Abstract

The climate sensitivity question is examined from the viewpoint of surface energy balance considerations. This approach clarifies the role of ocean-atmosphere interactions in determining the surface warming to an increase in CO₂. The study uses a one-dimensional, 17-layer, coupled ocean-atmosphere model. The primary contribution to the surface warming is from the enhanced tropospheric IR emission, which is an order of magnitude greater than the direct CO₂ radiative heating at the surface. The source for this enhancement is the increased H₂O evaporation from the warmer oceans in the CO₂ rich atmosphere and, hence, ocean-atmosphere interactions play a crucial role in determining the magnitude of the surface warming as well as its transient response.

This ocean-atmosphere feedback is implicitly included in model studies, but past analysts of model results have not highlighted this aspect of the problem. Consequently, published empirical approaches based on surface energy balance considerations (unaware of the ocean-atmosphere feedbacks that contribute to surface warming) have obtained results differing significantly from climate model results. Several experiments are performed with the coupled one-dimensional model to identify the various sources of the discrepancy between the modeling and empirical approaches. The paper also examines the influence of ocean-atmospheric interactions on the transient response of the climate system, which reveals the deficiencies in present schemes of asynchronously coupling the ocean and the atmosphere in three-dimensional climate models.

Abstract

The large increase in polar stratospheric temperatures during sudden warming events results in an enhancement in the downward longwave radiation flux emitted by the stratosphere. It is shown that the enhancement in the downward longwave flux has an appreciable warming effect on the winter polar troposphere and surface and also leads to a reduction in the tropospheric available potential energy. The results of the previous analyses indicate that the stratospheric warming is initiated by the upward flux of geopotential energy from the troposphere. Based on these results it is suggested that the planetary-scale tropospheric waves accomplish part of the poleward transport of heat in the troposphere indirectly through the sudden warming phenomenon. It is also shown that roughly 80–90% of the downward longwave radiative flux emitted by ozone in the polar stratosphere reaches the.polar surface and the importance of this downward ozone flux to the polar surface energy budget is qualitatively discussed. The results indicate the importance of the stratospheric circulation and radiative energy transfer to the tropospheric circulation and energy budget.

Abstract

This paper examines the interactions between ice-albedo, lapse-rate and cloud-top feedbacks with the aid of GCM climate experiments published by Wetherald and Manabe (1975). First we establish that the long-wave modification effect (the so-called “greenhouse effect”) of clouds depends largely on the temperature difference T_sc between surface and cloud tops. If T_sc changes with a change in the surface temperature T_s , then the longwave modification effect of clouds would change which would result in a modification of the initial change in T_s . This feedback between the longwave modification effect of clouds and T_s , is referred to as the cloud top feedback in this paper. The sign of this feedback is considered positive (negative) when it amplifies (decreases) an initial change in T_s and it is shown that the sign is determined by the sign of dT_sc /dT_s .

In the GCM climate experiments of Wetherald and Manabe (1975), dT_sc /dT_s < 0 at low latitudes and dT_sc /dT_s > 0 at high latitudes; consequently, cloud tops exert a positive feedback at high latitudes and a negative feedback at low latitudes. We then demonstrate that the interaction between cloud-top, ice-albedo and lapse-rate feedbacks induces a nonlinear response of surface temperature changes in a GCM climate model.

The significance of the H₂O e-type absorption to the surface energy budget of the tropics is stressed. The results suggest that the e-type absorption will be important in determining the sensitivity of the hydrological cycle to changes in surface temperature.

Abstract

A simplified radiative transfer model for the earth's atmosphere is presented. The simplification is achieved by a combination of band absorptance and emissivity formulation for treating radiative transfer due to H₂O, CO₂ and O₃. The model incorporates the major and minor radiative transfer processes due to H₂O, CO₂ and O₃. The radiative model is used to develop an efficient and accurate radiative-convective model. Results for the global surface temperature, stratospheric thermal structure, and the net outgoing longwave flux are presented.

The computed thermal structure of the stratosphere and the stratospheric cooling rates are in excellent agreement with previous studies. The amplitude of the diurnal temperature difference in the upper stratosphere obtained from the present model is larger by about 50% than Leovy's (1964) results. This difference is due to the inclusion of Doppler broadening effects and CO₂ hot and minor isotopic bands in the present model.

