[1] Eddy covariance measurements of sensible (H) and latent (LE) heat fluxes were made over a large southern open water surface of Ross Barnett Reservoir (the Reservoir hereafter) in Mississippi during the cool season with frequent incursions of cold fronts from 1 September 2007 to 31 January 2008. The eddy covariance tower was located in the middle of the main body of the Reservoir with the tower fetches exceeding 2.0 km in all directions. The Reservoir was ice-free in winter and the water temperatures always decreased with depth. Over the entire cool season, the averaged water surface temperatures were 1.8°C higher than the overlying air (i.e., positive temperature gradients that led to thermally convective conditions) and the averaged vapor pressure near the water surface was 0.8 kPa greater than the overlying air (i.e., positive vapor pressure gradients), though occasionally negative gradients for temperature and vapor pressure were also observed for short periods. On average, the wind speeds were considerably large (3.9 m s À1 ) to maintain adequate turbulent mixing mechanically. As a consequence of the combined effect of thermally and mechanically generated turbulent mixing, consistently positive H (with a mean H of 20.0 W m À2 ) and LE (with a mean LE of 80.0 W m À2 ) occurred during the entire season. These continuous energy losses via H and LE resulted in release of a large amount of energy stored in the water to the atmosphere. The mean Bowen ratio was low for this open water surface (i.e., 0.3), suggesting that most of the energy released from the water fueled evaporation rather than sensible heating of the atmosphere. Nighttime evaporative water losses were substantial, contributing to 45% of the total evaporative water loss in this cool season. Frequent incursions of cold fronts with windy, cold, and dry air masses significantly promoted turbulent exchanges of sensible and latent heat through enhanced turbulent mixing thermally and mechanically, leading to large H and LE events. Daily H and LE (i.e., evaporation) during the passages of cold fronts were on average 2.7 and 7.3 times those in nonevent days, respectively. Given the fact that large H and LE events occurred 26% of the time for our site, these cold front events caused an increase in the seasonal H by 42% and LE by 157%. Therefore changes in frequency, intensity, and duration of synoptic weather events, particularly the incursions of cold fronts, have significant impacts on the surface energy budget and evaporation over water at this site.

In this study, the rainfall asymmetries in tropical cyclones (TCs) that made landfall in the Hainan (HN), Guangdong (GD), Fujian (FJ), and Zhejiang (ZJ) provinces of mainland China and Taiwan (TW) from 2001 to 2009 were analyzed on the basis of TRMM satellite 3B42 rainfall estimates. The results reveal that in landfalling TCs, the wavenumber 1 rainfall asymmetry shows the downshear to downshear-left maximum in environmental vertical wind shear (VWS), which is consistent with previous studies for TCs over the open oceans. A cyclonic rotation from south China to east China in the location of the rainfall maximum has been identified. Before landfall, the location of the rainfall maximum rotated from southwest to southeast of the TC center for TCs making landfall in the regions from HN to GD, TW, FJ, and ZJ. After landfall, the rotation became from southwest to northeast of the TC center from south China to east China. It is shown that this cyclonic rotation in the location of the rainfall maximum is well correlated with a cyclonic rotation from south China to east China in the environmental VWS between 200 and 850 hPa, indicating that the rainfall asymmetry in TCs that made landfall over China is predominantly controlled by the large-scale VWS. The cyclonic rotation of VWS is found to be related to different interactions between the midlatitude westerlies and the landfalling TCs in different regions. The results also indicate that the axisymmetric (wavenumber 0) component of rainfall generally decreased rapidly after landfall in most studied regions.

