Cardington, England (Angell. 1964). However. the 10-min period obtained
w hcn the tetroons pass across Oklahoma City appears rather small. possibly
pointing up subtle irrhiin inlluenws on the Laprangian period of oscillation
as well.
Thc right-hand diagram of Fig. 4 illustrates the mean variation with
height of Lagrangian and Eulerian spectral-peak frequencies. 8s estimated
from tctroon flights iicross the city and from tower data downwind of the
city. I n both cases the dominant period of oscillation increases with incrcasing height, hut equally so, implying little change in /{ with height. Because
the mean tctroon hcight was 35 m greater than the anemometer height of
355 m, the mean value of /I for the urban sample is really I2 rather than 13.
In the following. the individual Lagrnngian spectral-peak frequency values
have hecn "normalized" to ;I height of 355 m by assuming a variation with
height given by the L line in Fig. 4. This. of course, also results in the
normalization of /{ to the 355-m level.
3. VAHIArION Of p WlTtl MET~~OHOI.OGICAI. PARAMETERS
Figure 5 illustrates the variation of /j with wind spced (measured at ;I
height of 355 m on the tower), as determined from indikidual Lagrangian
Euleriiin spectral comparisons. When the tower is downwind of the city iind
thc tctroons pass over the city, the trace is labeled "urhan". whereas when
tho tout'r is upwind of the city and the tetroons do not pass over the city the
trilcc is labeled "rural." In both cases. /I increases with increase in wind
specd. ;I S would be anticipated from the well-known shift of the Euleriiin
SpcctriII pcak towitrd higher frequency with increase in wind speed (the
turhulcncc clements are transported more rapidly past the fixed point in
stroiig winds). and the observation that the frequency of the Lagrangian
SpCCtIill pcak is almost indepcndcnt of wind speed (being related more to
ntniosphcric stability).
Note that while there is a discontinuous jump in the ahsolute value of/)
hetween urhiin and rural areas. the percentage change in p per unit
( I in sec I ) increase in wind speed remains nearly the same, with average
ViilUcS of I1 and 13 9;, in urban and rural areas, respectively. Thus, the
iiicrcasd surface roughness of the urban areas does not appreciably affect
tlic w h i w variation of /{ with wind speed. The small circles in Fig. 5 represent our estimate of the Actual mean values of /I for wind speeds exceeding
10 ni scc I , obtaincd on the assumption that the derived rural values of /{
are representative iind that the percentage change in /I (per unit increment of
wind speed) is 12 I,:;,. With this assumption. is indicated to vary from u
mean value of 2 at a wind specd of 6 m sec- ' to ii mean value of 6 at a wind
spt.ed of 15 ni scc '. Thc letters in Fig. 5 show the mean values of/Iat given
w hcn the tetroons pass across Oklahoma City appears rather small. possibly
pointing up subtle irrhiin inlluenws on the Laprangian period of oscillation
as well.
Thc right-hand diagram of Fig. 4 illustrates the mean variation with
height of Lagrangian and Eulerian spectral-peak frequencies. 8s estimated
from tctroon flights iicross the city and from tower data downwind of the
city. I n both cases the dominant period of oscillation increases with incrcasing height, hut equally so, implying little change in /{ with height. Because
the mean tctroon hcight was 35 m greater than the anemometer height of
355 m, the mean value of /I for the urban sample is really I2 rather than 13.
In the following. the individual Lagrnngian spectral-peak frequency values
have hecn "normalized" to ;I height of 355 m by assuming a variation with
height given by the L line in Fig. 4. This. of course, also results in the
normalization of /{ to the 355-m level.
3. VAHIArION Of p WlTtl MET~~OHOI.OGICAI. PARAMETERS
Figure 5 illustrates the variation of /j with wind spced (measured at ;I
height of 355 m on the tower), as determined from indikidual Lagrangian
Euleriiin spectral comparisons. When the tower is downwind of the city iind
thc tctroons pass over the city, the trace is labeled "urhan". whereas when
tho tout'r is upwind of the city and the tetroons do not pass over the city the
trilcc is labeled "rural." In both cases. /I increases with increase in wind
specd. ;I S would be anticipated from the well-known shift of the Euleriiin
SpcctriII pcak towitrd higher frequency with increase in wind speed (the
turhulcncc clements are transported more rapidly past the fixed point in
stroiig winds). and the observation that the frequency of the Lagrangian
SpCCtIill pcak is almost indepcndcnt of wind speed (being related more to
ntniosphcric stability).
Note that while there is a discontinuous jump in the ahsolute value of/)
hetween urhiin and rural areas. the percentage change in p per unit
( I in sec I ) increase in wind speed remains nearly the same, with average
ViilUcS of I1 and 13 9;, in urban and rural areas, respectively. Thus, the
iiicrcasd surface roughness of the urban areas does not appreciably affect
tlic w h i w variation of /{ with wind speed. The small circles in Fig. 5 represent our estimate of the Actual mean values of /I for wind speeds exceeding
10 ni scc I , obtaincd on the assumption that the derived rural values of /{
are representative iind that the percentage change in /I (per unit increment of
wind speed) is 12 I,:;,. With this assumption. is indicated to vary from u
mean value of 2 at a wind specd of 6 m sec- ' to ii mean value of 6 at a wind
spt.ed of 15 ni scc '. Thc letters in Fig. 5 show the mean values of/Iat given