• Allan, R. P., , and M. A. Ringer, 2003: Inconsistencies between satellite estimates of longwave cloud forcing and dynamical fields from reanalyses. Geophys. Res. Lett., 30 .1491, doi:10.1029/2003GLI017019.

    • Search Google Scholar
    • Export Citation
  • Allan, R. P., , M. A. Ringer, , and A. Slingo, 2003: Evaluation of moisture in the Hadley Centre climate model using simulations of HIRS water-vapour channel radiances. Quart. J. Roy. Meteor. Soc., 129 , 33713389.

    • Search Google Scholar
    • Export Citation
  • Allan, R. P., , M. A. Ringer, , J. A. Pamment, , and A. Slingo, 2004: Simulation of the Earth’s radiation budget for Medium-Range Weather Forecasts 40-year analysis (ERA40). J. Geophys. Res., 109 .D18107, doi:10.1029/2004JD004816.

    • Search Google Scholar
    • Export Citation
  • Barkstrom, B., and Coauthors, 1989: Earth Radiation Budget Experiment (ERBE) archival and April 1985 results. Bull. Amer. Meteor. Soc., 70 , 12541262.

    • Search Google Scholar
    • Export Citation
  • Bony, S., , K-M. Lau, , and Y. C. Sud, 1997: Sea surface temperature and large-scale circulation influences on tropical greenhouse effect and cloud radiative forcing. J. Climate, 10 , 20552077.

    • Search Google Scholar
    • Export Citation
  • Cess, R. D., , G. L. Potter, , W. L. Gates, , J-J. Morcrette, , and L. Corsetti, 1992: Comparison of general circulation models to Earth Radiation Budget Experiment data: Computation of clear-sky fluxes. J. Geophys. Res., 97 , 2042120426.

    • Search Google Scholar
    • Export Citation
  • Chung, E. S., , B. J. Sohn, , and V. Ramanathan, 2004: Moistening processes in the upper troposphere by deep convection: A case study over the tropical Indian Ocean. J. Meteor. Soc. Japan, 82 , 959965.

    • Search Google Scholar
    • Export Citation
  • Collins, W. D., , and A. K. Inamdar, 1995: Validation of clear-sky fluxes for tropical oceans from the Earth Radiation Budget Experiment. J. Climate, 8 , 569578.

    • Search Google Scholar
    • Export Citation
  • Fasullo, J., , and D-J. Sun, 2001: Radiative sensitivity to water vapor under all-sky conditions. J. Climate, 14 , 27982807.

  • Harshvardhan, and Randall, D., , T. G. Corsetti, , and D. A. Dazlich, 1989: Earth radiation budget and cloudiness simulations with a general circulation model. J. Atmos. Sci., 46 , 19221942.

    • Search Google Scholar
    • Export Citation
  • Iacono, M. J., , J. S. Delamere, , E. J. Mlawer, , and S. A. Clough, 2003: Evaluation of upper tropospheric water vapor in the NCAR Community Climate Model (CCM3) using modeled and observed HIR radiances. J. Geophys. Res., 108 .4037, doi:10.1029/2002JD002539.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Kiehl, J. T., 1994: On the observed near cancellation between longwave and shortwave cloud forcing in tropical regions. J. Climate, 7 , 559565.

    • Search Google Scholar
    • Export Citation
  • Klenk, K. F., , P. K. Bhartia, , E. Hilsenrath, , and A. J. Fleig, 1983: Standard ozone profile from balloon and satellite data sets. J. Climate Appl. Meteor., 22 , 20122022.

    • Search Google Scholar
    • Export Citation
  • Lanzante, J. R., , and G. E. Gahrs, 2000: The “clear-sky bias” of TOVS upper-tropospheric humidity. J. Climate, 13 , 40344041.

  • Manabe, S., , and R. T. Wetherald, 1967: Thermal equilibrium of the atmosphere with a given distribution of relative humidity. J. Atmos. Sci., 24 , 241259.

    • Search Google Scholar
    • Export Citation
  • McClatchey, R. A., , R. W. Fenn, , J. E. A. Selby, , F. E. Volz, , and J. S. Garing, 1972: Optical properties of the atmosphere. 3d ed. AFCRL-72-0497, AD 753075, Air Force Geophysical Laboratory, L.G. Hanscom Field, MA, 108 pp.

  • Pierrehumbert, R. T., , and R. Roca, 1998: Evidence for control of Atlantic subtropical humidity by large scale advection. Geophys. Res. Lett., 25 , 45374540.

    • Search Google Scholar
    • Export Citation
  • Ramanathan, V., , R. D. Cess, , E. F. Harrison, , P. Minnis, , B. R. Barkstrom, , E. Ahmad, , and D. Hartmann, 1989: Cloud-radiative forcing and climate: Results from the Earth Radiation Budget Experiment. Science, 243 , 5763.

    • Search Google Scholar
    • Export Citation
  • Ramanathan, V., and Coauthors, 2001: The Indian Ocean Experiment: An integrated assessment of the climate forcing and effects of the Great Indo-Asian haze. J. Geophys. Res., 106 , 2837128399.

    • Search Google Scholar
    • Export Citation
  • Roca, R., , S. Louvet, , L. Picon, , and M. Desbois, 2005: A study of convective systems, water vapor and top of the atmosphere cloud radiative forcing over the Indian Ocean using INSAT-1B and ERBE data. Meteor. Atmos. Phys., 90 , 4965.

    • Search Google Scholar
    • Export Citation
  • Rossow, W., , and R. Schiffer, 1991: ISCCP cloud data products. Bull. Amer. Meteor. Soc., 72 , 220.

  • Schmetz, J., , C. Geijo, , W. P. Menzel, , K. Strabala, , L. van de Berg, , K. Holmlund, , and S. Tjemkes, 1995: Satellite observations of upper tropospheric relative humidity, clouds and wind field divergence. Contrib. Atmos. Phys., 68 , 345357.

    • Search Google Scholar
    • Export Citation
  • Schmetz, J., , B. J. Sohn, , M. König, , H. J. Lutz, , and R. Stuhlmann, 2002: “Background” upper tropospheric hunidity and the greenhouse effect observed from satellite. Proc. 2002 EUMETSAT Meteorological Satellite Conf., EUM P 36, Dublin, Ireland, EUMETSAT, 609–616.

