Observed sea level pressure (SLP) trends for 1901–10, 1951–10, and 1981–2010 are assessed using two observed data sources (HadSLP2_lowvar and 20CRv3) compared to a CMIP5 multimodel ensemble. The CMIP5 simulations include runs with (i) no external forcing (Control runs), (ii) natural external forcing only (Natural-Forcing), or (iii) natural plus anthropogenic forcings combined (All-Forcings). We assess whether the CMIP5 All-Forcing ensemble is consistent with observations and whether there is model-based evidence for detectable anthropogenic influence for the observed SLP trends. For the 1901–2010 and 1951–2010 trends, a robustly detectable anthropogenic signal in both observational data products is a zonal band of SLP increase extending over much of the Southern Hemisphere extratropics (30°–50°S). In contrast, the HadSLP2_lowvar and 20CRv3 observed data products disagree on the sign of the century-scale trends in SLP over much of the low-latitude region 25°N–25°S. These differences will limit confident detection/attribution/consistency conclusions for lower-latitude regions, at least until the observational data product discrepancies are better reconciled. The Northern Hemisphere extratropics remains a difficult region for identifying any detectable anthropogenic influence for annual- or seasonal-mean SLP trends. Overall, our results highlight the difficulty in detecting and attributing anthropogenic signals in SLP for relatively short time scales. The observed 1981–2010 regional trends typically have a different pattern and magnitude from the simulated externally forced trends. Consequently, our results suggest that internal variability is likely the dominant driver of most observed 1981–2010 regional trend features, including the pronounced increase in SLP over the central and eastern equatorial Pacific.
We describe the baseline coupled model configuration and simulation characteristics of GFDL's Earth System Model Version 4.1 (ESM4.1), which builds on component and coupled model developments at GFDL over 2013–2018 for coupled carbon‐chemistry‐climate simulation contributing to the sixth phase of the Coupled Model Intercomparison Project. In contrast with GFDL's CM4.0 development effort that focuses on ocean resolution for physical climate, ESM4.1 focuses on comprehensiveness of Earth system interactions. ESM4.1 features doubled horizontal resolution of both atmosphere (2° to 1°) and ocean (1° to 0.5°) relative to GFDL's previous‐generation coupled ESM2‐carbon and CM3‐chemistry models. ESM4.1 brings together key representational advances in CM4.0 dynamics and physics along with those in aerosols and their precursor emissions, land ecosystem vegetation and canopy competition, and multiday fire; ocean ecological and biogeochemical interactions, comprehensive land‐atmosphere‐ocean cycling of CO2, dust and iron, and interactive ocean‐atmosphere nitrogen cycling are described in detail across this volume of JAMES and presented here in terms of the overall coupling and resulting fidelity. ESM4.1 provides much improved fidelity in CO2 and chemistry over ESM2 and CM3, captures most of CM4.0's baseline simulations characteristics, and notably improves on CM4.0 in (1) Southern Ocean mode and intermediate water ventilation, (2) Southern Ocean aerosols, and (3) reduced spurious ocean heat uptake. ESM4.1 has reduced transient and equilibrium climate sensitivity compared to CM4.0. Fidelity concerns include (1) moderate degradation in sea surface temperature biases, (2) degradation in aerosols in some regions, and (3) strong centennial scale climate modulation by Southern Ocean convection.
One of the most consequential impacts of anthropogenic warming on humans may be increased heat stress, combining temperature and humidity effects. Here we examine whether there are now detectable changes in summertime heat stress over land regions. As a heat stress metric we use a simplified wet bulb globe temperature (WBGT) index. Observed trends in WBGT (1973–2012) are compared to trends from CMIP5 historical simulations (eight-model ensemble) using either anthropogenic and natural forcing agents combined or natural forcings alone. Our analysis suggests that there has been a detectable anthropogenic increase in mean summertime heat stress since 1973, both globally and in most land regions analyzed. A detectable increase is found over a larger fraction of land for WBGT than for temperature, as WBGT summertime means have lower interannual variability than surface temperature at gridbox scales. Notably, summertime WBGT over land has continued increasing in recent years--consistent with climate models--despite the apparent ‘hiatus’ in global warming and despite a decreasing tendency in observed relative humidity over land since the late 1990s.
Lau, Ngar-Cheung, and Jeff J Ploshay, December 2013: Model projections of the changes in atmospheric circulation and surface climate over North America, North Atlantic and Europe in the 21st century. Journal of Climate, 26(23), DOI:10.1175/JCLI-D-13-00151.1. Abstract
The impacts of climate change on the North America-North Atlantic-Europe sector are studied using a coupled general circulation model (CM3) and a high-resolution atmosphere-only model (HiRAM), both developed at the Geophysical Fluid Dynamics Laboratory. The CM3 experiment is conducted under two climate change scenarios for the 1860-2100 period. The sea surface temperature (SST) forcing prescribed in the ‘time-slice’ integrations with HiRAM is derived from observations for the 1979-2008 period, and projection by CM3 for the 2086-2095 period.
