a complex terrain environment. For each hour 120 profiles of wind components data is obtained, these have been averaged for each hour. NCEP/NCAR reanalysis II wind components of zonal and meridional for 6 hourly averaged data at 1000 mb in the grid of 12.5°–15°N and 77.5°–80°E has been taken to study the comparison between NARL sodar and NCEP/NCAR reanalysis II wind components [Kalnay et al., 1996; Kistler et al., 2001; Kanamitsu et al., 2002] (see also http://www.cpc.ncep.noaa.gov/products/wesley/reanalysis2/kana/reanl2-1.htm). For this purpose hourly data of NARL sodar at 120 m is averaged for every 6 hours. Missing data either for NARL sodar or for NCEP/NCAR Reanalysis II is not considered for both the systems. Temperature data from the Automatic Weather Station located in NARL is used to classify the stable and unstable conditions. After the quality check, in stable conditions 668 profiles for zonal case, 391 profiles for meridional case and during unstable conditions 673 profiles for the zonal wind component, 389 profiles for meridional case were available for the present study.

3. Results and Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Site, System Description and Data Used
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[8] Figure 2 shows the correlation plot between the 6 hourly averaged wind from NCEP/NCAR reanalysis II at 1000 mb and NARL Sodar 6 hourly averaged wind at an altitude of 120m during stable conditions both for zonal and meridional cases respectively. A correlation coefficient of 0.71 with a standard deviation of 1.11 ms−1 is observed for the available 668 profiles in zonal case whereas low correlation is observed in the meridional case of 0.56 with a standard deviation of 1.02 ms−1 for the available 391 profiles. Figure 3 represents the same as Figure 2 but for unstable conditions. A correlation coefficient of 0.78 with a standard deviation of 1.24 ms−1 is observed for the available 673 profiles in zonal case whereas lower correlation is observed in the meridional case of 0.66 with a standard deviation of 1.16 ms−1 for the available 389 profiles. This shows that the zonal wind component has better correlation for both the cases even in a complex terrain environment where the orography is very unevenly distributed. This type of terrain always leads to flow modifications due to the valley and mountain ridges. These types of structures are responsible for generation of mountain waves and gravity waves. Data taken for the present study is always under the influence of the topographic effect. This station is also under the influence of the katabatic flows and valley winds, due to the mountain ridges present near to the station. But our study is only under the influence of the along valley axis flow; i.e. the valley winds which range from 10 m to 400 m and the mountains in the nearby location are exactly in this range. The winds are averaged at 0, 6, 12 and 18 hrs respectively; these will represent the cold winds at the night time known as drainage winds and the warm air during the day time valley winds. Terrain influence on the winds derived is clearly inferred from the results obtained. Higher correlation is observed in unstable conditions than the stable conditions for both the wind components. During the unstable conditions, due to the surface heating vertical mixing of air masses increases and this brings the cool air down and moves the warm air aloft in the valley environment that mask terrain influences to a certain extent. Whereas during the stable conditions, terrain slopes introduces the valley-slope circulations and local winds are generated. Slightly poor correlation is observed in the stable condition in comparison with unstable condition may be due to terrain influence which may not have well represented in the model.

image

Figure 2. Correlation plot of NARL Sodar and NCEP Reanalysis II data during stable conditions: (a) zonal wind component and (b) meridional wind component.

Download figure to PowerPoint

image

Figure 3. Correlation plot of NARL Sodar and NCEP Reanalysis II data during unstable conditions: (a) zonal wind component and (b) meridional wind component.

Download figure to PowerPoint

[9] Figure 4 shows the correlation plot for all the data between NCEP/NCAR reanalysis II and NARL Sodar both for zonal and meridional cases respectively. A reasonable correlation of 0.74 with a standard deviation of 1.24 ms−1 is observed for the available 1341 profiles in the zonal case. Whereas, in the meridional case a correlation of 0.61 with a standard deviation of 1.1 ms−1 is observed for the total 781 profiles. Low correlation is observed in the meridional case than the zonal. During the months of October, November and December data from the NCEP/NCAR Reanalysis II is not available for the meridional component. During the monsoon season easterly winds are dominant over this station. Is the flow direction responsible for the poorer correlation or the data redundancy? During the pre-monsoon season north-easterly winds are dominant over this station, data is available from both the systems at that time and data is included in the correlation analysis of both zonal and meridional. Hence flow direction is not responsible for the poorer correlation in the meridional case. The range of values observed between both the wind components is also different. Zonal values were reported almost twice larger than the meridional component, so this might be one of the reasons for the smaller correlation coefficient in the meridional case. Many research applications relating to complex terrain such as horizontal mass humidity transport in the Atmospheric Boundary Layer, modeling of pollutants advection and dispersion are very important in Lower ABL. Wind flow is the main source for the transport of pollutants, heat flux, water vapor and momentum in the horizontal direction. Remote sensors which give continuous observations of wind data in the lower ABL are very sparsely available in many complex terrain environments over the globe. NCEP/NCAR Reanalysis winds whose data represents a spatial domain of 250 km have shown a good correlation with the NARL Doppler Sodar in the Lower ABL. Even in this complex type of environment a reasonable correlation of 0.74 in zonal and 0.61 in meridional cases from both the systems prove that NCEP/NCAR Reanalysis winds are reliable for using them in research applications for the ABL and pollution dispersion models.

image

Figure 4. Correlation plot of NARL Sodar and NCEP Reanalysis II data for the total observation period: (a) zonal wind component and (b) meridional wind component.

Download figure to PowerPoint

4. Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Site, System Description and Data Used
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[10] NARL Sodar observations with high vertical and temporal resolutions have been compared with 2.5° × 2.5° grid NCEP/NCAR Reanalysis II wind data. One year long sodar measurements have been used to compute 6 hourly averaged zonal and merdional winds at an height of 120 m and are compared with NCEP/NCAR Reanalysis II data of Zonal and Meridional components at a pressure level of 1000 mb in the grid of 12.5°–15°N and 77.5°–80°E. Complex terrain influence on the NCEP/NCAR Reanalysis II data in the lower ABL during stable and unstable conditions have been investigated. Higher correlation is observed in unstable conditions than the stable conditions for both the wind components. This infers that terrain influence is higher on the winds derived during stable conditions due to the slope-valley circulations and the mountain-wave activity. Correlation of 0.74 has been observed in the zonal and a little lower of 0.61 has been observed in the meridional case for the total data studied. Despite the complex terrain environment of the station surrounded by a block of hillocks, measurements have shown good comparison. This study shows that NCEP/NCAR Reanalysis winds are reliable for using in various research applications in the Lower ABL in complex terrain environments.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Site, System Description and Data Used
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[11] The authors would like to thank the NOAA group for providing the NCEP Reanalysis data by the NOAA/OAR/ESRL/PSD, Boulder, Colorado, USA, from their Website http://www.cdc.noaa.gov/. The authors also would like to thank two reviewers for their constructive comments on the manuscript.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Site, System Description and Data Used
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Site, System Description and Data Used
  5. 3. Results and Discussion
  6. 4. Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
grl25399-sup-0001-t01.txtplain text document0KTab-delimited Table 1.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.