  • Slingo, A., , J. A. Pamment, , and M. J. Webb, 1998: A 15-year simulation of the clear-sky greenhouse effect using the ECMWF reanalyses: Fluxes and comparisons with ERBE. J. Climate, 11 , 690708.

    • Search Google Scholar
    • Export Citation
  • Smith, E. A., , and L. Shi, 1992: Surface forcing of the infrared cooling profile over the Tibetan Plateau. Part I: Influence of relative longwave radiative forcing at high latitude. J. Atmos. Sci., 49 , 805822.

    • Search Google Scholar
    • Export Citation
  • Soden, B. J., , and R. Fu, 1995: A satellite analysis of deep convection, upper tropospheric humidity, and the greenhouse effect. J. Climate, 8 , 23332351.

    • Search Google Scholar
    • Export Citation
  • Soden, B. J., , and J. R. Lanzante, 1996: An assessment of satellite and radiosonde climatologies of upper-tropospheric water vapor. J. Climate, 9 , 12351250.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., 1994: Temperature-moisture biases in ECMWF analyses based on clear-sky longwave simulations constrained by SSM/I and MSU measurements and comparisons to ERBE estimates. J. Climate, 7 , 17071718.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., 1999: Cloud-induced infrared radiative heating and its implications for large-scale tropical circulations. J. Atmos. Sci., 56 , 26572672.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., , and J. Schmetz, 2004: Water vapor induced OLR variations associated with high cloud changes over the Tropics: A study from Meteosat-5 observations. J. Climate, 17 , 19871996.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., , J. Schmetz, , S. Tjemkes, , M. Koenig, , A. Arriaga, , and E. S. Chung, 2000: Intercalibration of the Meteosat-7 water vapor channel with SSM/T-2. J. Geophys. Res., 105 , 1567315680.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., , E. S. Chung, , J. Schmetz, , and E. A. Smith, 2003: Estimating upper tropospheric water vapor from SSM/T-2 measurements. J. Appl. Meteor., 42 , 488504.

    • Search Google Scholar
    • Export Citation
  • Udelhofen, P., , and D. L. Hartmann, 1995: Influence of tropical convective cloud systems on the relative humidity in the upper troposphere. J. Geophys. Res., 100 , 74237440.

    • Search Google Scholar
    • Export Citation
  • van de Berg, L., , A. Pyomjamsri, , and J. Schmetz, 1991: Monthly mean upper-tropospheric humidities in cloud-free areas from METEOSAT observations. Int. J. Climatol., 11 , 819826.

    • Search Google Scholar
    • Export Citation
  • Weaver, C. P., , W. D. Collins, , and H. Grassl, 1994: Relationship between clear-sky atmospheric greenhouse effect and deep convection during the Central Equatorial Pacific Experiment: Model calculations and satellite observations. J. Geophys. Res., 99 , 2589125901.

    • Search Google Scholar
    • Export Citation
  • Wielicki, B. A., , and R. N. Green, 1989: Cloud identification for ERBE radiative flux retrieval. J. Appl. Meteor., 28 , 11331146.

  • Wielicki, B. A., , B. R. Barkstrom, , E. F. Harrison, , R. B. Lee III, , G. L. Smith, , and J. E. Cooper, 1996: Clouds and the Earth’s Radiant Energy System (CERES): An Earth Observing System experiment. Bull. Amer. Meteor. Soc., 77 , 853868.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Scatter diagram of instantaneous radiosonde observed (raob) vs SSM/T-2 retrieved upper-tropospheric water vapor (i.e., integrated water vapor for the 500–200-hPa layer). Radiosonde observations were taken during INDOEX during January–March 1999.

  • View in gallery

    Geographical distribution of explained variances of UTW by vertically stratified ISCCP clouds for (a) the summer of 1997–98, and (b) the winter of 1996/97 and 1997/98.

  • View in gallery

    Scatterplot of measured vs predicted UTWs for (a) the summer of 1999 and (b) the winter of 1999/2000.

  • View in gallery

    Geographical distributions of (a) clear-sky UTW, (b) all-sky UTW, (c) clear-sky minus all-sky UTW, and (d) clear-sky minus all-sky UTH for the summer of 1997–98. Units are kg m−2 for UTWs, and UTH difference is given in %.

  • View in gallery

    Same as in Fig. 3, but for the winter of 1996/97 and 1997/98.

  • View in gallery

    Water vapor contribution to longwave cloud radiative forcing for (a) the summer of 1997–98 and (b) the winter of 1996/97 and 1997/98.

  • View in gallery

    Scatterplot of CERES longwave cloud forcing (CRFCERES) and water vapor contribution to the longwave CRF.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 27 27 5
PDF Downloads 15 15 3

Dry Bias in Satellite-Derived Clear-Sky Water Vapor and Its Contribution to Longwave Cloud Radiative Forcing

View More View Less
  • 1 School of Earth and Environmental Sciences, Seoul National University, Seoul, South Korea
  • 2 European Organization for the Exploitation of Meteorological Satellites, Darmstadt, Germany
  • 3 School of Earth and Environmental Sciences, Seoul National University, Seoul, South Korea
© Get Permissions
Full access

Abstract

In this paper, the amount of satellite-derived longwave cloud radiative forcing (CRF) that is due to an increase in upper-tropospheric water vapor associated with the evolution from clear-sky to the observed all-sky conditions is assessed. This is important because the satellite-derived clear-sky outgoing radiative fluxes needed for the CRF determination are from cloud-free areas away from the cloudy regions in order to avoid cloud contamination of the clear-sky fluxes. However, avoidance of cloud contamination implies a sampling problem as the clear-sky fluxes represent an area drier than the hypothetical clear-sky humidity in cloudy regions. While this issue has been recognized in earlier works this study makes an attempt to quantitatively estimate the bias in the clear-sky longwave CRF. Water vapor amounts in the 200–500-mb layer corresponding to all-sky condition are derived from microwave measurements with the Special Sensor Microwave Temperature-2 Profiler and are used in combination with cloud data for determining the clear-sky water vapor distribution of that layer. The obtained water vapor information is then used to constrain the humidity profiles for calculating clear-sky longwave fluxes at the top of the atmosphere. It is shown that the clear-sky moisture bias in the upper troposphere can be up to 40%–50% drier over convectively active regions. Results indicate that up to 12 W m−2 corresponding to about 15% of the satellite-derived longwave CRF in tropical regions can be attributed to the water vapor changes associated with cloud development.