The wintertime response in the late 21st century is characterized by an enhancement of the positive phase of the North Atlantic Oscillation in sea level pressure (SLP), and poleward and eastward displacements of the Atlantic jetstream and storm track. The forcing pattern due to eddy vorticity fluxes in the perturbed storm track matches well with the response pattern of the SLP field in the late 21st century. The model results suggest that the above circulation changes are linked to the gradient of the altered SST forcing in the North Atlantic.
In summer, the projected enhancement of convection over the eastern tropical Pacific is accompanied by a wavetrain spanning the North America-North Atlantic-Europe sector. This quasi-stationary circulation pattern is associated with diminished storm track activity at 40°-50°N, and an eddy forcing pattern that is similar to the summertime SLP response in the late 21st century.
Miyakoda, Kikuro, Annalisa Cherchi, A Navarra, S Masina, and Jeff J Ploshay, February 2012: ENSO and its effects on the atmospheric heating processes. Journal of the Meteorological Society of Japan, 90(1), DOI:10.2151/jmsj.2012-103. Abstract
El Ni˜no-Southern Oscillation (ENSO) is an important air-sea coupled phenomenon that plays a dominant role in the
variability of the tropical regions. Observations, atmospheric and oceanic reanalysis datasets are used to classify ENSO
and non-ENSO years to investigate the typical features of its periodicity and atmospheric circulation patterns. Among
non-ENSO years, we have analyzed a group, called type-II years, with very small SST anomalies in summer that tend
to weaken the correlation between ENSO and precipitation in the equatorial regions. A unique character of ENSO is
studied in terms of the quasi-biennial periodicity of SST and heat content (HC) fields over the Pacific-Indian Oceans.
While the SST tends to have higher biennial frequency along the Equator, the HC maximizes it into two centers in the
western Pacific sector. The north-western center, located east of Mindanao, is strongly correlated with SST in the NINO3
region. The classification of El Ni˜no and La Ni˜na years, based on NINO3 SST and north-western Pacific HC respectively,
has been used to identify and describe temperature and wind patterns over an extended-ENSO region that includes the
tropical Pacific and Indian Oceans.
The description of the spatial patterns within the atmospheric ENSO circulation has been extended to tropospheric
moisture fields and low-level moisture divergence during November–December–January, differentiating the role of El
Ni˜no, when large amounts of condensational heat are concentrated in the central Pacific, from La Ni˜na that tends to
mainly redistribute heat to Maritime Continents and higher latitudes. The influence of the described mechanisms on
equatorial convection in the context of the variability of ENSO on longer timescales for the end of the 20th century is
questioned. However, the inaccuracy of the atmospheric reanalysis products in terms of precipitation and the shorter time
length of more reliable datasets hamper a final conclusion on this issue.
The Geophysical Fluid Dynamics Laboratory (GFDL) has developed a coupled general circulation model (CM3) for atmosphere, oceans, land, and sea ice. The goal of CM3 is to address emerging issues in climate change, including aerosol-cloud interactions, chemistry-climate interactions, and coupling between the troposphere and stratosphere. The model is also designed to serve as the physical-system component of earth-system models and models for decadal prediction in the near-term future, for example, through improved simulations in tropical land precipitation relative to earlier-generation GFDL models. This paper describes the dynamical core, physical parameterizations, and basic simulation characteristics of the atmospheric component (AM3) of this model.
Relative to GFDL AM2, AM3 includes new treatments of deep and shallow cumulus convection, cloud-droplet activation by aerosols, sub-grid variability of stratiform vertical velocities for droplet activation, and atmospheric chemistry driven by emissions with advective, convective, and turbulent transport. AM3 employs a cubed-sphere implementation of a finite-volume dynamical core and is coupled to LM3, a new land model with eco-system dynamics and hydrology.
Most basic circulation features in AM3 are simulated as realistically, or more so, than in AM2. In particular, dry biases have been reduced over South America. In coupled mode, the simulation of Arctic sea ice concentration has improved. AM3 aerosol optical depths, scattering properties, and surface clear-sky downward shortwave radiation are more realistic than in AM2. The simulation of marine stratocumulus decks and the intensity distributions of precipitation remain problematic, as in AM2.
The last two decades of the 20th century warm in CM3 by .32°C relative to 1881-1920. The Climate Research Unit (CRU) and Goddard Institute for Space Studies analyses of observations show warming of .56°C and .52°C, respectively, over this period. CM3 includes anthropogenic cooling by aerosol cloud interactions, and its warming by late 20th century is somewhat less realistic than in CM2.1, which warmed .66°C but did not include aerosol cloud interactions. The improved simulation of the direct aerosol effect (apparent in surface clear-sky downward radiation) in CM3 evidently acts in concert with its simulation of cloud-aerosol interactions to limit greenhouse gas warming in a way that is consistent with observed global temperature changes.
