The role of each nonlinear term in the vorticity and divergence equations derived from the primitive equations to the steady nonlinear Ekman pumping is examined for uni-directional flows with a small Rossby number comprehensively. For uni-directional flows with a constant horizontal shear, the perturbation analysis in the first order of a small Rossby number show that the contribution from the nonlinear divergence term in the vorticity equation to the Ekman pumping is the most dominant. That from the vertical advection term in the vorticity equation is the second dominant. The nonlinear terms from the divergence equation are less important for the nonlinear corrections to the classical Ekman pumping. In addition, the effect of the nonlinear divergence term to the Ekman pumping is opposite from that of the other nonlinear terms. The latter result also can be verified by qualitative discussions. Furthermore, the analytical solutions for constant horizontal shear flows can be utilized even for uni-directional flows with general horizontal shear. In this case, it is found that the contribution from the nonlinear divergence and horizontal advection terms in the vorticity equation to the Ekman pumping is the most dominant. That from the vertical advection and tilting terms in the vorticity equation is the second dominant. Some comments on the pre-existing approximate models of the nonlinear Ekman pumping are made.
The motion of binary cyclonic vortices with equal structure and strength in no environmental flow is investigated with a focus on what determines their merger or separation. For this, proposals in previous studies on the critical separation distance of binary vortices are validated using a nondivergent barotropic model on an f-plane. Results from numerical experiments indicate that the sign of the initial relative vorticity in the mid-region between two vortices largely determines their merger or separation. This is consistent with the proposal of Falkovich et al. When the initial relative vorticity in the mid-region between two vortices is negative, the two vortices separate. However, it is found that an initial positive relative vorticity between the two vortices does not guarantee their merger. Only when the initial positive relative vorticity exceeds a certain threshold value, the two vonices merge. Based upon the results of numerical experiments, it is suggested that the critical separation distance of binary vortices is slightly smaller than twice the radius at which the relative vorticity of one vortex becomes zero. The merger or separation of binary vortices can be conceptually explained by the advection of the symmetric relative vorticity of one vortex by the symmetric tangential flow of the other vortex. This relative vorticity advection produces vorticity anomalies with different magnitudes to the north and south of each vortex and results in net eastward or westward secondary flow near its center when the two vortices are initially placed in the east-west direction.
This study explores sub-monthly variation for the circulation and convection in the western North Pacific during July and August. The sub-monthly variability is dominated by a wave-like pattern that propagates north-northwestward from the northeast of Papua New Guinea, to the East China Sea. The wavelength and phase speed are about 4000 km and 5 m s−1, respectively. This feature fluctuates in a 7-30 day timescale, with the spectral peak at 12.5 days. The vertical structure exhibits a barotropic structure below 500 hPa, which gradually tilts northwestward in upper levels near Taiwan and southern Japan. For more than 70% of cases which reveal the wave-like fluctuation, recurving tropical cyclones are found to occur concurrently with the wave-like pattern. These recurving tropical cyclones are likely part of the wave-like pattern, and may not be regarded as isolated vortices. The wave-like pattern occurs in a large-scale flow, which is characterized by an enhanced monsoon trough extending eastward into the Philippine Sea, and a strong ridge protruding westward from the subtropical high. It is suggested that the cyclonic phase of the wave-like pattern provides a favorable environment for the genesis, growth, and movement of the recurving tropical cyclones. The monsoon trough/subtropical ridge fluctuation, the wave-like pattern, and tropical cyclones, which represent phenomena in three different time and spatial scales, form a multi-scale system that affects the sub-monthly variability of the circulation, and convection, in the western North Pacific during the boreal summer.
A new long-term reanalysis project, JRA-25, has set as one of its main goals the realistic representation of tropical cyclones (TCs) in the reanalysis. To supplement in situ observations near TCs, wind profile data (TCR data) are retrieved based on best track information and assimilated. This paper addresses the benefits of using TCR data for the analysis of TCs. TC representation in the JRA-25 reanalysis is compared with other reanalysis data, and an experimental reanalysis (Control) without TCR data. The general result is that JRA-25 successfu11y represents the location and intensity of each TC. Among TC basins in the Northern Hemisphere, the TC representation in the JRA-25 data in the eastern North Pacific, and in the tropical Atlantic, where upper observations are sparse, is superior compared to other datasets that did not assimilate TCR data. In the western North Pacific, especially around Japan and the East China Sea, TCR data give little improvement in TC representation, since conventional upper observations are dense there and upstream of the active TCregtons. TC detection rates in the Northern basins and over the entire globe, are computed using an objective procedure from the reanalysis datasets. The rate in JRA-25 is the highest in all basins among the datasets, and consistent through the period. The high qualified TC representation in intensity and location has a positive effect on a flow field, and hydrologic cycle around TCs. Case studies of TC track forecasts for a recurving TC, in which reanalysis data are used as initial conditions, suggest the JRA-25 data are more realistic than the Control. For these reasons, TCR data is effective in representing TCs and sur rounding atmospheric conditions in JRA-25.
We examine the interannual variability of the sea ice area in the Sea of Okhotsk in terms of surface heat flux, using highly resolved flux data with the ice concentration taken into account. On the northwest and east Sakhalin shelves, where the initial ice formation occurs, the onset of ice formation is simply determined by the local heat flux in fall (October-November); the degree to which the ocean is cooled by the atmosphere. Consequently, this heat flux mostly determines the interannual variability of sea ice area in the Okhotsk Sea in the initial stage. The air temperature anomaly is the main cause of the heat flux anomaly in fall. The fall heat flux has persistently affected the ice area anomaly, particularly until mid-January. In the later sea ice season, the relationship between the ice area and heat flux is obscured, particularly after mid-January, by the heat insulating effect of sea ice. The ice area variabilities in the southern and northeast regions are correlated with the local heat flux in the months preceding ice appearance, but not strongly. Our analysis suggests that the heat flux does not determine to what extent the ice advances finally in the Okhotsk Sea. Typical heat flux distributions in heavy and light ice years are also presented from our heat flux data: the heat loss center is located in the central Okhotsk Sea in light ice years, and moves to the eastern Okhotsk Sea near the Kuril Straits in heavy ice years.
A numerical investigation of the temperature variation in urban and suburban areas due to the presence of buildings was carried out using a one-dimensional canopy model combined with a meso-scale meteorological model. Since temperature increases in an urban area are caused by sensible heat from building surfaces besides anthropogenic heat and reduction of wind speed due to buildings’ drag, we estimated each cause separately to determine the contribution by each to the temperature variation. The simulation was performed for Kanda, an urban area, and for Nerima, a suburban area of Tokyo. Comparisons were made with actual temperatures before the estimation. The comparison indicated that the measured temperatures in the Kanda and Nerima areas were nearly reproduced by the model. The sensitivity analysis indicated that, in a comparison with the temperature in no building environment, the contribution of (i) sensible heat flux from building surfaces to temperature rise was 49% in Kanda and 20% in Nerima, (ii) wind reduction due to drag was 41% in Kanda and 59% in Nerima, and (iii) the effect of the interaction between (I) and (II) was +10% in Kanda and +20% in Nerima.
Atmospheric particles were collected in southwestern Japan during dust storm events in spring 2000, and the samples were treated with water-dialysis. In this study, size change of dust particles before and after dialysis is discussed with respect to their sea salt and sulfur contents. It is confirmed that size increase of dust particles has a strong correlation with their sea salt content but is independent from their non-sea-salt sulfur content. This suggests that the growth of dust particles in size during their dispersion in the marine atmosphere is dominated by the combination with sea salt rather than by other processes such as surface uptake of sulfate.