Corresponding author address: Dr. B. J. Sohn, School of Earth and Environmental Sciences, Seoul National University, Mail Code NS80, Kwanak-Gu Shillim-Dong, Seoul 151-747, South Korea. Email: sohn@snu.ac.kr

Abstract

In this paper, the amount of satellite-derived longwave cloud radiative forcing (CRF) that is due to an increase in upper-tropospheric water vapor associated with the evolution from clear-sky to the observed all-sky conditions is assessed. This is important because the satellite-derived clear-sky outgoing radiative fluxes needed for the CRF determination are from cloud-free areas away from the cloudy regions in order to avoid cloud contamination of the clear-sky fluxes. However, avoidance of cloud contamination implies a sampling problem as the clear-sky fluxes represent an area drier than the hypothetical clear-sky humidity in cloudy regions. While this issue has been recognized in earlier works this study makes an attempt to quantitatively estimate the bias in the clear-sky longwave CRF. Water vapor amounts in the 200–500-mb layer corresponding to all-sky condition are derived from microwave measurements with the Special Sensor Microwave Temperature-2 Profiler and are used in combination with cloud data for determining the clear-sky water vapor distribution of that layer. The obtained water vapor information is then used to constrain the humidity profiles for calculating clear-sky longwave fluxes at the top of the atmosphere. It is shown that the clear-sky moisture bias in the upper troposphere can be up to 40%–50% drier over convectively active regions. Results indicate that up to 12 W m−2 corresponding to about 15% of the satellite-derived longwave CRF in tropical regions can be attributed to the water vapor changes associated with cloud development.

Corresponding author address: Dr. B. J. Sohn, School of Earth and Environmental Sciences, Seoul National University, Mail Code NS80, Kwanak-Gu Shillim-Dong, Seoul 151-747, South Korea. Email: sohn@snu.ac.kr

1. Introduction

Clouds reflect incoming solar radiation back to space and reduce the energy available within the earth–atmosphere system. On the other hand, emitted infrared radiation from the earth surface and the lower moist atmosphere is trapped by clouds, thereby reducing the longwave energy loss to space. Through these competing processes the presence of clouds fundamentally modifies atmospheric thermodynamic structure, and thus earth’s general circulation and climate (e.g., Manabe and Wetherald 1967; Bony et al. 1997; Sohn 1999, among many others). Thus, it is important to measure the cloud influence upon the radiation budget using globally available satellite data in order to better understand the global climate system.

For a given cloudy area with a cloud fraction Ac, the longwave radiation flux L* at the top of the atmosphere (TOA) can be expressed as follows:
i1520-0442-19-21-5570-e1
where L*f and L*c are clear-sky and overcast-sky longwave (LW) fluxes at the TOA, respectively. Then, the term Ac(L*fL*c) in Eq. (1) can be referred to as the cloud radiative forcing (CRF). In a satellite approach, CRF is generally determined by differencing the observed total flux from the clear-sky flux (e.g., Ramanathan et al. 1989). Thus, LW cloud radiative forcing [CRF(L)] is
i1520-0442-19-21-5570-e2
where L* is a direct measurement by the radiometer at the TOA while L*f is estimated from the composite of clear-sky pixels, which for longwave infrared measurements must be away from the cloudy regions.

Recent earth radiation budget estimation missions such as the Earth Radiation Budget Experiment (ERBE; Barkstrom et al. 1989) and the Clouds and the Earth’s Radiant Energy System (CERES; Wielicki et al. 1996) estimated CRF by determining the clear-sky flux L*f . The ERBE clear-sky radiation fluxes have been derived by synthesizing all measurements identified as clear scenes using the classification of each measurement in terms of percent cloudiness into clear, partly cloudy, mostly cloudy, and overcast scenes (Wielicki and Green 1989). Since then clear-sky and CRF products have been widely used for diagnosing various physical processes or assessing climate model performance. However, such a satellite clear-sky composite method implicitly assumes that the atmospheric conditions for the clear-sky and total overall sky (or all sky) are the same as indicated in Eq. (1). It is intuitively clear that differences in atmospheric conditions exist because clouds only develop under favorable dynamic and thermodynamic conditions particularly those associated with moist air. Thus, the hypothetical clear-sky atmospheric state for the cloud-free area (e.g., in a 2.5° × 2.5° grid area) should generally not be the same as in the adjacent cloud area.

It has been noted that ERBE-type clear-sky longwave fluxes (and then LW CRF), particularly over convective regions, are different from what is generally obtained from reanalysis products or radiosonde observations (Sohn 1994; Collins and Inamdar 1995; Slingo et al. 1998; Allan and Ringer 2003), suggesting that there exists a nonnegligible bias in the ERBE clear-sky flux causing a potential problem in interpreting CRF in climate studies. A satellite bias in clear-sky longwave fluxes has also been identified by modeling studies (e.g., Harshvardhan et al. 1989; Cess et al. 1992). Consistent with these findings, a dry bias in the upper-tropospheric humidity (UTH) from 6.7-μm water vapor channel satellite measurements was also noted when compared with the cloudy-sky radiosonde observations (Soden and Lanzante 1996; Lanzante and Gahrs 2000), although the dry bias was quoted as “modest“ (i.e., less than 10% of relative humidity). Lanzante and Gahrs (2000) further noted that the bias is less noticeable for shorter time scales but becomes more obvious as the averaging time increases to the climate scale.

Such biases seem to be largely related to the upper-tropospheric moisture because the LW trapping by water vapor is most significant when water vapor is abundant in the upper-atmospheric layer. From the analysis of UTH from Geostationary Operational Environmental Satellite 6.7-μm brightness temperatures Udelhofen and Hartmann (1995) noted that UTH seems to increase linearly with the high cloud amount. This observational evidence is consistent with the notion that outgoing longwave radiation (OLR) over the deep convection area also decreases with upper-tropospheric moistening associated with deep convection (Weaver et al. 1994; Sohn and Schmetz 2004).

Although the above studies recognized the possible water vapor influence on the longwave changes over the cloud area, it still remains unclear to what extent upper tropospheric moistening associated with deep convection contributes to satellite-derived longwave cloud forcing. The recent advent of microwave measurements from space such as the Special Sensor Microwave Temperature-2 Profiler (SSM/T-2) measurements allows us to shed light on that question because microwave radiation is able to penetrate cloud layers and thus provides water vapor information in both clear and cloudy areas.