TRMM satellite 3B42 rainfall estimates for 133 landfalling tropical cyclones (TCs) over China during 2001–15 are used to examine the relationship between TC intensity and rainfall distribution. The rain rate of each TC is decomposed into axisymmetric and asymmetric components. The results reveal that, on average, axisymmetric rainfall is closely related to TC intensity. Stronger TCs have higher averaged peak axisymmetric rain rates, more averaged total rain, larger averaged rain areas, higher averaged rain rates, higher averaged amplitudes of the axisymmetric rainfall, and lower amplitudes of wavenumbers 1–4 relative to the total rainfall. Among different TC intensity change categories, rapidly decaying TCs show the most rapid decrease in both the total rainfall and the axisymmetric rainfall relative to the total rain. However, the maximum total rain, maximum rain area, and maximum rain rate are not absolutely dependent on TC intensity, suggesting that stronger TCs do not have systematically higher maximum rain rates than weaker storms. Results also show that the translational speed of TCs has little effect on the asymmetric rainfall distribution in landfalling TCs. The maximum rainfall of both the weaker and stronger TCs is generally located downshear to downshear left. However, when environmental vertical wind shear (VWS) is less than 5 m s−1, the asymmetric rainfall maxima are more frequently located upshear and onshore, suggesting that in weak VWS environments the coastline could have a significant effect on the rainfall asymmetry in landfalling TCs.

To evaluate the abilities of satellite retrievals in reflecting precipitation features related to tropical cyclones (TCs) affecting mainland China, four years of 6-and 24-h precipitation retrievals from three datasets, namely the Tropical Rainfall Measuring Mission satellite algorithm 3B42, version 6 (3B42), Climate Prediction Center morphed (CMORPH) product, and one based on the Geostationary Meteorological Satellite-5 infrared brightness temperature (GMS5-TBB), are compared statistically with direct measurements from surface gauge rainfall data during the periods affected by TCs. The GMS5-TBB dataset was set up by a method of considering the GMS5-TBB characteristics, hourly precipitation intensity, and horizontal distribution for landfalling TCs. The results show that in a general sense, all three satellite-retrieved rainfall datasets give quite reasonable 6-and 24-h rainfall distributions, with skill decreasing with the increase in both latitude and rainfall amount. The 3B42 has a little bit better skill than CMORPH, which is likely related to the fact that the 3B42 product has a rain gauge adjustment and CMORPH does not. Further analyses show that both 3B42 and CMORPH considerably underestimate the moderate and heavy rainfall and overestimate the very light precipitation. The overestimation of the GMS5-TBB data for the light rain is larger than that for 3B42 and CMORPH, probably due to the fact that the GMS5-TBB method considers stratiform and convective rainfall separately with a fixed stratiform rain rate of 2 mm h 21 . For the heavy rainfall events, the GMS5-TBB data perform much better than the 3B42 and CMORPH data with an almost halved bias, owing to the fact that the GMS5-TBB method adopted the adjustment of the convective rain rate by considering TBB characteristics of landfalling TCs and using hourly gauge rainfall in the setup process. Since the heavy rainfall events associated with landfalling TCs are of the most concern, the compared GMS5-TBB data could be useful as an operational/research reference.

Typhoon Fitow made landfall south of Shanghai, China, on 6 October 2013. During the following two days, precipitation in excess of 300 mm day−1 occurred 400 km to the north of the typhoon center. The rain-producing systems included (i) outward-spiraling rainbands, which developed in the storm’s north sector in favorable environmental wind shear, and (ii) frontal cloud as a result of coastal frontogenesis. Over the rain area, in addition to enhanced ascent, there were increases in low-level moisture, convective instability, and midlevel relative vorticity. There is evidence of a preconditioning period prior to the rain when midlevel subsidence and boundary layer moistening occurred. From analysis of low-level equivalent potential temperature the following observations were made: (i) after landfall, a cold, dry airstream wrapped into Fitow’s circulation from the north, limiting the inner-core rainfall and producing a cold-air boundary, and (ii) an extended warm, moist airstream from the east converged with the cold-air intrusion over the rain area. The heavy rain occurred as the large-scale flow reorganized. Major anticyclones developed over China and the North Pacific. At upper levels, a large-amplitude trough relocated over central China with the entrance to a southwesterly jet positioned near Shanghai. Back trajectories from the rain area indicate that four environmental interactions developed: (i) increasing midlevel injection of moist potential vorticity (PV) from Fitow’s circulation; (ii) low-level warm, moist inflow from the east; (iii) midlevel inflow from nearby Typhoon Danas; and (iv) decreasing mid- to upper-level injection of PV from the midlatitude trough. The authors propose that the resultant PV structure change provided a very favorable environment for the development of rain systems.