Environ. Res. Lett. 10 (2015) 064011 I Takayabu et al Figure 4. (a) Radar reflectivity at z = 2 km in the m02 (1001) experiment with the 1 km WRF model (dBZ) (11 UTC on 11 August 2014). (b) Radar reflectivity data for Guiuan station (dBZ), elevation angle 1° (0416 UTC on 11 August 2014). To evaluate how well the experiments could reproduce Typhoon Haiyan, we used the results of experiment m02 (#1001), in which the simulated track was nearest to the observed track (best track). The structure of the eye-wall clouds was represented by cascade downscaling to a 1 km grid scale. Because the inner core convective band just around the eye is the engine of a typhoon, it is necessary to represent the structure of the eye-wall clouds for the sufficient intensification of a typhoon. Compared with weaker TCs, however, the radius of the eye of category 4 or 5 typhoons becomes smaller (about 20 km) (Weatherford and Gray 1988). Therefore, to represent the structure of the eye-wall clouds, the model must have a horizontal resolution of less than 10 km, and 5 use of a model with a 2 km or 1 km grid is better because the simulation can be performed without a cumulus convective scheme (Gentry and Lackmann 2010, Kanada et al 2012). Supplementary figure S1 depicts the convective rain bands within and around the TC core in the 20, 5 and 1 km models. In the coarser resolution models, the eye-wall structure appears smoothed. To simulate the intensity of the disturbance better, the structure within the TC’s core must be sufficiently resolved. Comparison of our simulation results (experiment m02 (#1001)) with the radar reflectivity data for Typhoon Haiyan obtained by Guiuan station (figure 4) showed that WRF 1 km was able to reproduce the structure around the eye-wall clouds well.