TROPICAL AND MONSOON
RESEARCH
Tropical Research in CGAM
Our tropical research programme has continued to make progress on a wide range of topics which are outlined below. Over the past six months we have contributed to CLIVAR related research in Europe through the organization of a Euroclivar Workshop on 'Climatic impact of scale interactions for the tropical ocean-atmosphere system', and through the drafting of the recommendations for research in 'Global Teleconnections' for the Euroclivar report on 'Climate Variability and Predictability Research in Europe, 1999-2004'.
In January we welcomed Rich Neale to the group. Rich has just completed a Ph.D. with Brian Hoskins on studies of tropical convection and the Hadley circulation using the aquaplanet version of the UM. Rich will be working on tropical teleconnections in the atmosphere-only and coupled versions of the UM.
Pete Inness will be joining the group in March from the Met. Office College. Pete worked at the Hadley Centre for several years on convective parametrization in the UM and on the simulation of the MJO. Pete will be working on the intraseasonal variability in the tropical Pacific and its relationship with El Niño.
Monsoons
Studies of the characteristics of monsoon subseasonal variability in the ECMWF Reanalysis and the UM L58 AMIP II integration, as well as the sensitivity of the subseasonal behaviour to soil moisture feedbacks and idealised El Niño forcing, are described in separate articles.
Much of our research has focused on the predictability of the Asian Summer Monsoon on seasonal timescales. The principal scientific basis for seasonal prediction in the atmosphere is that lower boundary conditions, such as SSTs, are less chaotic and may therefore be at least partially predictable. Even though the atmosphere is chaotic, these lower boundary conditions can 'lend' predictability to the atmosphere.
There has been considerable discussion recently on whether the intraseasonal variability of the monsoon is essentially chaotic, or whether its behaviour can be influenced by external forcing and hence contribute to predictability on the seasonal timescale. In collaboration with Ken Sperber at PCMDI, Livermore, we have used the NCEP 40-year Reanalysis to investigate intraseasonal variability of the Asian Summer Monsoon and its relationship to interannual variability and predictability. This work follows on from a preliminary study with ERA data, reported in Annamalai et al. (1999).
In this work we have considered the two hypotheses, firstly that the boundary forcing predisposes the monsoon system towards drought or flood on a seasonal mean basis, and secondly that the boundary forcing influences the intraseasonal variability, such that the probability of being in one weather regime or another is no longer equally likely. The latter hypothesis basically says that the active/break cycles of the monsoon may be chaotic, but the probability of one regime being more dominant than the other is affected by the low frequency external forcing in a predictable manner. We find that the boundary forcing, specifically the phase of ENSO, has little impact on the behaviour of the intraseasonal variability and the population of active vs. break regimes. We have concluded that the boundary forcing introduces a seasonal mean bias that predisposes the monsoon towards flood or drought conditions.
The predictability of the Indian Summer Monsoon has also been investigated by using the summer ensemble of ECMWF seasonal forecasts which span the ECMWF reanalysis period (1979-1993), and were performed as part of the EU Seasonal Prediction project, PROVOST. The results suggest that the current level of predictability is low, certainly much lower than that suggested by statistical methods. However, the model displays considerable systematic errors in its basic simulation of the monsoon's rainfall distribution which may be influencing the result. We hope to continue to reassess the level of predictability using updated model versions.
Although SST is considered to be the principal factor that influences the monsoon's interannual variability, there is evidence that land surface anomalies, such as springtime Eurasian snow cover, may also be a factor. This has led to a collaborative project with ECMWF and CINECA to investigate the influence of land surface temperature/Eurasian snow cover on monsoon predictability. Preliminary results suggest that land surface anomalies can influence the monsoon, but that their effects are secondary to those of SST.
1997 was characterised by the rapid development of an El Niño whose strength exceeded any previously observed this century. The basic understanding of the influence of El Niño on the Asian Summer Monsoon suggested that the monsoon should be substantially deficient, yet the All India Rainfall was slightly above normal. The reasons for this have been investigated in terms of both the seasonal mean, large scale circulation anomalies and the subseasonal, regional weather events. By comparing the results with a similar analysis of two previous major El Niño events in 1982 and 1987, the common and disparate features of the response have been identified.
On the large scale, the basic hypothesis that, in El Niño years, the strength of the monsoon is determined by a modulation of the Walker circulation, in which there is implied additional subsidence over the West Pacific and S.E. Asia, is generally supported by the results. However, with the exception of 1987, this simple scenario is not sufficient to explain the regional and subseasonal response, in particular the above normal AIR in 1997. Our results have shown that the modulation of the local Hadley circulation may play an important role. In both 1982 and 1997, the suppression of convection over the maritime continent and the equatorial Indian Ocean was marked. As a consequence, the ITCZ to the north over the continent was preferentially more active, particularly in 1997, allowing deep monsoon depressions to develop. The seasonal mean AIR anomalies are thus the consequence of a subtle balance between the suppressing effects of the Walker circulation and the enhancing effects of the local Hadley circulation. In 1997, despite a much weaker monsoon circulation and a substantial modulation of the Walker circulation, the changes in the local Hadley circulation dominated the monsoon activity, particularly during July and August.