In this study we provide evidence that upper-tropospheric water vapor (UTW) changes occurring with cloud formation are important by examining the difference in water vapor amount between the clear sky and the all sky, and their associated longwave forcing difference. In doing so, we generate a clear-sky UTW climatology that reflects the ERBE clear-sky condition by relating UTWs retrieved from SSM/T-2 brightness temperatures to International Satellite Cloud Climatology Project (ISCCP) cloud data (Rossow and Schiffer 1991). Then longwave radiation fluxes with the clear-sky and all-sky UTW fields are estimated. Their differences in turn provide a quantitative estimation of the bias in the longwave cloud radiative forcing due to sampling the humidity field of clear-sky area away from cloudy regions.

2. Data description

The water vapor amount in the layer between 200 and 500 mb (UTW) has been retrieved from the microwave measurements by SSM/T-2 on board the Defense Meteorological Satellite Program satellite by applying a statistical-physical algorithm developed by Sohn et al. (2003). The water vapor profile can be inferred from the SSM/T-2 instrument because the sensor carries three channels around the 183-GHz strong water vapor absorption band.

Although detailed information on the retrieval method and validation of UTWs is found in Sohn et al. (2003), it is worthwhile to provide the characteristics of retrieved UTW for the all-sky condition. The retrieved SSM/T-2 UTWs are compared against well-calibrated radiosonde observations taken during the Indian Ocean Experiment (INDOEX; Ramanathan et al. 2001), which are available from the Joint Office for Science Support of the University Corporation for Atmospheric Research. The comparison results for the January–March 1999 period (Fig. 1) show that the rms error is about 0.65 kg m−2 with a correlation coefficient of 0.87. Statistical results given in Fig. 1 are similar to those presented in Sohn et al. (2003). More importantly the UTW does not show the bias that exists in the IR-based upper-tropospheric water vapor measurements (e.g., Soden and Lanzante 1996; Lanzante and Gahrs 2000).

Pentad mean UTW data for the six summer months [June–August (JJA)] of 1997–98, and the six winter months [December–February (DJF)] of 1996/97 and 1997/98 were produced for this analysis. The pentad means were provided in a 2.5° × 2.5° grid format consistent with ISCCP cloud data.

To relate clouds to water vapor in the upper troposphere retrieved from SSM/T-2, ISCCP cloud data are employed. We use the ISCCP D1 dataset, which provides various retrieved and calculated parameters with a spatial resolution of 2.5° × 2.5° and a 3-h sampling interval. Daily averages of low, middle, and high clouds were computed for the JJA 1997–98 and DJF 1996/97 and 1997/98 periods by averaging eight observations per day. Then pentad means of low, middle, and high cloud were prepared to match the pentad mean UTWs.

3. Determination of clear-sky UTW

Instead of explicitly determining clear-sky UTW from larger SSM/T-2 footprints about 50 km at the nadir, we use cloud information in conjunction with SSM/T-2 measured all-sky UTW data in order to determine the “ERBE-like” clear-sky UTW. It has been suggested that high clouds can be used as a surrogate for the upper-tropospheric water vapor because the latter is due to detrained hydrometeors from cumulus towers (e.g., Udelhofen and Hartmann 1995). However, in this study, to account for the water vapor distribution under cloud-free conditions, we employ a regression scheme that uses vertically stratified clouds. Specifically, we use low-, medium-, and high-level cloud amounts from ISCCP as predictors for the simultaneous UTW observations derived from SSM/T-2.

Introducing low Al, middle Am, and high Ah cloud amounts and relating to UTW, the following regression equation can be formulated:
i1520-0442-19-21-5570-e3
where a is the interception point and the bs are regression coefficients. Equation (3) can be further expressed as the following equation:
i1520-0442-19-21-5570-e4
where UTWclr is the clear-sky UTW. The novel aspect of this study is that from a multiple linear regression relating multilevel cloud distribution to UTW, the clear-sky UTW can be obtained when Al = Am = Ah = 0. Because UTWclr is the amount of water vapor within the upper-tropospheric layer when the cloud is absent at a given location and ERBE-type clear-sky fluxes are from cloud-free scenes, UTWclr should be analogous to the UTW distribution that the EBRE clear-sky may have. Therefore, we refer to UTWclr as the “clear-sky UTW.”

In this study “all-sky UTW” represents the SSM/T-2 estimated UTW and it is given on the left-hand side of Eq. (4), indicating that the all-sky UTW is expressed by clear-sky UTW plus the water vapor change associated with cloud formation. Therefore the difference between the all-sky UTW and the clear-sky UTW can be interpreted as the water vapor change due to the presence of clouds.

4. Clear-sky UTW fields

The regression coefficients of Eq. (3) were determined at each 2.5° × 2.5° grid point from 36 pentad means of UTW and cloud amounts for JJA and DJF. To examine how well the UTW is predicted by vertically stratified clouds (here low, middle, and high clouds), explained variances for the JJA and DJF periods are given in Fig. 2. During the summer, explained variances higher than 40% are found over most of the tropical convective areas including the Asian summer monsoon region, the western Pacific warm pool area, and the ITCZ extending from the western Pacific to the eastern Pacific. Because UTW is strongly correlated with high-level clouds over the convective area (e.g., Soden and Fu 1995), relatively high values of explained variance can be expected over the convectively active regions. In contrast, relatively smaller explained variances are noted over the north and south flanks of the convective areas, implying that the correlation between clouds and UTW is weak over the dominant descending regions in which the moisture variation is mainly controlled by the horizontal advection of water vapor associated with the large-scale circulation (Pierrehumbert and Roca 1998).

The same interpretation can be applied for the DJF period. However, in comparison with the JJA period, the DJF period shows higher explained variances over the most of the Tropics except the North African and the Arabian dry regions and cold oceanic regions over the southeast Pacific and the South Atlantic off South Africa. These regions of lower explained variance also appear to be in the descending branch of the Hadley-type circulation where moisture variations are largely due to moisture advection and to a lesser extent cloud occurrence.