Using the historical surface temperature data set compiled by Climatic Research Unit of University of East Anglia and Hadley Centre of UK, this study examines the seasonal and latitudinal profile of the surface temperature change observed during the last several decades. It reveals that the recent change in zonal mean surface air temperature is positive at practically all latitudes. In the Northern Hemisphere, the warming increases with increasing latitude and is large in the Arctic Ocean during much of the year except in summer, when it is small. At the Antarctic coast and in the northern part of the Circumpolar Ocean (near 55°S), where limited data are available, the changes appear to be small during most seasons, though the warming is notable at the coast in winter. This warming is, however, much less than the warming over the Arctic Ocean. The seasonal variation of the surface temperature change appears to be broadly consistent with the result from a global warming experiment, which was conducted some time ago using a coupled atmosphere-ocean-land model.
The simulation of the diurnal cycle (DC) of precipitation and surface wind
circulation by a global general circulation model (GCM) with a horizontal resolution of
50 km is evaluated. The model output is compared with observational counterparts based
on datasets produced by the Tropical Rainfall Measurement Mission and the European
Centre for Medium-Range Weather Forecasts. The summertime diurnal characteristics
over tropical regions in Asia, the Americas and Africa are portrayed using the amplitude
and phase of the first harmonic of the 24-h cycle, departures of data fields during selected
hours from the daily mean, and differences between extreme phases of the DC.
There is general agreement between model and observations with respect to the
large-scale land-sea contrasts in the DC. Maximum land precipitation, onshore flows and
landward migration of rainfall signals from the coasts occur in afternoon, whereas peak
maritime rainfall and offshore flows prevail in morning. Seaward migration of
precipitation is discernible over western Bay of Bengal and South China Sea during
nocturnal and morning hours. The evolution from low-intensity rainfall in morning/early
afternoon to heavier precipitation several hours later is also evident over selected
continental sites.
Although the model provides an adequate simulation of the daytime upslope and
nighttime downslope winds in the vicinity of mountain ranges, valleys and basins, there
are notable discrepancies between model and observations in the DC of precipitation near
some of these orographic features. The model does not reproduce the observed seaward
migration of precipitation from the western coasts of Myanmar and India, and from
individual islands of the Indonesian Archipelago at nighttime.
Lau, Ngar-Cheung, and Jeff J Ploshay, February 2009: Simulation of synoptic- and subsynoptic-scale phenomena associated with the East Asian summer monsoon using a high-resolution GCM. Monthly Weather Review, 137(1), DOI:10.1175/2008MWR2511.1. Abstract
A 20-yr simulation using a global
atmospheric general circulation model with a resolution of 0.5° latitude ×
0.625° longitude is compared with observational findings. The primary goal
of this survey is to assess the model performance in reproducing various
summertime phenomena related to the continental-scale Asian monsoon in
general, and the regional-scale East Asian monsoon in particular. In both
model and observed atmospheres, the seasonal march of the precipitation
centers associated with the Asian summer monsoon is characterized by onsets
occurring earliest over the southeastern Bay of Bengal, followed by rapid
northeastward advances over Indochina, the South China Sea–Philippine Sea
and the western Pacific, northward evolution in the East Asian sector, as
well as northwestward development over the Bay of Bengal, the Indian
subcontinent, and the Arabian Sea. This onset sequence is accompanied by
southwesterly low-level flows over the rainy regions, as well as
northwestward migration of the 200-mb Tibetan anticyclone. Analysis of the
heat sources and sinks in various regions illustrates the prominent role of
condensational heating in the local energy budget during the mature phases
of monsoon development. In accord with observations, the simulated monsoon
rains in the East Asian sector are organized about zonally elongated
“mei-yu–baiu” (plum rain) systems. These precipitation features advance to
higher latitudes during the June–July period, in conjunction with
displacements of the axis of the low-level anticyclone over the subtropical
western Pacific. A detailed case study is performed on a prominent rainy
episode in the simulation. The model is capable of reproducing the observed
intense gradients in temperature, humidity, and moist static stability in
the vicinity of the mei-yu–baiu front, as well as the spatial relationships
between the rainband and the three-dimensional flow field. The axis of the
mei-yu–baiu rainband in this event is aligned with the trajectory of a
succession of mesoscale cyclonic vortices, which originate from southwestern
China and travel northeastward over the Yangtze River basin.
