El Niño has been observed to exhibit decadal changes in its properties; the cause and implication of such changes are strongly debated. Various mechanisms proposed to be responsible for the decadal modulation of ENSO in the late 1970s are examined using coupled atmosphere-ocean models, with a focus on the roles of decadal changes in ocean thermal structure observed in the tropical Pacific. Two types of coupled models are used, one intermediate coupled model (ICM) and another hybrid coupled model (HCM), which consist of the same intermediate ocean model (IOM) with an empirical parameterization for the temperature of subsurface water entrained into the mixed layer (Te), which is constructed via a singular value decomposition (SVD) analysis of historical data. The differences in the ICM and HCM are in the atmospheric component: the former is a statistical one estimating wind stress (_) anomalies based on a SVD analysis, and the latter is an atmospheric general circulation model (AGCM; ECHAM4.5). Signal part of wind stress response to SST anomalies is constructed by a 24 member ensemble ECHAM4.5 simulations forced by observed SST; stochastic wind stress forcing is estimated from the AGCM ensemble and coupled simulations. The ICM and HCM simulations can very well reproduce interannual variability associated with El Niño in the tropical Pacific. Sensitivity model simulations are made using the coupled models with different realizations of stochastic forcing in the atmosphere, and observed changes in ocean thermal structure in the late 1970s which are introduced into the coupled systems through the Te parameterizations from two sub-periods before (1963-79) and after (1980-96) the climate shift (Te 63-79and Te 80-96), respectively. It is demonstrated that the properties of ENSO are modulated differently by the decadal Te changes in the ocean and stochastic forcing in the atmosphere. The former is responsible for changes in the phase propagation of ENSO, while the latter can contribute to the amplitude and period modulation.

By definition, many scientific issues in Earth Science are global in nature. Therefore, when we study such issues with the aid of observations and models/data assimilation we need to deal with massive datasets. As a result, there is great utility in a high degree of automation at many levels in many areas. The high degree of
automation allows us to focus our attention on the scientific issues. We will show examples of automation in model creation, observation cross-validation and calibration, data analysis and web site creation. The concept is also being used in designing prototypes of the next generation of earth observing system with real time
autonomous observation direction.

Results of GCM simulations will be presented which isolate the influence of the new techniques as implemented in the MITgcm. The direct result of using Extended Mosaic was an increase in the local turbulent kinetic energy. This resulted in a change in character of a reinforcing feedback which is manifest in either a drying or a wetting regime, depending on the local surface conditions. Extended Mosaic also resulted in a strengthened the land-atmosphere coupling in the regions where this change in the feedback was manifest. The results of a series of aquaplanet simulations using GridAlt show that it captures most of the effects of increased vertical resolution. Realistic simulations with GridAlt show an increased atmospheric response to the underlying surface temperature.

Max Planck Institute for Meteorology, Hamburg, ESSC, University of Reading, UK

Tropical Cyclones (TC) under different climate conditions in the Northern Hemisphere have been investigated with the Max Planck Institute (MPI) coupled   (ECHAM5/MPI-OM) and atmosphere (ECHAM5) climate models at different resolutions. The intensity and size of TC depends crucially on resolution with higher wind speed and smaller scales at higher resolution. The typical size of a TC is reduced by a factor of 2.3 from T63 to T319. The structure of the storms is increasingly realistic at higher resolution.

The model has been tested using observed initial data from ERA-40 and operational data from 2005 and 2006. There is a realistic response to ENSO with less Atlantic TC during El Nino.

Series of 30-year experiments have been undertaken using conditions from the end of the 19 th and 20 th century and IPCC SRES scenario A1B for the end of the 21st century.

There are no significant differences between the 19 th and 20 th century but a reduction of TC at the end of the 21 st century. Reduction in the number of storms occurs in all regions. The high-resolution experiments indicate intensification at the upper end of TC, but this is not noticeable at the low resolution (T63).

We identify two competing processes effecting TC in a warmer climate. First, the increase in the static stability and reduced vertical circulation is suggested to contribute to the reduction in the number of storms. Second, the increase in temperature and water vapor provide more energy for the storms so when favorable conditions occur, higher SST and higher specific humidity will contribute to more intense storms. However, this will require higher resolution to have its full effect.

