Open Access

Modified, threshold-based circulation type classification for Central Europe, on the basis of Lityński’s classification

Kinga Kulesza

Kinga Kulesza

Faculty of Geography and Regional Studies, University of WarsawWarsaw, Poland
Search for this author on
Sciendo|Google Scholar
Kulesza, Kinga
  
Jan 31, 2019
Modified, threshold-based circulation type classification for Central Europe, on the basis of Lityński’s classification's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Published Online: Jan 31, 2019
Page range: 53 - 62
Received: Apr 24, 2018
Accepted: Oct 11, 2018
DOI: https://doi.org/10.2478/mgrsd-2018-0033
Keywords
© 2019 Kinga Kulesza, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Figure 1

Example of a spatial domain for the Lityński-based circulation type classification.Source: own elaboration.
Example of a spatial domain for the Lityński-based circulation type classification.Source: own elaboration.

Figure 2

The thresholds for the classes of the indices Wp (a), Ws (b) and Cp (c) for 16 January 1948-2015. ORG_u – the upper threshold for the ORG classification; NEW_u – the upper threshold for the NEW classification; MOD20_u – the upper threshold for the MOD20 classification; ORG_I – the lower threshold for the ORG classification; NEW_I – the lower threshold for the NEW classification; and MOD20_I – the lower threshold for the MOD20 classification.Source: own elaboration.
The thresholds for the classes of the indices Wp (a), Ws (b) and Cp (c) for 16 January 1948-2015. ORG_u – the upper threshold for the ORG classification; NEW_u – the upper threshold for the NEW classification; MOD20_u – the upper threshold for the MOD20 classification; ORG_I – the lower threshold for the ORG classification; NEW_I – the lower threshold for the NEW classification; and MOD20_I – the lower threshold for the MOD20 classification.Source: own elaboration.

Figure 3

The value distribution of the Wp, Ws, and Cp indices for January (JAN), April (APR), July (JUL), and October (OCT) 1986-2015. Vertical dashed lines denote the class thresholds according to the formula x̄ ±0.433σ (the NEW1 variant); vertical continuous lines denote the class thresholds according to the percentiles 33 and 67 (the MOD30 variant). Variant names are explained in the text.Source: own elaboration.
The value distribution of the Wp, Ws, and Cp indices for January (JAN), April (APR), July (JUL), and October (OCT) 1986-2015. Vertical dashed lines denote the class thresholds according to the formula x̄ ±0.433σ (the NEW1 variant); vertical continuous lines denote the class thresholds according to the percentiles 33 and 67 (the MOD30 variant). Variant names are explained in the text.Source: own elaboration.

Figure 4

Deviations (%) of the frequencies of individual classes for the indices Wp (a), Ws (b), and Cp (c) from the theoretical equally-likely frequency (where the frequency of each class equals 33.33%), in individual classification variants. Variant names are explained in the text.Source: own elaboration.
Deviations (%) of the frequencies of individual classes for the indices Wp (a), Ws (b), and Cp (c) from the theoretical equally-likely frequency (where the frequency of each class equals 33.33%), in individual classification variants. Variant names are explained in the text.Source: own elaboration.

Figure 5

Mean absolute error (MAE) for the sub-variants MOD5, MOD10, MOD15, MOD20, MOD25, MOD30, and MOD35 at different periods of time on which the class thresholds are calculated. Variant names are explained in the text.Source: own elaboration.
Mean absolute error (MAE) for the sub-variants MOD5, MOD10, MOD15, MOD20, MOD25, MOD30, and MOD35 at different periods of time on which the class thresholds are calculated. Variant names are explained in the text.Source: own elaboration.

Figure 6

Atmospheric circulation types determined using the ORG and MOD20 classifications in comparison with synoptic charts for selected days at 12:00 UTC. Variant names are explained in the text.Source: own work on the basis of synoptic charts obtained from: Koninklijk Nederlands Meteorologisch Instituut 2018.
Atmospheric circulation types determined using the ORG and MOD20 classifications in comparison with synoptic charts for selected days at 12:00 UTC. Variant names are explained in the text.Source: own work on the basis of synoptic charts obtained from: Koninklijk Nederlands Meteorologisch Instituut 2018.

Mean absolute error (MAE) for the different variants of the classification algorithm_ Variant names are explained in the text_

VariantORGNEWNEW1MOD30
MAB (pp)1.821.051.010.74

The matrix of agreement between the results of individual classification schemes (%)_ Variant names are explained in the text_

ORG
89.3NEW
88.095.6NEW1
84.991.193.0MOD30
80.986.387.689.2MOD20
Language:
English
Publication timeframe:
4 times per year
Journal Subjects:
Geosciences, Geography, Geosciences, other
Journal RSS Feed