A model diagnosis has been performed on the nocturnal Great Plains low-level jet (LLJ), which is one of the key elements of the warm season regional climate over North America. The horizontal–vertical structure, diurnal phase, and amplitude of the LLJ are well simulated by an atmospheric general circulation model (AGCM), thus justifying a reevaluation of the physical mechanisms for the formation of the LLJ based on output from this model. A diagnosis of the AGCM data confirms that two planetary boundary layer (PBL) processes, the diurnal oscillation of the pressure gradient force and of vertical diffusion, are of comparable importance in regulating the inertial oscillation of the winds, which leads to the occurrence of maximum LLJ strength during nighttime. These two processes are highlighted in the theories for the LLJ proposed by Holton (1967) and Blackadar (1957). A simple model is constructed in order to study the relative roles of these two mechanisms. This model incorporates the diurnal variation of the pressure gradient force and vertical diffusion coefficients as obtained from the AGCM simulation. The results reveal that the observed diurnal phase and amplitude of the LLJ can be attributed to the combination of these two mechanisms. The LLJ generated by either Holton’s or Blackadar’s mechanism alone is characterized by an unrealistic meridional phase shift and weaker amplitude.
It is also shown that the diurnal phase of the LLJ exhibits vertical variations in the PBL, more clearly at higher latitudes, with the upper PBL wind attaining a southerly peak several hours earlier than the lower PBL. The simple model demonstrates that this phase tilt is due mainly to sequential triggering of the inertial oscillation from upper to lower PBL when surface cooling commences after sunset. At lower latitudes, due to the change of orientation of prevailing mean wind vectors and the longer inertial period, the inertial oscillation in the lower PBL tends to be interrupted by strong vertical mixing in the following day, whereas in the upper PBL, the inertial oscillation can proceed in a low-friction environment for a relatively longer duration. Thus, the vertical phase tilt initiated at sunset is less evident at lower latitudes.
Lee, M-I, S D Schubert, M J Suarez, Isaac M Held, Arun Kumar, T L Bell, J-K E Schemm, Ngar-Cheung Lau, Jeff J Ploshay, H-K Kim, and S-H Yoo, May 2007: Sensitivity to horizontal resolution in the AGCM simulations of warm season diurnal cycle of precipitation over the United States and Northern Mexico. Journal of Climate, 20(9), DOI:10.1175/JCLI4090.1. Abstract
This study examines the sensitivity of the North American warm season diurnal cycle of precipitation to changes in horizontal resolution in three atmospheric general circulation models, with a primary focus on how the parameterized moist processes respond to improved resolution of topography and associated local/regional circulations on the diurnal time scale. It is found that increasing resolution (from approximately 2° to ½° in latitude–longitude) has a mixed impact on the simulated diurnal cycle of precipitation. Higher resolution generally improves the initiation and downslope propagation of moist convection over the Rockies and the adjacent Great Plains. The propagating signals, however, do not extend beyond the slope region, thereby likely contributing to a dry bias in the Great Plains. Similar improvements in the propagating signals are also found in the diurnal cycle over the North American monsoon region as the models begin to resolve the Gulf of California and the surrounding steep terrain. In general, the phase of the diurnal cycle of precipitation improves with increasing resolution, though not always monotonically. Nevertheless, large errors in both the phase and amplitude of the diurnal cycle in precipitation remain even at the highest resolution considered here. These errors tend to be associated with unrealistically strong coupling of the convection to the surface heating and suggest that improved simulations of the diurnal cycle of precipitation require further improvements in the parameterizations of moist convection processes.
Lee, M-I, S D Schubert, M J Suarez, Isaac M Held, Ngar-Cheung Lau, Jeff J Ploshay, Arun Kumar, H-K Kim, and J-K E Schemm, June 2007: An analysis of the warm-season diurnal cycle over the continental United States and Northern Mexico in general circulation models. Journal of Hydrometeorology, 8(3), DOI:10.1175/JHM581.1. Abstract
The diurnal cycle of warm-season rainfall over the continental United States and northern Mexico is analyzed in three global atmospheric general circulation models (AGCMs) from NCEP, GFDL, and the NASA Global Modeling Assimilation Office (GMAO). The results for each model are based on an ensemble of five summer simulations forced with climatological sea surface temperatures.
Although the overall patterns of time-mean (summer) rainfall and low-level winds are reasonably well simulated, all three models exhibit substantial regional deficiencies that appear to be related to problems with the diurnal cycle. Especially prominent are the discrepancies in the diurnal cycle of precipitation over the eastern slopes of the Rocky Mountains and adjacent Great Plains, including the failure to adequately capture the observed nocturnal peak. Moreover, the observed late afternoon–early evening eastward propagation of convection from the mountains into the Great Plains is not adequately simulated, contributing to the deficiencies in the diurnal cycle in the Great Plains. In the southeast United States, the models show a general tendency to rain in the early afternoon—several hours earlier than observed. Over the North American monsoon region in the southwest United States and northern Mexico, the phase of the broad-scale diurnal convection appears to be reasonably well simulated, though the coarse resolution of the runs precludes the simulation of key regional phenomena.