Because the regression coefficients in Eq. (3) are indicative of UTW sensitivity to a cloud amount change, it is worthwhile to examine the magnitudes of regression coefficients and associated UTW changes. To provide an insight into how UTWs are forced by cloud changes, regression coefficients in Eq. (3) are averaged over a convectively active region (0°–15°S, 90°E–150°W) during the winter. The averaged coefficients for the low, middle, and high clouds are 0.0016, 0.0228, 0.0353, respectively (see Table 1). Taking the mean cloud amounts into consideration, the UTW contributions by low, middle, and high clouds are 0.03, 0.45, and 1.06 kg m−2, respectively. While UTWs are most sensitive to the high cloud development, as expected, a significant UTW change can also be related to midlevel clouds because of a relatively high regression coefficient (i.e., 0.0228); this is understood by considering the broad vertical depth of the contribution function (see Sohn et al. 2000). Negligible contribution is found for low clouds.

Above we discussed how closely UTWs are related to vertically stratified clouds in terms of explained variances; however, it is also interesting to show the accuracy of the predicted UTW from clouds. In doing so, the UTWs for JJA 1999 and DJF 1999/2000 periods are diagnosed from the corresponding means of low, middle, and high cloud amounts on a seasonal time scale by applying the obtained regression coefficients. Different data periods are chosen in order to know whether regression coefficients are valid for different time periods. Results are given in Fig. 3 with UTWs retrieved from SSM/T-2 measurements. In Fig. 3, each data point represents the 3-month mean UTW at a given 2.5° × 2.5° grid point. High correlations of 0.98 and 0.97 in Fig. 3 prove the close relationship between measured and predicted UTWs, suggesting that the water vapor amounts in the tropical upper troposphere are closely related to the presence of clouds at least on seasonal time scales, however, variations seem to be larger on the pentad time scale as indicated by the lower explained variances in Fig. 2.

In the scattergrams for the summer of 1999 and the winter of 2000 (Fig. 3), the rms errors are 0.27 and 0.29 kg m−2, respectively. Assuming that pentad averages are statistically independent of each other, the rms error for a 6-month average result is about 0.19 kg m−2 for the summer and 0.20 kg m−2 for the winter because the rms error magnitude is proportional to N−1/2, where N is the total number of sample. Then the maximum error range of the clear-sky UTW during the summer is about 20% [approximately 0.20 kg m−2 (1 kg m−2)−1] over the very dry area and about 4% [approximately 0.20 kg m−2 (5 kg m−2)−1] over the moist area. This interpretation is consistent with the fact that larger explained variances are found over the convective areas while smaller explained variances are found over the cold oceans or dry subtropical high pressure regions as shown in Fig. 2.

Clear-sky and all-sky UTWs, their difference, and the corresponding UTH difference for JJA of 1997–98 (i.e., 6-month mean) are presented in Fig. 4. Clear-sky and all-sky UTHs were estimated by combining UTWs and NCEP seasonal mean atmospheric profiles. In so doing the relative humidity of the 200–500-mb layer was altered to satisfy the estimated UTW. The shaded areas represent UTWs greater than 3 kg m−2 in Figs. 4a,b, and values greater than 1 kg m−2 and greater than 15% in the difference fields of Figs. 4c and 4d, respectively. In the clear-sky UTW field (Fig. 4a), local maxima are found over the convectively active regions such as the Asian summer monsoon region and the area extending from the equatorial eastern Pacific to Colombia. Higher clear-sky UTW values over the convective areas are not surprising when we consider that the clear-sky area near the cloud edge is significantly influenced by the outflow of moist air from the cumulus tower or evaporation of dissipating clouds near the convection cell. Indeed, Udelhofen and Hartmann (1995) showed that the high relative humidity in the upper-tropospheric layer in convective regions is confined to within 500 km from the cloud edge. Thus, the clear-sky composite from cloud-free pixels should comprise relatively moister conditions in the convective area because clear areas between cloud clusters or near convective clouds are likely moist.

Because the all-sky UTW includes cloud influence on the moisture profile and the air is saturated within clouds, it is obvious that the measured UTW field for JJA in Fig. 4b shows maximum areas over the convectively active regions. As clearly indicated in the UTW difference map (Fig. 4c), the clear-sky UTW is drier by 1 kg m−2 over most of the Asian monsoon region, the western Pacific, central Africa, and the equatorial eastern Pacific–Central America area. It is noted that the dry bias of the clear sky reaches up to 40% of the mean UTW over the convectively active regions. The larger moisture in the all-sky condition is apparently due to the inclusion of extra water vapor associated with cloud formation as explained in Eq. (4). Considering that the UTW difference of 1 kg m−2 is about a 15% UTH change, the UTH biases are expected to be around 20%–30% over the highly convective area (Fig. 4d).

The finding corroborates the early UTH climatology results by van de Berg et al. (1991); they derived a “clear-sky UTH climatology” by discarding all areas with any high level and/or midlevel clouds on a scale of 32 × 32 Meteosat pixels (about 200 × 200 km). This was done to have a clean basis for validation with radiosonde observations. As a result the maximum values of UTH were around 50%–60% (with respect to ice). Interestingly Schmetz et al. (1995) found, with the same UTH retrieval method maximum, values over deep convective regions between 70% and 80% UTH when clear-sky pixels from areas with clouds were included.

During the DJF periods of 1996/97 and 1997/98, the maxima regions of clear-sky UTW and all-sky UTW are found in the Maritime Continent to the central Pacific, and in rain forest areas of South Africa and South America, due to the shift of the convective zone associated with the seasonal change (see Fig. 4). However, UTW magnitudes seem smaller than shown during the summer. Because of the shift of the deep convection zone, the difference map (Figs. 4c) shows the increased UTW for the all-sky greater than 1 kg m−2 over the most of the tropical latitudes between 10°N and 20°S. Up to 20%–30% UTH changes are also noted during the DJF period.

5. UTW contribution to the CRF

The main objectives of this study are (i) to assess how the UTW changes associated with cloud influence the satellite-derived LW CRF and (ii) to separate the contribution of increased moisture associated with clouds from the LW CRF. Equation (4) provides a basis for this approach. When only atmospheric gas contribution to the OLR is considered, the OLR difference between the UTW and UTWclr conditions can be interpreted as the contribution by water vapor change due to the cloud development. We calculated the OLR with UTW and UTWclr as inputs to constrain the humidity profile in the upper troposphere. For the OLR calculation, we first constructed the mean atmospheric conditions for JJA 1997–1998 and DJF 1996/97 and 1997/98 using the National Centers for Environmental Prediction (NCEP) reanalysis data (Kalnay et al. 1996). The NCEP data provide temperature at 17 designated pressure levels (1000, 925, 850, 700, 600, 500, 400, 300, 250, 200, 150, 100, 70, 50, 30, 20, and 10 mb) and relative humidity at lowest 8 pressure levels up to the 300-mb, with a horizontal resolution of 2.5° × 2.5° latitude and longitude. The TOA OLR flux is obtained from radiative transfer calculations for a fixed relative humidity between the 200- and 500-mb level whose precipitable water is equal to the assigned UTW, while the humidity below the 500-mb height is kept at the mean atmospheric conditions determined by NCEP data. Above the 200-mb level, water vapor was fixed with the reference atmosphere for the Tropics (McClatchey et al. 1972).