The flux calculations indicate that the relatively minor bands like the CO₂ hot and minor isotopic bands and the e-type absorption by the H₂O continuum band have to be included in order to compute the outgoing flux F to within 1% accuracy. Results are also presented for the sensitivity of F to surface temperature. It is shown that the H₂O e-type absorption has a substantial influence on the sensitivity parameter dF/dT^s .

Abstract

Radiative transfer calculations employing observed values of the ice crystal size distribution demonstrate that the absorption of solar radiation by cirrus clouds can make a significant contribution to the diabatic heating of the upper troposphere. The effects due to this absorption on the upper tropospheric (100–300 mb) thermal profile are investigated in a general circulation model (GCM) with interactive clouds; guided by observations, two experiments are performed assuming vastly different vertical profiles of the ice water density. Solar heating rates within the extensive cirrus decks associated with monsoon and other convective clouds reach values of 1.5 K day⁻¹. Thus, cirrus solar heating can be an important source for east-west asymmetries in the tropical diabatic heating. Furthermore, because of the latitudinal gradients in the solar insolation, cirrus solar absorption can also influence the meridional beating gradients within the upper troposphere.

In spite of the significant east-west asymmetries in the imposed cirrus solar heating, the change in the GCM tropical temperatures is nearly zonally uniform. The magnitude of the zonal mean tropical temperature changes in the GCM (up to 5°K at P ≈ 165 mb) indicate that lack of cirrus solar heating may be one reason for the cold bias of the GCMS. Furthermore, the shortwave beating can also account for the observed lapse rate stabilization in the upper troposphere.

In addition to the solar effect, the longwave radiative effects of cirrus can also be important but their sign and magnitude are very sensitive to the vertical distribution of clouds. Cirrus longwave heating rates can range from large negative values (cooling) when overlying optically thick clouds (for example, in “deep” extended systems with base below the upper troposphere) to large positive values (heating) for “anvil” type cirrus located in the upper troposphere and with no other clouds below. For the overcast portions of the tropics, if “anvil” type cirri are the only clouds of significance in the upper troposphere, the longwave heating would be the dominant radiative component and this effect becomes more pronounced with increasing altitude of cloud location. Hence, for the tropical zone as a whole, the sign and magnitude of the longwave effect depends on the relative composition of the “deep” and “anvil” clouds. Radiation model calculations that employ climatological values of the vertical distribution of clouds yield a longwave heating effect for the cirrus with the magnitude being comparable to the solar effect.

Thus, our results suggest a significant role for the cirrus radiative effects in maintaining the zonal mean thermal structure of the upper troposphere. This inference should be contrasted with the notion that the steep positive gradient in the tropical upper-troposphere potential temperatures is maintained by the latent heat released in penetrating cumulus towers.

Abstract

The primary interest of the present study is to examine the sensitivity of climate to radiative perturbations such as increases in CO₂ and solar insolation for surface temperatures warmer than present day global averaged values (T_s > 290 K). The climate sensitivity, defined here as the change in T_s , is examined with the aid of a one-dimensional radiative-convective model. The solar insolation in the model is varied from 880 to 1840 W m⁻² to obtain a wide range of T_s , from 255 to 325 K. We examine in detail the dependence of the computed ΔT_s , on the following processes which are known to be important in the warmer regions (e.g., tropics) of the present day atmosphere: convective parameterizations (fixed lapse-rate, moist-adiabatic adjustment and cumulus adjustment); H₂O vertical distribution; and H₂O longwave radiative treatment.

The climate sensitivity is shown to vary nonlinearly with T_s and to depend strongly on: (i) convective processes; (ii) H₂O continuum absorption; and (iii) upper tropospheric (pressure, p<500 mb) relative humidity. With one major exception, for all the cases considered in this paper, the climate sensitivity increases with T_s , for T_s <300 K and decreases by more than a factor of 2 as T_s , increases from 300 to 325 K. Hence, for both fixed lapse-rate and moist lapse-rate models, our calculations clearly rule out the possibility (but for the exception noted below) of a runaway greenhouse effect. The one major exception is when the upper tropospheric relative humidity value is allowed to attain values of 50% as opposed to the traditionally assumed one-dimensional model profile in which the relative humidity decreases linearly with P from a value of about 80% at the surface to about 15% at about 200 mb. In the instance, when the upper tropospheric relative humidity is held fixed (as T_s changes) at 50%, the sensitivity seems to increase even when T_s exceeds 300 K.

In order to facilitate theoretical interpretation of the numerical results, the climate feedback parameter λ is inferred from surface and from top-of-the-atmosphere energy balance considerations. The inferred λ illustrate the consistency between the two approaches of interpreting climate sensitivity.