All three models employ deep convection schemes that assume fundamentally the same buoyancy closure based on simplified versions of the Arakawa–Schubert scheme. Nevertheless, substantial differences between the models in the diurnal cycle of convection highlight the important differences in their implementations and interactions with the boundary layer scheme. An analysis of local diurnal variations of convective available potential energy (CAPE) shows an overall tendency for an afternoon peak—a feature well simulated by the models. The simulated diurnal cycle of rainfall is in phase with the local CAPE variation over the southeast United States and the Rocky Mountains where the local surface boundary forcing is important in regulating the diurnal cycle of convection. On the other hand, the simulated diurnal cycle of rainfall tends to be too strongly tied to CAPE over the Great Plains, where the observed precipitation and CAPE are out of phase, implying that free atmospheric large-scale forcing plays a more important role than surface heat fluxes in initiating or inhibiting convection.
Multicentury integrations from two global coupled ocean–atmosphere–land–ice models [Climate Model versions 2.0 (CM2.0) and 2.1 (CM2.1), developed at the Geophysical Fluid Dynamics Laboratory] are described in terms of their tropical Pacific climate and El Niño–Southern Oscillation (ENSO). The integrations are run without flux adjustments and provide generally realistic simulations of tropical Pacific climate. The observed annual-mean trade winds and precipitation, sea surface temperature, surface heat fluxes, surface currents, Equatorial Undercurrent, and subsurface thermal structure are well captured by the models. Some biases are evident, including a cold SST bias along the equator, a warm bias along the coast of South America, and a westward extension of the trade winds relative to observations. Along the equator, the models exhibit a robust, westward-propagating annual cycle of SST and zonal winds. During boreal spring, excessive rainfall south of the equator is linked to an unrealistic reversal of the simulated meridional winds in the east, and a stronger-than-observed semiannual signal is evident in the zonal winds and Equatorial Undercurrent.
Both CM2.0 and CM2.1 have a robust ENSO with multidecadal fluctuations in amplitude, an irregular period between 2 and 5 yr, and a distribution of SST anomalies that is skewed toward warm events as observed. The evolution of subsurface temperature and current anomalies is also quite realistic. However, the simulated SST anomalies are too strong, too weakly damped by surface heat fluxes, and not as clearly phase locked to the end of the calendar year as in observations. The simulated patterns of tropical Pacific SST, wind stress, and precipitation variability are displaced 20°–30° west of the observed patterns, as are the simulated ENSO teleconnections to wintertime 200-hPa heights over Canada and the northeastern Pacific Ocean. Despite this, the impacts of ENSO on summertime and wintertime precipitation outside the tropical Pacific appear to be well simulated. Impacts of the annual-mean biases on the simulated variability are discussed.
for climate research developed at the Geophysical Fluid Dynamics Laboratory (GFDL) are presented. The atmosphere model, known as AM2, includes a new gridpoint dynamical core, a prognostic cloud scheme, and a multispecies aerosol climatology, as well as components from previous models used at GFDL. The land model, known as LM2, includes soil sensible and latent heat storage, groundwater storage, and stomatal resistance. The performance of the coupled model AM2–LM2 is evaluated with a series of prescribed sea surface temperature (SST) simulations. Particular focus is given to the model's climatology and the characteristics of interannual variability related to E1 Niño– Southern Oscillation (ENSO).
One AM2–LM2 integration was performed according to the prescriptions of the second Atmospheric Model Intercomparison Project (AMIP II) and data were submitted to the Program for Climate Model Diagnosis and Intercomparison (PCMDI). Particular strengths of AM2–LM2, as judged by comparison to other models participating in AMIP II, include its circulation and distributions of precipitation. Prominent problems of AM2– LM2 include a cold bias to surface and tropospheric temperatures, weak tropical cyclone activity, and weak tropical intraseasonal activity associated with the Madden–Julian oscillation.
An ensemble of 10 AM2–LM2 integrations with observed SSTs for the second half of the twentieth century permits a statistically reliable assessment of the model's response to ENSO. In general, AM2–LM2 produces a realistic simulation of the anomalies in tropical precipitation and extratropical circulation that are associated with ENSO.
Ploshay, Jeff J., and Jeffrey L Anderson, 2002: Large sensitivity to initial conditions in seasonal predictions with a coupled ocean-atmosphere general circulation model. Geophysical Research Letters, 29(8), DOI:10.1029/2000GL012710. Abstract PDF
An ensemble of one-year forecasts differing only in details of the atmospheric initial conditions was produced with a coupled ocean-atmosphere general circulation model (GCM) in order to investigate the predictability of the coupled system. For some ocean initial conditions, the evolution of the tropical Pacific ocean thermal structure seems to be relatively deterministic for lead times out to one year. However, there are other ocean initial conditions, mostly in the mid 1990's for which coupled model forecasts of the tropical Pacific are much more sensitive to details of the atmosphere initial conditions. In some cases, the ensemble forecasts appear to split, with some ensemble members predicting El Niño-like conditions, and others predicting La Niña. Very large ensembles were run for several of these cases. Very slight perturbations added to the atmospheric initial conditions led to large spread in predicted SST anomalies in some years. These are model results; however, they do suggest the possibility that seasonal predictions of the coupled tropical system may be highly non-deterministic in some years.