The vertical distribution of ozone amount is obtained from the Klenk et al. (1983) parameterization based on results of the Nimbus-4 Backscattered Ultraviolet experiment for the upper layer and on balloon measurements for the lower stratosphere and troposphere. Because this parameterization expresses the ozone profile as a function of time of year and latitude, zonal means for each season are specified. The vertical distribution of CO2 concentration is kept constant at 360 ppm throughout all atmospheric layers. Surface values are also obtained from NCEP reanalysis data.

The radiative transfer calculations are performed with a narrowband model that considers all relevant gaseous atmospheric absorbers (Smith and Shi 1992). Two sets of OLR were calculated corresponding to clear-sky and all-sky UTWs, and a difference field is presented to diagnose the LW CRF caused by the different moisture distribution.

OLR differences arising from UTW difference between the clear sky and the all sky show in Figs. 4 and 5 are presented in Fig. 6. As expected OLR fluxes determined from clear-sky UTW are always higher than those from the all-sky estimation because of drier conditions at least in the tropical region under study. Differences larger than 8 W m−2 are found in most of the convectively active regions. Because the CRF is determined by subtracting the measured total OLR from the clear-sky OLR, the contribution of longwave flux made by water vapor changes correlated with cloud presence is effectively included in the CRF. In other words, the satellite-estimated CRF represents radiation perturbation not only due to the cloud properties, but also due to the water vapor changes related to cloud formation.

To assess what degree cloud forcing can be contributed by the water vapor increase associated with cloud development a scatterplot of the CERES CRF and UTW forced flux was made (Fig. 7). In this diagram, the CERES CRF is only from one summer season of 1998 because of the limited availability of the Tropical Rainfall Measuring Mission/CERES data while the OLR difference between satellite and model approaches is based on the two summer seasons of 1997 and 1998. Here we assume the different averaging periods may not much detract from this comparison.

Figure 7 indicates that the water vapor effect on the CRF tends to increase rapidly and then levels off with respect to the CRF increase. It is noted that up to 15% of the LW CRF can be attributed to the UTW change induced by cloud occurrence when the CRF is at the high end. A much larger percentage of water vapor contribution is found when cloud forcing is smaller. This result is consistent with Schmetz et al. (2002) who degraded the original Meteosat-7 water vapor pixel resolution gradually from 5 km to coarser resolution, and indeed they found differences in clear-sky fluxes in cloudy regions of typically 10 W m−2. This is understood invoking the result of Udelhofen and Hartmann (1995) who showed that humidity in the upper troposphere decreases exponentially with the distance from the cloud edge, suggesting that moist cloud-free areas are predominantly located near the cloud edge. If the satellite has a higher spatial resolution to detect cloud-free pixels near cloud edges or small areas between clouds, the clear-sky composite would be moister and thus would give rise to a smaller clear-sky longwave flux. The results obtained in this study quantify that a significant water vapor contribution to CRF would render a quantitative comparison of climate model simulation with the satellite estimate questionable, unless the model incorporates a clear-sky determination method used by satellite. Noting the consequences of a dry bias in the satellite data, there have been efforts to reduce the bias by sampling models in a way more consistent with the satellite approach (e.g., Iacono et al. 2003; Allan et al. 2003, 2004), which brought about a better agreement in clear-sky OLR distributions.

6. Summary and conclusions

This study quantitatively assessed how much of the satellite-derived longwave CRF is due to the upper-tropospheric water vapor change associated with cloud formation. A clear-sky water vapor distribution corresponding to the ERBE-type clear-sky radiation field ought to be drier than the overcast area in which upper-tropospheric moistening occurs in association with the development of deep convective clouds because the ERBE-type clear sky is estimated from clear-sky observations away from clouds. To assess the water vapor changes associated with cloud development we estimated the hypothetical clear-sky UTW by relating UTWs retrieved from SSM/T-2 to ISCCP cloud data through a multiple linear regression. The obtained clear-sky UTW has been compared with measured all-sky UTW and the largest differences are found in convective regions. Corresponding OLR fluxes were calculated using NCEP reanalysis profiles and the two UTW fields.

It is concluded that the difference between two UTW fields (clear sky versus all sky) is largely due to a moistening process associated with the development of deep convective clouds. Because of that the satellite clear-sky determination method puts clear-sky fluxes toward the higher side of OLR by selecting and composing only cloud-free pixels, which are likely drier than the nearby cloud area for a given location. Therefore the common satellite-inferred LW CRF is not only due to the cloud hydrometers but also due to the UTW changes correlated with clouds. The largest contributions of up to 12 W m−2 are shown over convectively active tropical regions, corresponding to about 15% of the satellite-derived LW CRF. These results are consistent with clear-sky sampling bias up to 15 W m−2 over warm regions of strong ascent (Allan and Ringer 2003) and 5 W m−2 in the zonal mean over the Tropics (Cess et al. 1992).

It is suggested that the use of satellite-derived values of LW CRF for climate model validation requires a very careful analysis before one embarks on direct quantitative comparison, including the test of whether the climate model is capable of reproducing the UTH field (see Chung et al. 2004). We show that differences of 10 W m−2 or more in the LW CRF can be explained by how the clear-sky LW flux is derived.

This study also helps us to understand the physical mechanism inducing longwave cloud radiative forcing. Results suggest that satellite-estimated CRF is not only contributed by the cloud’s macrophysical (cloud amount, height, and cloud-top temperature) and microphysical properties (cloud droplet size distribution and particle shape) but also through the water vapor increase associated with cloud formation. This is in line with previous works on the role of water vapor in net cloud forcing balance/imbalance associated with tropical convective systems (Roca et al. 2005) and OLR sensitivity to given water vapor distributions (Fasullo and Sun 2001). Therefore one needs to be careful not to interpret the near cancellation between OLR and solar radiation only from the perspective of cloud properties (e.g., Kiehl 1994), as also noted in Allan and Ringer (2003) and Roca et al. (2005).