The EU project on the Asian Summer Monsoon, SHIVA, concludes this spring. One of the outcomes of our research is that there is potential for coupled ocean-atmosphere processes to influence monsoon variability on timescales from days to decades. Much of our future research will focus on understanding these processes, beginning with a detailed study of monsoon variability in the coupled version of the UM, HadCM3.
The Madden-Julian Oscillation
Observations from TOGA COARE have indicated that the surface fluxes and sea surface temperatures (SST) in the Western Pacific are modulated on intraseasonal timescales in association with the Madden Julian Oscillation (MJO). Consequently, mechanisms for the maintenance of the MJO have been proposed which involve a coupling between the atmosphere and the ocean. Using ERA data, satellite OLR data and weekly SST analyses, we have investigated the relationship between the surface fluxes, convection and SST on intraseasonal timescales throughout the tropical Indian and Pacific Oceans, for the 15 year period, 1983-1997. We have also examined the robustness of these relationships from region to region and from year to year.
We have found that there are significant lag correlations (at the 95% level) between OLR and SST in the tropical Indian and Pacific Oceans between about 60E and 180E. Within this region there are also significant lag correlations between OLR and zonal wind stress, latent heat flux and shortwave flux. Outside this region the lag correlations between OLR and SST are weaker and the phase relationship between OLR and latent heat flux changes due to the alternation in the basic state winds from westerly to easterly. In general, we find that the SSTs are warmer than normal about 10 days prior to the maximum in convective activity. This warming is associated with increased solar radiation, reduced surface evaporation and light winds. Following the convective maximum, the SSTs cool due to reduced solar radiation and enhanced evaporation associated with stronger winds.
The spatial scale of the anomalies in OLR, SST and fluxes is about 60 in longitude from maximum to minimum. The magnitudes of the perturbations are near 20Wm-2 for the fluxes, with the shortwave flux anomalies being about 50% larger than the latent heat flux anomalies. Averaged over a large number of MJO events, the perturbation to the SST is small, being between 0.1 and 0.2K. However, on an individual basis the SST anomalies can reach 1K.
The results of this study provide firm evidence of a relationship between convection, surface fluxes and SST which will form the basis for a series of experiments with the aquaplanet version of the UM to investigate the mechanisms of ocean/atmosphere coupling at intraseasonal timescales, particularly with regard to the simulation of the MJO.
The impact of atmospheric intraseasonal variability on the tropical Pacific Ocean is an area of increasing interest. The rapid growth and exceptional amplitude of the 1997/98 El Niño seem to be due, in part, to strong MJO activity during the winter of 1996/97 which excited ocean Kelvin waves. Our future research will focus on the dynamical response of the ocean to intraseasonal forcing. In collaboration with Mike Davey and Sarah Ineson (UKMO) we have already started a series of experiments with their tropical Pacific (TOGA) high resolution ocean model forced with observed winds for 1997/98. We also plan to participate in an international intercomparison project (MJOMIP) which will study the response of the tropical Pacific Ocean to idealized MJO-related forcing.
Diurnal to Synoptic Timescale Variability
We have been participating in the EU funded project, CLAUS, which is constructing an archive of global window brightness temperature (BT) on a regular 0.5 x0.5 degree grid at 3-hourly intervals for the period 1983-present. We have been using the CLAUS data to investigate the diurnal cycle of tropical radiation and convection. The phase and amplitude of the diurnal cycle for different regimes, such as oceanic and land convection and clear sky land conditions, have been studied. The results suggest that the mechanisms to explain the observed diurnal convective variation over land and ocean are very different. The diurnal cycle of convection over land is basically a response to the surface radiative heating cycle, with convection tending to peak in the late afternoon/early evening. To explain deep oceanic convection with a late night or early morning maximum, several mechanisms may be involved: e.g., the direct radiation-convection interaction (Randall et al. 1991); the cloud vs. cloud free radiation difference (Gray and Jacobson 1977) and the rather complex surface-cloud-radiation interaction (Chen and Houze 1997).
As a basic forced mode of the climate system, an accurate representation of the phase and amplitude of the diurnal cycle over land and ocean should be a key test of surface, boundary layer and convective parametrizations. The simulation of the diurnal cycle in tropical convection and land surface temperature by the UM has been evaluated using the results from CLAUS as verification. We have found that the UM has various systematic errors in its simulation which can be related to aspects of the physical parametrizations. In particular, the phase of the diurnal cycle is not well simulated by the model. Over land this seems to be related to a tendency for the surface to warm up too rapidly during the morning. We plan to investigate the sensitivity of the diurnal cycle to changes in the model physics, such as the closure of the convection scheme.
