Accesso libero

Facilitating long-term 3D sonic anemometer measurements in hemiboreal forest ecosystems

Steffen M. Noe
Steffen M. Noe
, 
Alisa Krasnova
Alisa Krasnova
, 
Dmitrii Krasnov
Dmitrii Krasnov
, 
H. Peter
H. Peter
, 
E. Cordey
E. Cordey
 e 
Ahto Kangur
Ahto Kangur
   | 04 giu 2022
Forestry Studies's Cover Image
INFORMAZIONI SU QUESTO ARTICOLO
Articolo precedente
Articolo Successivo
Cita
Article Category: Research paper
Pubblicato online: 04 giu 2022
Pagine: 140 - 149
Ricevuto: 30 dic 2021
Accettato: 31 dic 2021
DOI: https://doi.org/10.2478/fsmu-2021-0016
Parole chiave
© 2021 Steffen M. Noe et al., published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1

Logic decision scheme to provide the intermittent heating capacity of sensor heads of 3D ultrasonic anemometers at SMEAR Estonia. The system default is heater off and it switches to heater on only if the temperature is below the set threshold temperature ts and if the number of errors exceed the set error count ec for a given time period.
Logic decision scheme to provide the intermittent heating capacity of sensor heads of 3D ultrasonic anemometers at SMEAR Estonia. The system default is heater off and it switches to heater on only if the temperature is below the set threshold temperature ts and if the number of errors exceed the set error count ec for a given time period.

Figure 2

Temperature measured at 2 m height at the SMEAR Estonia measurement station from August 2014 until October 2019 and the mean long-term Estonian temperature record from the Estonian Weather Service (www.ilmateenistus.ee/kliima/kliimanor-mid/?lang=en). Data shown for SMEAR Estonia are averages over 10 minutes.
Temperature measured at 2 m height at the SMEAR Estonia measurement station from August 2014 until October 2019 and the mean long-term Estonian temperature record from the Estonian Weather Service (www.ilmateenistus.ee/kliima/kliimanor-mid/?lang=en). Data shown for SMEAR Estonia are averages over 10 minutes.

Figure 3

Ice formation probability density functions per instrumentation height and combined data from all instrumentation heights. A skewed mixture distributions of the icing temperatures are observable at all instrumentation heights. Most had at minimum two peaks. Detailed peak information is presented in Table 1.
Ice formation probability density functions per instrumentation height and combined data from all instrumentation heights. A skewed mixture distributions of the icing temperatures are observable at all instrumentation heights. Most had at minimum two peaks. Detailed peak information is presented in Table 1.

Figure 4

Probability density functions of data combined from all instrumentation heights and split by ice forming months. The PDF for November has a 4-peak shape. Single-peaked PDFs were found for October, December, March, and April while January and February were double-peaked. Detailed peak information is presented in Table 2.
Probability density functions of data combined from all instrumentation heights and split by ice forming months. The PDF for November has a 4-peak shape. Single-peaked PDFs were found for October, December, March, and April while January and February were double-peaked. Detailed peak information is presented in Table 2.

Figure 5

Box-Whisker chart for the monthly distribution of the median time interval of sensor head heating of the 3D ultrasonic anemometers. The median length of heating is at about 25 to 30 seconds; the longest heating intervals were found in April with 32 seconds and the shortest in November with 18 seconds. The minimum heating interval for all data analysed was 3 seconds and the maximum was 72 seconds. The boxes’ width is scaled related to the number of datapoints in each month.
Box-Whisker chart for the monthly distribution of the median time interval of sensor head heating of the 3D ultrasonic anemometers. The median length of heating is at about 25 to 30 seconds; the longest heating intervals were found in April with 32 seconds and the shortest in November with 18 seconds. The minimum heating interval for all data analysed was 3 seconds and the maximum was 72 seconds. The boxes’ width is scaled related to the number of datapoints in each month.

To assess probable icing temperatures per month from measured data we used mixed distributions limited to be combinations of normal distributions N(μ, σ). The Pearson χ2 criterion denotes the probability that the data could have been drawn from the distribution.

Month Temperature °C Pearson χ2 p-value Distribution
October 0.6 0.88 N(0.642105, 1.26088)
November −6.9 0.99 N(−6.9067, 2.52087)
−2.3 N(−2.28275, 0.646585)
0 N(0.0306805, 0.759967)
1.6 N(1.65582, 0.205358)
December −3.1 0.84 N(−3.12821, 3.49842)
January −18.2 0.079 N(−18.2457, 4.54108)
−3.8 N(−3.84678, 3.97783)
February −9.5 0.56 N(−9.56731, 2.44771)
−0.9 N(−0.921769, 2.28932)
March −1.4 0.22 N(−1.42623, 2.60906)
April −0.2 0.90 N(−0.2, 1.34759)

To determine the most probable icing temperatures from measured data we used mixed distributions limited to be combinations of normal distributions N(μ, σ). The Pearson χ2 criterion denotes the probability that the data could have been drawn from the distribution.

Height m Tempe-rature °C Pearson χ2 p-value Distribution
110 −21.2 0.73 N(−21.2062, 0.353645)
−6.4 N(−6.92617, 0.409879)
−0.6 N(−0.38294, 1.95312)
90 −22.3 0.04 N(−22.3468, 2.72057)
−3.7 N(−3.71077, 3.79747)
70 −11.1 0.91 N(−11.1222, 7.40277)
−0.4 N(−0.373057, 2.2027)
50 −10.1 0.54 N(−10.1285, 5.72445)
−1.3 N(−1.26539, 1.89357)
30 −19.6 0.90 N(−19.5834, 4.00944)
−7.9 N(−7.93453, 2.19069)
−1.3 N(−1.24393, 1.63284)
all −9.4 0.00047 N(−9.55811, 6.68105)
−1.2 N(−1.12501, 2.04707)
eISSN:
1736-8723
Lingua:
Inglese
Frequenza di pubblicazione:
2 volte all'anno
Argomenti della rivista:
Life Sciences, Plant Science, Ecology, other
Feed RSS della rivista