Acknowledgments

The authors express their appreciation to Ms. Hye-Suk Park, and Mr. Eui-Seok Chung of Seoul National University for his assistance with the analysis. The authors thank two anonymous reviewers for their constructive and valuable comments, which led to an improved version of the manuscript. This work has been supported by the Korea Meteorological Administration Research and Development Program under the Grant CATER 2006-2103, and by the BK21 Project of the Korean government.

REFERENCES

  • Allan, R. P., , and M. A. Ringer, 2003: Inconsistencies between satellite estimates of longwave cloud forcing and dynamical fields from reanalyses. Geophys. Res. Lett., 30 .1491, doi:10.1029/2003GLI017019.

    • Search Google Scholar
    • Export Citation
  • Allan, R. P., , M. A. Ringer, , and A. Slingo, 2003: Evaluation of moisture in the Hadley Centre climate model using simulations of HIRS water-vapour channel radiances. Quart. J. Roy. Meteor. Soc., 129 , 33713389.

    • Search Google Scholar
    • Export Citation
  • Allan, R. P., , M. A. Ringer, , J. A. Pamment, , and A. Slingo, 2004: Simulation of the Earth’s radiation budget for Medium-Range Weather Forecasts 40-year analysis (ERA40). J. Geophys. Res., 109 .D18107, doi:10.1029/2004JD004816.

    • Search Google Scholar
    • Export Citation
  • Barkstrom, B., and Coauthors, 1989: Earth Radiation Budget Experiment (ERBE) archival and April 1985 results. Bull. Amer. Meteor. Soc., 70 , 12541262.

    • Search Google Scholar
    • Export Citation
  • Bony, S., , K-M. Lau, , and Y. C. Sud, 1997: Sea surface temperature and large-scale circulation influences on tropical greenhouse effect and cloud radiative forcing. J. Climate, 10 , 20552077.

    • Search Google Scholar
    • Export Citation
  • Cess, R. D., , G. L. Potter, , W. L. Gates, , J-J. Morcrette, , and L. Corsetti, 1992: Comparison of general circulation models to Earth Radiation Budget Experiment data: Computation of clear-sky fluxes. J. Geophys. Res., 97 , 2042120426.

    • Search Google Scholar
    • Export Citation
  • Chung, E. S., , B. J. Sohn, , and V. Ramanathan, 2004: Moistening processes in the upper troposphere by deep convection: A case study over the tropical Indian Ocean. J. Meteor. Soc. Japan, 82 , 959965.

    • Search Google Scholar
    • Export Citation
  • Collins, W. D., , and A. K. Inamdar, 1995: Validation of clear-sky fluxes for tropical oceans from the Earth Radiation Budget Experiment. J. Climate, 8 , 569578.

    • Search Google Scholar
    • Export Citation
  • Fasullo, J., , and D-J. Sun, 2001: Radiative sensitivity to water vapor under all-sky conditions. J. Climate, 14 , 27982807.

  • Harshvardhan, and Randall, D., , T. G. Corsetti, , and D. A. Dazlich, 1989: Earth radiation budget and cloudiness simulations with a general circulation model. J. Atmos. Sci., 46 , 19221942.

    • Search Google Scholar
    • Export Citation
  • Iacono, M. J., , J. S. Delamere, , E. J. Mlawer, , and S. A. Clough, 2003: Evaluation of upper tropospheric water vapor in the NCAR Community Climate Model (CCM3) using modeled and observed HIR radiances. J. Geophys. Res., 108 .4037, doi:10.1029/2002JD002539.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Kiehl, J. T., 1994: On the observed near cancellation between longwave and shortwave cloud forcing in tropical regions. J. Climate, 7 , 559565.

    • Search Google Scholar
    • Export Citation
  • Klenk, K. F., , P. K. Bhartia, , E. Hilsenrath, , and A. J. Fleig, 1983: Standard ozone profile from balloon and satellite data sets. J. Climate Appl. Meteor., 22 , 20122022.

    • Search Google Scholar
    • Export Citation
  • Lanzante, J. R., , and G. E. Gahrs, 2000: The “clear-sky bias” of TOVS upper-tropospheric humidity. J. Climate, 13 , 40344041.

  • Manabe, S., , and R. T. Wetherald, 1967: Thermal equilibrium of the atmosphere with a given distribution of relative humidity. J. Atmos. Sci., 24 , 241259.

    • Search Google Scholar
    • Export Citation
  • McClatchey, R. A., , R. W. Fenn, , J. E. A. Selby, , F. E. Volz, , and J. S. Garing, 1972: Optical properties of the atmosphere. 3d ed. AFCRL-72-0497, AD 753075, Air Force Geophysical Laboratory, L.G. Hanscom Field, MA, 108 pp.

  • Pierrehumbert, R. T., , and R. Roca, 1998: Evidence for control of Atlantic subtropical humidity by large scale advection. Geophys. Res. Lett., 25 , 45374540.

    • Search Google Scholar
    • Export Citation
  • Ramanathan, V., , R. D. Cess, , E. F. Harrison, , P. Minnis, , B. R. Barkstrom, , E. Ahmad, , and D. Hartmann, 1989: Cloud-radiative forcing and climate: Results from the Earth Radiation Budget Experiment. Science, 243 , 5763.

    • Search Google Scholar
    • Export Citation
  • Ramanathan, V., and Coauthors, 2001: The Indian Ocean Experiment: An integrated assessment of the climate forcing and effects of the Great Indo-Asian haze. J. Geophys. Res., 106 , 2837128399.

    • Search Google Scholar
    • Export Citation
  • Roca, R., , S. Louvet, , L. Picon, , and M. Desbois, 2005: A study of convective systems, water vapor and top of the atmosphere cloud radiative forcing over the Indian Ocean using INSAT-1B and ERBE data. Meteor. Atmos. Phys., 90 , 4965.

    • Search Google Scholar
    • Export Citation
  • Rossow, W., , and R. Schiffer, 1991: ISCCP cloud data products. Bull. Amer. Meteor. Soc., 72 , 220.