Synoptic activity in the tropics is often associated with waves which can be related to the preferred equatorially-trapped modes of the atmospheric circulation based on shallow water theory. Various modes (inertio-gravity, equatorial Rossby, mixed Rossby-gravity and Kelvin waves) can be detected, in many cases related to the active phase of the MJO. We have used the global brightness temperature data (from the EU CLAUS project) to identify and characterise the equatorial waves based on time/space spectral analysis. The various modes noted above can be identified. The results have been used to evaluate the tropical variability in the AMIP II integration of the L58 UM. Preliminary results suggest that the model fails almost completely to capture these equatorial waves. The model also appears to have a very weak MJO. We plan to investigate the possibility that the introduction of convective momentum transport in HadAM3 may be contributing to these errors.
Tropical Teleconnections
Interannual variability in tropical SSTs, particularly El Niño, influences the global circulation through teleconnection patterns, a good example being the well-known Pacific-North American (PNA) pattern. In the tropics these teleconnections occur primarily through changes in the Walker Circulation. For example, in El Niño conditions, convective activity tends to be suppressed in the Atlantic and Indian Ocean regions giving rise to weak Atlantic hurricane seasons and deficient Indian monsoon rainfall. Correspondingly, SSTs in the Atlantic and Indian Oceans tend to warm as a remote response to El Niño. This tropics-wide response to El Niño is well known and can lead to global mean temperature anomalies. It is clear that a good simulation of tropical variability and its associated global teleconnections is a necessary requirement for climate change and seasonal to decadal prediction.
Through a 3-year contract with the Hadley Centre we are commencing a study of the tropical teleconnections in atmosphere-only and coupled versions of the UM. In HadAM3 and HadCM3 the tropics-wide response to ENSO is exaggerated resulting in a greater than observed temporal variability in the spatially averaged surface temperature. Initial assessment of ENSO in the AMIP II integration with the L58 UM confirms that the local response to El Niño in the Pacific is exaggerated and that the remote response over the Indian and Atlantic Oceans is not well simulated. We plan to investigate the mechanisms of the excessive ENSO response in HadAM3 and HadCM3 using sensitivity experiments with the aquaplanet model, with idealized El Niño forcing experiments and with the version of the UM with regional coupling between the ocean and atmosphere.
Recent Papers
Woolnough, S.J., J. M. Slingo and B. J. Hoskins, 1999: The relationship between convection and sea surface temperature on intraseasonal timescales. To be submitted.
Slingo, J. M. and H. Annamalai, 1999: 1997: The El Niño of the century and the response of the Indian Summer Monsoon. Submitted to Mon. Wea. Rev.
Hoskins, B. J. and G-Y. Yang, 1999: The equatorial response to higher latitude forcing. Submitted to J. Atmos. Sci.
Hoskins, B. J., R. B. Neale, M. J. Rodwell and G-Y. Yang, 1999: Aspects of the large-scale tropical atmospheric circulation. Tellus (in press)
Matthews, A. J., J. M. Slingo, B. J. Hoskins and P. M. Inness, 1999: Fast and slow Kelvin waves in the Madden-Julian Oscillation of a GCM. Q. J. R. Meteorol. Soc. (in press)
Ferranti, L., J. M. Slingo, T. N. Palmer and B. J. Hoskins, 1999: The effect of land surface feedbacks on the monsoon circulation. Q. J. R. Meteorol. Soc. (in press).
Annamalai, H., J. M. Slingo, K.R. Sperber and K. Hodges, 1999: The mean evolution and variability of the Asian Summer Monsoon: Comparison of ECMWF and NCEP/NCAR Reanalyses. Mon. Wea. Rev. (in press)
Slingo, J. M., D. P. Rowell, K. R. Sperber and F. Nortley, 1999: On the predictability of the interannual behaviour of the Madden-Julian Oscillation and its relationship with El Niño. Q. J. R. Meteorol. Soc., January Part B.
Yang, G. and J. M. Slingo, 1998: The seasonal mean and diurnal cycle of tropical convection as inferred from CLAUS data and the Unified Model. UGAMP Tech. Report No. 47.
Slingo, J. M., 1998: The 1997/98 El Niño. Weather, 53, 274-281.
Annamalai, H. and J. M. Slingo, 1998: The Asian Summer Monsoon, 1997. Weather, 53, 284-286.
Slingo, J. M., 1998: The Indian Summer Monsoon and its Variability. Chapter in 'Beyond El Niño: Decadal Variability in the Climate System' Edited by Antonio Navarra. To be published by Springer. In press.
Slingo, J. M. and P. Delecluse, 1998: Goal 4: Scale interactions and the tropical atmosphere-ocean system. Invited review paper to appear in the Proceedings of the COARE98 International Conference, Boulder.
Slingo, J. M., P. Delecluse and G. Komen, 1998: Climatic impact of scale interactions for the tropical ocean-atmosphere system. Euroclivar Workshop Report, Paris, September 1998.
Slingo, J. M., 1998: Studies of the Hydrology, Influence and Variability of the Asian Summer Monsoon (SHIVA). Proceedings of the EC Climate Conference, Vienna, October 1998.
Julia Slingo, H. Annamalai, Bernd Becker,
Rich Neale, Steve Woolnough, Guiying Yang
CGAM, University of Reading
J.M.Slingo@rdg.ac.uk