It is well understood that quality short and long range weather forecasts require a valid stratosphere.  The Climate Prediction Center (CPC) has a long history of providing guidance to NCEP's Environmental Modeling Center on the quality of the stratosphere in the Global Forecast System (GFS).  This includes the quality of ozone, temperature, height and wind analyses and forecasts.  It was a natural extension of our capabilities to perform the same type of stratospheric assessment of forecasts generated from the Climate Forecast System (CFS).  CPC has since evaluated the basic stratospheric state of the Operational (T62L64) CFS forecasts and a CFS version of the current GFS (GSI-Hybrid T382L64).  We will present temperature, height, and ozone comparisons of these two models with respect to the DOE NCEP Reanalysis 2 and CPC analyses.  We will also present comparisons of the poleward heat flux and how well the structure of the wind fields are maintained with forecast time.  This includes the maintenance of the QBO.   Future evaluation plans will be discussed.

The Arctic climate is complex due to numerous nonlinear interactions between and within the atmosphere, cryosphere, ocean, and land. A dynamic downscaling experiment for the atmosphere in Arctic using ARPEGE climate model has been done, the systematic errors in the model are discussed in the meanwhile.   Furthermore, the poleward energy transports based on NCEP Reanalysis data and ARPEGE climate model simulation results are calculated, a long term variability of energy transport in North hemisphere based on Ncep data is showed as well. Finally, Cyclone tracks that occurred in the Northern Hemisphere during 50 extended winters has been identified from NCEP reanalyses and model simulation results by means of an objective tracking method (Hodges 1995, 1996). The tracks have been divided into two classes, one with northward moving cyclones and a second with southward moving cyclones.

Recent advances in computing technology have allowed a new kind of general circulation model (GCM) to emerge, a Multi-scale Modeling Framework (MMF). It is becoming an increasingly practical tool in the area of climate modeling. In MMF, most of conventional sub-GCM-grid-scale parameterizations in each grid-column are replaced with imbedded cloud-resolving model (CRM), which in this context is often called a super-parameterization. The MMF approach allows clouds, aerosols, turbulence, and radiation processes interact on the cloud-scale, as they do in nature. One particular strength of the MMF approach is that it provides a direct link among the GCM, CRM, and cloud microphysics communities. The results of a 19-year long AMIP-style climate simulation and simple climate sensitivity experiments will be discussed. The MMF does a good job of reproducing the interannual variability in terms of spatial structure and magnitude of major atmospheric anomalies associated with the ENSO. The subseasonal variability of tropical climate associated with the Maddan-Julian Oscillation (MJO) and equatorially trapped waves are particular strengths of the simulation. Strong sensitivity of equatorially trapped disturbances such as MJO and Kelvin waves to 2K SST perturbation will be demonstrated using a simplified aqua-planet setup. Simulated cloud regimes defined by the value of 500-mb pressure vertical velocity in Tropics will be compared to observations. Preliminary analysis of simulated cloud-scale vertical velocity statistics from the MMF don't seem to show any significant land/ocean contrast in convective intensity that would explain large contrast in lightning frequency

Presence of snow strongly affects the surface energy budget in mid-to-high latitudes during the transition period of winter to spring. In addition, synoptic waves propagating over the Northern Plains are a major source of snow storms and flash flooding during this period. The presence of snow and its melt from winter to spring affects the propagation of synoptic waves and the amount of precipitation over this region. Therefore, it is crucial to include the cold land physics in high- and temperate-latitudes for regional climate studies. Here we investigate the effect of snow cover, soil frost, and snowmelt on the atmosphere over cold land and implement a multi-layer Soil-Snow Model (SSM) coupled to an atmospheric model - Purdue Regional Climate Model (PRCM). We have applied a one-dimensional, multi-layer land surface model based on the conservation of heat and water substance inside the soil and snow for use with the regional climate model. Compared to the current land-surface scheme, the new SSM shows significant improvement in both moisture and temperature simulation for the months of March and April 1997, which affects the surface energy budget and the hydrological cycle. Overall, the regional climate simulation of March and April 1997 with the inclusion of detailed frozen soil and snow processes improves the synoptic and local circulations during the cold season. The partitioning of incoming radiative energy into sensible and latent heat fluxes is shown to be sensitive to snowmelt and soil freeze/thaw conditions during the early stages of the model simulations. Both spatial and temporal analyses of PRCM-SSM indicate that the coupling of these processes is important when simulating cold season regional climate with frozen soil and snow cover.