  • Schmetz, J., , C. Geijo, , W. P. Menzel, , K. Strabala, , L. van de Berg, , K. Holmlund, , and S. Tjemkes, 1995: Satellite observations of upper tropospheric relative humidity, clouds and wind field divergence. Contrib. Atmos. Phys., 68 , 345357.

    • Search Google Scholar
    • Export Citation
  • Schmetz, J., , B. J. Sohn, , M. König, , H. J. Lutz, , and R. Stuhlmann, 2002: “Background” upper tropospheric hunidity and the greenhouse effect observed from satellite. Proc. 2002 EUMETSAT Meteorological Satellite Conf., EUM P 36, Dublin, Ireland, EUMETSAT, 609–616.

  • Slingo, A., , J. A. Pamment, , and M. J. Webb, 1998: A 15-year simulation of the clear-sky greenhouse effect using the ECMWF reanalyses: Fluxes and comparisons with ERBE. J. Climate, 11 , 690708.

    • Search Google Scholar
    • Export Citation
  • Smith, E. A., , and L. Shi, 1992: Surface forcing of the infrared cooling profile over the Tibetan Plateau. Part I: Influence of relative longwave radiative forcing at high latitude. J. Atmos. Sci., 49 , 805822.

    • Search Google Scholar
    • Export Citation
  • Soden, B. J., , and R. Fu, 1995: A satellite analysis of deep convection, upper tropospheric humidity, and the greenhouse effect. J. Climate, 8 , 23332351.

    • Search Google Scholar
    • Export Citation
  • Soden, B. J., , and J. R. Lanzante, 1996: An assessment of satellite and radiosonde climatologies of upper-tropospheric water vapor. J. Climate, 9 , 12351250.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., 1994: Temperature-moisture biases in ECMWF analyses based on clear-sky longwave simulations constrained by SSM/I and MSU measurements and comparisons to ERBE estimates. J. Climate, 7 , 17071718.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., 1999: Cloud-induced infrared radiative heating and its implications for large-scale tropical circulations. J. Atmos. Sci., 56 , 26572672.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., , and J. Schmetz, 2004: Water vapor induced OLR variations associated with high cloud changes over the Tropics: A study from Meteosat-5 observations. J. Climate, 17 , 19871996.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., , J. Schmetz, , S. Tjemkes, , M. Koenig, , A. Arriaga, , and E. S. Chung, 2000: Intercalibration of the Meteosat-7 water vapor channel with SSM/T-2. J. Geophys. Res., 105 , 1567315680.

    • Search Google Scholar
    • Export Citation
  • Sohn, B. J., , E. S. Chung, , J. Schmetz, , and E. A. Smith, 2003: Estimating upper tropospheric water vapor from SSM/T-2 measurements. J. Appl. Meteor., 42 , 488504.

    • Search Google Scholar
    • Export Citation
  • Udelhofen, P., , and D. L. Hartmann, 1995: Influence of tropical convective cloud systems on the relative humidity in the upper troposphere. J. Geophys. Res., 100 , 74237440.

    • Search Google Scholar
    • Export Citation
  • van de Berg, L., , A. Pyomjamsri, , and J. Schmetz, 1991: Monthly mean upper-tropospheric humidities in cloud-free areas from METEOSAT observations. Int. J. Climatol., 11 , 819826.

    • Search Google Scholar
    • Export Citation
  • Weaver, C. P., , W. D. Collins, , and H. Grassl, 1994: Relationship between clear-sky atmospheric greenhouse effect and deep convection during the Central Equatorial Pacific Experiment: Model calculations and satellite observations. J. Geophys. Res., 99 , 2589125901.

    • Search Google Scholar
    • Export Citation
  • Wielicki, B. A., , and R. N. Green, 1989: Cloud identification for ERBE radiative flux retrieval. J. Appl. Meteor., 28 , 11331146.

  • Wielicki, B. A., , B. R. Barkstrom, , E. F. Harrison, , R. B. Lee III, , G. L. Smith, , and J. E. Cooper, 1996: Clouds and the Earth’s Radiant Energy System (CERES): An Earth Observing System experiment. Bull. Amer. Meteor. Soc., 77 , 853868.

    • Search Google Scholar
    • Export Citation
Fig. 1.
Fig. 1.

Scatter diagram of instantaneous radiosonde observed (raob) vs SSM/T-2 retrieved upper-tropospheric water vapor (i.e., integrated water vapor for the 500–200-hPa layer). Radiosonde observations were taken during INDOEX during January–March 1999.

Citation: Journal of Climate 19, 21; 10.1175/JCLI3948.1

Fig. 2.
Fig. 2.

Geographical distribution of explained variances of UTW by vertically stratified ISCCP clouds for (a) the summer of 1997–98, and (b) the winter of 1996/97 and 1997/98.

Citation: Journal of Climate 19, 21; 10.1175/JCLI3948.1

Fig. 3.
Fig. 3.

Scatterplot of measured vs predicted UTWs for (a) the summer of 1999 and (b) the winter of 1999/2000.

Citation: Journal of Climate 19, 21; 10.1175/JCLI3948.1

Fig. 4.
Fig. 4.

Geographical distributions of (a) clear-sky UTW, (b) all-sky UTW, (c) clear-sky minus all-sky UTW, and (d) clear-sky minus all-sky UTH for the summer of 1997–98. Units are kg m−2 for UTWs, and UTH difference is given in %.

Citation: Journal of Climate 19, 21; 10.1175/JCLI3948.1

Fig. 5.
Fig. 5.

Same as in Fig. 3, but for the winter of 1996/97 and 1997/98.

Citation: Journal of Climate 19, 21; 10.1175/JCLI3948.1

Fig. 6.
Fig. 6.

Water vapor contribution to longwave cloud radiative forcing for (a) the summer of 1997–98 and (b) the winter of 1996/97 and 1997/98.

Citation: Journal of Climate 19, 21; 10.1175/JCLI3948.1

Fig. 7.
Fig. 7.

Scatterplot of CERES longwave cloud forcing (CRFCERES) and water vapor contribution to the longwave CRF.

Citation: Journal of Climate 19, 21; 10.1175/JCLI3948.1

Table 1.

Regression coefficients and cloud amounts for three ISCCP cloud types and their respective contributions to the all-sky UTWs averaged over a region (0°–15°S, 90°E–150°W) during the DJF 1996/97 and 1997/98 periods.

Table 1.
Save