Anderson, Jeffrey L., and Jeff J Ploshay, 2000: Impact of initial conditions on seasonal simulations with an atmospheric general circulation model. Quarterly Journal of the Royal Meteorological Society, 126(567), 2241-2264. Abstract PDF
Many previous studies have examined the use of very long integrations of atmospheric general circulation models (AGCMs) forced by observed sea surface temperatures (SSTs) as proxies for seasonal atmospheric predictions. These long simulations explore a boundary-value problem in which significant deviations from the model's long-term climatology must be a result of the SST forcing. Seasonal lead simulations starting with observed initial conditions (ICs) for the atmosphere and land surface while retaining observed SST forcing are an intermediate step between the pure boundary-value problem and the pure initial-value forecast problem in which SSTs are also predicted. As part of the Dynamical Seasonal Prediction (DSP) experiment, an ensemble of AGCM integrations with observed atmospheric ICs and model climatology land surface ICs was integrated from mid-December through March for 16 years. These DSP simulation ensembles are compared to ensembles of long boundary-value simulations from the same AGCM in a perfect-model setting (no comparisons of simulations to observations are attempted). Significant differences must be due to the impact of the DSP ICs. Surprisingly large and long-lived differences are found in both the mean and the variance of the ensembles. Many appear to occur because the ICs of the DSP runs are inconsistent with the AGCM climatology; an extended period of model 'spinup' is the result. Some differences are related to local impacts of the land surface ICs while others, like shifts in the distribution of tropical precipitation and a cooling of the northern hemisphere, are less obviously related to the ICs. The results suggest that care will be needed when inserting observed ICs into seasonal predictions in order to avoid the long-term effects of model spin-up.
Shukla, J, Jeffrey L Anderson, D Baumhefner, Y Chang, E Kalnay, L Marx, T N Palmer, D Paolino, and Jeff J Ploshay, et al., November 2000: Dynamical Seasonal Prediction. Bulletin of the American Meteorological Society, 81(11), DOI:10.1175/1520-0477(2000)081<2593:DSP>2.3.CO;2. Abstract
Dynamical Seasonal Prediction (DSP) is an informally coordinated multi-institution research project to investigate the predictability of seasonal mean atmospheric circulation and rainfall. The basic idea is to test the feasibility of extending the technology of routine numerical weather prediction beyond the inherent limit of deterministic predictability of weather to produce numerical climate predictions using state-of-the-art global atmospheric models. Atmospheric general circulation models (AGCMs) either forced by predicted sea surface temperature (SST) or as part of a coupled forecast system have shown in the past that certain regions of the extratropics, in particular, the Pacific–North America (PNA) region during Northern Hemisphere winter, can be predicted with significant skill especially during years of large tropical SST anomalies. However, there is still a great deal of uncertainty about how much the details of various AGCMs impact conclusions about extratropical seasonal prediction and predictability.
DSP is designed to compare seasonal simulation and prediction results from five state-of-the-art U.S. modeling groups (NCAR, COLA, GSFC, GFDL, NCEP) in order to assess which aspects of the results are robust and which are model dependent. The initial emphasis is on the predictability of seasonal anomalies over the PNA region. This paper also includes results from the ECMWF model, and historical forecast skill over both the PNA region and the European region is presented for all six models.
It is found that with specified SST boundary conditions, all models show that the winter season mean circulation anomalies over the Pacific–North American region are highly predictable during years of large tropical sea surface temperature anomalies. The influence of large anomalous boundary conditions is so strong and so reproducible that the seasonal mean forecasts can be given with a high degree of confidence. However, the degree of reproducibility is highly variable from one model to the other, and quantities such as the PNA region signal to noise ratio are found to vary significantly between the different AGCMs. It would not be possible to make reliable estimates of predictability of the seasonal mean atmosphere circulation unless causes for such large differences among models are understood.