Abstract

This study examines the spectral and diurnal variations in the planetary (i.e., top-of-atmosphere) clear sky albedo of the earth. The clear sky planetary albedo is calculated by a radiative transfer model which uses observed mean January and July earth properties on a global 5° × 5° grid. Our model calculations account for the regional, zonal and seasonal variations in humidity, temperature, sea ice and snow cover. In addition, seasonal and zonal variations in ozone are included. We calculate the diurnal cycle of clear sky planetary albedo in the following spectral intervals: 0.2–0.5, 0.5–0.7 and 0.7–4 μm. Model results reveal the strong wavelength dependence of planetary albedo. For all surfaces in our model, the planetary albedo decreases from morning to local noon, with the diurnal variations being particularly strong over water surfaces. We describe in detail the spectral and diurnal variations in planetary albedo over many natural surfaces, such as vegetation, snow, sea ice and ocean. The comprehensive model results presented in our study should find application in studies concerned with the estimation of potential spectral and diurnal sampling errors in satellite radiation budget measurements.

Abstract

Simplified band models are developed for methane (CH₄) and nitrous oxide (N₂O) bands in the longwave radiation spectrum. The band models are then employed in a radiation model to calculate the seasonally and latitudinally varying contributions of CH₄, and N₂O to the radiative energy balance of the earth-troposphere system. From the energy balance calculations, it is concluded that the longwave opacity (i.e., the so-called “greenhouse effect”) due to present-day observed concentrations of CH₄. and N₂O contribute nearly 2 K to hemispherical mean surface temperature with possible larger contributions to polar surface temperatures. The paper also discusses stratospheric effects of CH₄ and N₂O and examines the sensitivity of tropospheric radiation energy balance to large increases in CH₄ and N₂O.

Abstract

A simple moist model for the large-scale tropical atmospheric circulation is constructed by combining the simple models of Gill and Neelin and Held. The model describes the first baroclinic mode of the moist troposphere with variable “gross moist stability” in response to given thermodynamic forcing from surface evaporation and atmospheric cloud radiative forcing (CRF), which is a measure of the radiative effects of clouds in the atmospheric radiative heating. When the present model is forced solely by the observed atmospheric CRF, quantitatively reasonable Hadley and Walker circulations are obtained, such as the trades, the ascending branches in the intertropical convergence zone (ITCZ) and the South Pacific Convergence Zone (SPCZ), as well as the descending branches in the cold tongue and subtropics. However, when the model is forced only by the observed surface evaporation, the Walker circulation totally disappears, and the Hadley circulation reverses. These results indicate that, in the context of a moist dynamic model, the spatial variations of atmospheric CRF are more important in terms of driving and maintaining the Hadley and Walker circulations than the spatial variation of surface evaporation.

Abstract

In this paper diagnostic estimates of cloud radiative forcing (CRF) and clear-sky radiation budget at the surface and in the atmosphere, based on satellite-observed radiation budget at the top of the atmosphere (TOA) and empirical parameterizations derived from radiation models and field observations, are presented. This analysis is restricted to the tropical Pacific. High clouds over the intertropical convergence zone (ITCZ), the South Pacific convergence zone (SPCZ), and the warm pool (WP) exert a positive CRF of about 70 W m⁻² within the atmosphere and a negative CRF of about −70 W m⁻² at the surface, although with a negligible net CRF at the TOA. On the other hand, low clouds over the eastern subtropical Pacific and the equatorial cold tongue exert a negative CRF of about −20 W m⁻² at the surface as well as in the atmosphere. The spatial gradients of the clear-sky radiation budget at the surface and in the atmosphere are small. In particular, it is shown that the clear-sky radiative cooling in the atmosphere is larger over the ITCZ, the SPCZ, and the WP, when compared with that over the subtropics and the cold tongue. Next, based on these diagnostic estimates and available surface turbulent heat flux data, the role of atmospheric CRF in the large-scale atmospheric moist static energy (MSE) transport is quantified. It is found that the atmospheric CRF provides the major energy source for balancing the divergence of MSE transport (from the ITCZ, the SPCZ, and the WP to the subtropics and the cold tongue) by the large-scale atmospheric circulation. On the other hand, the clear-sky radiative flux convergence and the surface turbulent heat fluxes have just the reverse spatial pattern and hence cannot satisfy the large-scale atmospheric MSE transport requirements.