Search for: All records

Abstract In convective quasi-equilibrium theory, tropical tropospheric temperature perturbations are expected to follow vertical profiles constrained by convection, referred to as A-profiles here, often approximated by perturbations of moist adiabats. Differences between an idealized A-profile based on moist-static energy conservation and temperature perturbations derived from entraining and nonentraining parcel computations are modest under convective conditions—deep convection mostly occurs when the lower troposphere is close to saturation, thus minimizing the impact of entrainment on tropospheric temperature. Simple calculations with pseudoadiabatic perturbations about the observed profile thus provide useful baseline A-profiles. The first EOF mode of tropospheric temperature (TEOF1) from the ERA-Interim and AIRS retrievals below the level of neutral buoyancy (LNB) is compared with these A-profiles. The TEOF1 profiles with high LNB, typically above 400 hPa, yield high vertical spatial correlation (∼0.9) with A-profiles, indicating that tropospheric temperature perturbations tend to be consistent with the quasi-equilibrium assumption where the environment is favorable to deep convection. Lower correlation tends to occur in regions with low climatological LNB, less favorable to deep convection. Excluding temperature profiles with low LNB significantly increases the tropical mean vertical spatial correlation. The temperature perturbations near LNB exhibit negative deviations from the A-profiles—the convective cold-top phenomenon—with greater deviation for higher LNB. In regions with lower correlation, the deviation from A-profile shows an S-like shape beneath 600 hPa, usually accompanied by a drier lower troposphere. These findings are robust across a wide range of time scales from daily to monthly, although the vertical spatial correlation and TEOF1 explained variance tend to decrease on short time scales. 

Tropical rainfall variability is closely linked to meridional shifts of the Intertropical Convergence Zone (ITCZ) and zonal movements of the Walker circulation. The characteristics and mechanisms of tropical rainfall variations on centennial to decadal scales are, however, still unclear. Here, we reconstruct a replicated stalagmite-based 2,700-y-long, continuous record of rainfall for the deeply convective northern central Indo-Pacific (NCIP) region. Our record reveals decreasing rainfall in the NCIP over the past 2,700 y, similar to other records from the northern tropics. Notable centennial- to decadal-scale dry climate episodes occurred in both the NCIP and the southern central Indo-Pacific (SCIP) during the 20th century [Current Warm Period (CWP)] and the Medieval Warm Period (MWP), resembling enhanced El Niñ o -like conditions. Further, we developed a 2,000-y-long ITCZ shift index record that supports an overall southward ITCZ shift in the central Indo-Pacific and indicates southward mean ITCZ positions during the early MWP and the CWP. As a result, the drying trend since the 20 th century in the northern tropics is similar to that observed during the past warm period, suggesting that a possible anthropogenic forcing of rainfall remains indistinguishable from natural variability. 

This study used both observations and global climate model simulations to investigate the characteristics of winter extreme snowfall events along the coast (the Interstate 95 corridor) of the northeast United States where several mega-cities are located. Observational analyses indicate that, during 1980–2015, 110 events occurred when four coastal cities—Boston, New York City, Philadelphia, and Washington, D.C.—had either individually or collectively experienced daily snowfall exceeding the local 95th percentile thresholds. Boston had the most events, with a total of 69, followed by 40, 36, and 30 (moving southward) in the other three cities. The associated circulations at 200 and 850 hPa were categorized via K-means clustering. The resulting three composite circulations are characterized by the strength and location of the jet at 200 hPa and the coupled low pressure system at 850 hPa: a strong jet overlying the cities coupled with an inland trough, a weak and slightly southward shifted jet coupled with a cyclone at the coast, and a weak jet stream situated to the south of the cities coupled with a cyclone over the coastal oceans. Comparative analyses were also conducted using the GFDL High Resolution Atmospheric Model (HiRAM) simulation of the same period. Although the simulated extreme events do not provide one-to-one correspondence with observations, the characteristics nevertheless show consistency notably in total number of occurrences, intraseasonal and multiple-year variations, snow spatial coverage, and the associated circulation patterns. Possible future change in extreme snow events was also explored utilizing the HiRAM RCP8.5 (2075–2100) simulation. The analyses suggest that a warming global climate tends to decrease the extreme snowfall events but increase extreme rainfall events.

 

We incorporate a parameterization to quantify the effect of three‐dimensional (3‐D) radiation‐topography interactions on the solar flux absorbed by the surfaces, including multiple reflections between surfaces and differences in sunward/shaded slopes, in the Community Climate System Model version 4 (CCSM4). A sensitivity experiment is carried out using CCSM4 with the prescribed sea surface temperature for year 2000 to investigate its impact on energy budget and surface temperature over the Tibetan Plateau (TP). The results show that the topographic effect reduces the upward surface shortwave flux and, at the same time, enhance snowmelt rate over the central and southern parts of TP. Comparing to observations and the ensemble of Coupled Model Intercomparison Project Phase 5 (CMIP5), we found that CMIP5 models have a strong cold bias of 3.9 K over TP, partially induced by the strong reflection of shortwave fluxes. We show that the inclusion of topographic effect reduces the substantial biases of upward shortwave fluxes and surface air temperatures over TP by 13% in the CCSM4 model.