Anderson, Jeffrey L., H van den Dool, A Barnston, W Chen, William F Stern, and Jeff J Ploshay, 1999: Present-day capabilities of numerical and statistical models for atmospheric extratropical seasonal simulation and prediction. Bulletin of the American Meteorological Society, 80(7), 1349-1361. Abstract PDF
A statistical model and extended ensemble integrations of two atmospheric general circulation models (GCMs) are used to simulate the extratropical atmospheric response to forcing by observed SSTs for the years 1980 through 1988. The simulations are compared to observations using the anomaly correlation and root-mean-square error of the 700-hPa height field over a region encompassing the extratropical North Pacific Ocean and most of North America. On average, the statistical model is found to produce considerably better simulations than either numerical model, even when simple statistical corrections are used to remove systematic errors from the numerical model simulations. In the mean, the simulation skill is low, but there are some individual seasons for which all three models produce simulations with good skill.
An approximate upper bound to the simulation skill that could be expected from a GCM ensemble, if the model's response to SST forcing is assumed to be perfect, is computed. This perfect model predictability allows one to make some rough extrapolations about the skill that could be expected if one could greatly improve the mean response of the GCMs without significantly impacting the variance of the ensemble. These perfect model predictability skills are better than the statistical model simulations during the summer, but for the winter, present-day statistical forecasts already have skill that is as high as the upper bound for the GCMs. Simultaneous improvements to the GCM mean response and reduction in the GCM ensemble variance would be required for these GCMs to do significantly better than the statistical model in winter. This does not preclude the possibility that, as is presently the case, a statistical blend of GCM and statistical predictions could produce a simulation better than either alone.
Because of the primitive state of coupled ocean-atmosphere GCMs, the vast majority of seasonal predictions currently produced by GCMs are performed using a two-tiered approach in which SSTs are first predicted and then used to force an atmospheric model; this motivates the examination of the simulation problem. However, it is straightforward to use the statistical model to produce true forecasts by changing its predictors from simultaneous to precursor SSTs. An examination of the decrease in skill of the statistical model when changed from simulation to prediction mode is extrapolated to draw conclusions about the skill to be expected from good coupled GCM predictions.
Anderson, Jeffrey L., and Jeff J Ploshay, 1999: Impacts of land surface initial conditions on seasonal lead GCM simulations In Proceedings of the Twenty-Fourth Annual Climate Diagnostics and Prediction Workshop, Springfield, VA, NTIS, 319-322.
Anderson, Jeffrey L., H van den Dool, A Barnston, W Chen, William F Stern, and Jeff J Ploshay, 1998: Capabilities of dynamical and statistical methods for atmospheric extratropical seasonal prediction In Proceedings of the Twenty-Second Annual Climate Diagnostics and Prediction Workshop, Springfield, VA, NTIS, 46-49.
Anderson, Jeffrey L., Richard G Gudgel, and Jeff J Ploshay, 1998: Seasonal-interannual predictions from an ensemble of fully-coupled ocean-atmosphere GCM integrations. In Proceedings of the Twenty-Second Annual Climate Diagnostics and Prediction Workshop, Springfield, VA, NTIS, 18-20.
Miyakoda, Kikuro, Jeff J Ploshay, and Anthony Rosati, 1997: Preliminary study on SST forecast skill associated with the 1982/83 El Niño process, using coupled model data assimilation. Atmosphere-Ocean, 35(1), 469-486. Abstract PDF
A previous study by Rosati et al. (1997) has concluded that the specification of an adequate thermocline structure along the equatorial Pacific ocean is most crucial for El Niño forecasts. In that paper, the oceanic initial condition was generated by a data assimilation (DA) system (Derber and Rosati, 1989). However, the initial condition for the atmospheric part was taken from the National Meteorological Center's (NMC) operational analysis, which was simply attached to the oceanic part for the coupled model forecasts.
In the present paper, both the atmospheric and oceanic initial conditions are generated by a coupled DA system applied to a coupled air-sea general circulation model (GCM). The assimilation for the ocean is performed by the same system as mentioned above, in which the SST (sea surface temperature) and the subsurface temperatures are injected into a 15 vertical level oceanic GCM. The upper boundary condition, such as surface wind stress, is specified by the atmospheric DA. The assimilation for the atmosphere is performed by the continuous injection method of Stern and Ploshay (1992), using an 18 vertical level atmospheric GCM. The lower boundary condition, such as SST, is specified by the oceanic DA. The coupled model assimilations are carried out by switching the DA processes alternately every 6 hours between the ocean and the atmosphere.
The emphases of this study are: firstly, the effect of coupled air-sea model DA on the performance of subsequent forecsts; secondly, the impact of the coupled assimilation on improvement of the "spin-up" behavior of forecasts, i.e., to see whether a smooth start to the forecast is achieved by the coupled model DA process; and thirdly, investigation of the effect that the "spring barrier" has on predictability in the coupled GCM system. Preliminary results indicate that, in order to answer these questions, ensemble forecasts are necessary. Besides, the coupled assimilation could be important in improving the overall behavior of El Niño and La Niña forecasts.
