Corrigendum: Ultra-high-resolution numerical weather prediction with a large domain using the k computer: A case study of the Izu Oshima heavy rainfall event on October 15-16, 2013 (J. Meteor. Soc. Japan, (2018) 96, (25-54), 10.2151/jmsj.2018-006)

T. Oizumi, K. Saito, J. Ito, T. Kuroda, L. Duc

Research output: Contribution to journalComment/debate

Abstract

A correction is needed for Oizumi et al. (2018). Page Line 38 71–73 Figure 11 is the same as Fig. 10 except for the valid time which is 0300 JST on October 16than that of CM-500mDD (the dashed red line) below/above 500 m. → Corrected: Figure 11 is the same as Fig. 10 except for the valid time which is 0300 JST on October 16. At this time, in the horizontal distribution of the vertical flow for CM2kmMY (Fig. 11a1), a band-shaped updraft developed along the front. The updraft along the front in CM2kmDD (Fig. 11b1) was further intensified compared to that at 0100 JST, and a line-shaped downdraft appeared on the northwest side of the front. In the vertical cross section in both CM2kmMY (Fig. 11a2) and CM2kmDD (Fig. 11b2), clouds developed at the updraft area along the front (arrows), and rain areas were seen on the northwest side. In Fig. 11b2, the rain area corresponds to the downdraft area on the northwest side of the front. With a finer horizontal grid spacing of 500 m in CM500mMY (Fig. 11c1), the updraft became more distinct as small cells aligned along the front. In CM500mDD (Fig. 11d2), the vertical flow was further intensified and showed more detailed convection cell patterns. As for the clouds and rain (lower panels), similar correspondences between the cloud water, rainwater, updrafts, and downdrafts along the fronts (arrows) can be seen in the two experiments CM500mMY (Fig. 11c2) and CM500mDD (Fig. 11d2). Figure 12 shows the vertical cross section of the potential temperature at 0100 JST on October 16 along the line AB in Figs. 10a1 – 10d1. The arrows in this figure indicate the positions of the simulated front, as do the lines in Fig. 10. In CM2kmMY (Fig. 12a) and CM500mMY (Fig. 12c), the cold pool (292 – 294 K, indicated in blue) is only seen near the Izu Peninsula. Conversely, in CM2kmDD (Fig. 12b) and CM500mDD (Fig. 12d), the cold pool is more distinct and extends to the southeast. CM2kmDD had a thicker cold pool than CM500mDD. CM500mDD had a longer cold pool to the southeast than CM2kmDD. The experiments with DD had stronger cold pools than did the experiments with MYNN. The positional relationship between the cloud water and the rainwater shown in Fig. 10 was seen from the beginning to the end of the simulations. This tendency of the positional relationship indicates that the clouds developed with the updraft at the southeastern edge of the cold pool and the rain appeared on the northwest side of the front, especially in the experiments with DD. This means that the successive formation of new convection cells was reproduced along the front by the DD experiments and that the position of the rainband is likely affected by the position and strength of the cold pool. b. Vertical profiles Figure 13 shows the vertical profiles of the potential temperature (θ), the mixing ratio of the water vapor (qv), the equivalent potential temperature (θe), and the horizontal wind speed of CM500mMY and CM500mDD on the southeastern and northwestern sides of the front. Here on the inflow side, the vertical profiles were obtained via the spatiotemporal average of the warm and moist air in square C in Fig. 5 with horizontal sampling intervals of 2 km. The sampling period was 0000 – 0100 JST on October 16, and the sampling interval was 5 min. In total, 1452 data points (12 temporal and 121 spatial data points) were used for the vertical profiles. The vertical profiles on the northwestern side of the front were obtained in square D in Fig. 5 in the same way. As shown in Fig. 12, the profiles at the lower level reflect the cold air from the Kanto Plain, whereas the air above approximately 500 – 600 m may originate from the warm air, which blew over the lower level cold air. Figure 13a shows the vertical profiles of the potential temperatures in squares C and D. In square C, the potential temperature of CM500mMY (the bold black line) was slightly higher than that of CM500mDD (the bold red line) below 500 m, and the potential temperature of CM500mDD was slightly higher than that of CM500mMY between 500 m and 1.6 km. The vertical gradient of CM500mMY was smaller than that of CM500mDD from 300 m to 1.3 km. These differences are likely caused by the tendency of the MYNN scheme to yield larger mixing in the boundary layer. These results were consistent with previous studies (Kato 2011, 2012). In square D, the potential temperature of CM500mMY (the dashed black line) was greater/smaller than that of CM500mDD (the dashed red line) below/above 500 m. Supplementary: The corrected PDF file of Oizumi et al. (2018).

Original languageEnglish
Pages (from-to)291-292
Number of pages2
JournalJournal of the Meteorological Society of Japan
Volume96
Issue number3
DOIs
Publication statusPublished - 2018 Jan 1
Externally publishedYes

ASJC Scopus subject areas

  • Atmospheric Science

Fingerprint Dive into the research topics of 'Corrigendum: Ultra-high-resolution numerical weather prediction with a large domain using the k computer: A case study of the Izu Oshima heavy rainfall event on October 15-16, 2013 (J. Meteor. Soc. Japan, (2018) 96, (25-54), 10.2151/jmsj.2018-006)'. Together they form a unique fingerprint.

  • Cite this