The reanalysis of FGGE [First GARP (Global Atmospheric Research Program) Global Experiment] data for 128 days during two special observing periods has been performed, using an improved data-assimilation system and the revised FGGE level II dataset. The data-assimilation scheme features forward continuous (in time) data injection in both the original and the new systems. However, the major revisions in the new system include a better first guess and a more efficient dynamical balancing for the assimilation of observed data. The results of the implementation of this system are assessed by intercomparisons among the new FGGE analysis of other institutions such as ECMWF (European Centre for Medium-Range Weather Forecasts) and NMC (National Meteorological Center, Washington, D.C.), and also the original GFDL (Geophysical Fluid Dynamics Laboratory) analysis. The quality of the new GFDL analysis is now comparable to those of the other two institutions. However, the moisture analysis appears to be appreciably different, suggesting that the cumulus convection parameterizations and the boundary-layer moisture fluxes in the models are responsible for this discrepancy.
A detailed investigation of the results has been carried out by comparing the analyses with radiosonde observations. This verification reveals that temperature and wind differences have been reduced considerably from the original to the new GFDL analysis; they are now competitive with those of ECMWF and NMC, while with regard to the geopotential height, differences of the GFDL reanalysis are larger than the original GFDL as well as the ECMWF and the NMC. A comparative study is also made with UCLA analyses over Asia in connection with the Indian monsoon. The results indicate that the qualities of both analyses are comparable. The capability of representing Madden-Julian oscillations in the reanalysis and in the ECMWF and old GFDL analysis is investigated by comparing with satellite observations. It is revealed that these oscillations are successfully reproduced by the new analysis; however, the agreement with the satellite data is not quite satisfactory. The utilization of satellite-observed wind (satobs) and aircraft data (aireps) in the data assimilation needs particular care. It appears that the quality control of these data in the GFDL reanalysis is too restrictive; in other words, the toss-out criterion of wind data is too small. A consequence of the failure to accept some single-level data turns out to be a fairly large discrepancy in representing the maximum wind speed in the analysis. It is also discussed that the current forward continuous-injection scheme is not adequate to obtain diabatic quantities for the archive.
Major revisions to the Geophysical Fluid Dynamics Laboratory's (GFDL) continuous data assimilation system have been implemented and tested. Shortcomings noted during the original processing of data from FGGE [First GARP (Global Atmospheric Research Program) Global Experiment ] served as the basis for these improvements. This new system has been used to reanalyze the two FGGE special observing periods. The main focus here will be on assessing the changes to the assimilation system using comparisons of rerun test results with results from the original FGGE processing.
The key new features in the current system include: a reduction in the assimilation cycle from 12 to 6 h; the use of a 6-h forecast first guess for the OI (optimum-interpolation analysis) as opposed to the previous use of persistence as a first guess; an extension of the OI search range from 250 to 500 km with an increase in the maximum number of observations used per analysis point from 8 to 12; the introduction of incremental linear normal-mode initialization, eliminating the periodic nonlinear normal-mode initialization; and an increase in the horizontal resolution of the assimilating model from 30 waves to 42 waves, rhomboidally truncated. Tests of the new system show a significant reduction in the level of noise, improved consistency between mass and momentum analyses, and a better fit of the analyses to observations. In addition, the new system has demonstrated a greater ability to resolve rapidly moving and deepening transient features, with an indication of less rejection of surface pressure data.
In addition to the quantities archived during the original FGGE data processing, components of diabatic heating from the assimilating model have also been archived. They should be used with caution to the extent that they reflect model bias and spinup in addition to real features of the general circulation.
A series of one-month forecasts were carried out for eight January cases, using a particular prediction model and prescribing climatological sea-surface temperature as the boundary condition. Each forecast is a stochastic prediction that consists of three individual integrations. These forecasts start with observed initial conditions derived from datasets of three meteorological centers. The forecast skill was assessed with respect to time means of variables based on the ensemble average of three forecasts. The time of space filter is essential to suppress unpredictable components of atmospheric variabilities and thereby to make an attempt at extending the limit of predictability. The circulation patterns of the three individual integrations tend to be similar to each other on the one-month time scale, implying that forecasts for the 10 day (or 20 day) means are not fully stochastic.The overall results indicate that the 10-day mean height prognoses resemble observations very well in the first ten days, and then start to lose similarity to real states, and yet there is some recognizable skill in the last ten days of the month. The main interests in this study are the feasibility of one-month forecasts, the adequacy of initial conditions produced by a particular data assimilation, and the growth of stochastic uncertainty. An outstanding problem turns out to be a considerable degree of systematic error included in the prediction model, which is now known to be "climate drift". Forecast errors are largely due to the model's systematic bias. Thus, forecast skill scores are substantially raised if the final prognoses are adjusted for the model's known climatic drift.
Stern, William F., R T Pierrehumbert, Joseph J Sirutis, Jeff J Ploshay, and Kikuro Miyakoda, 1986: Recent developments in the GFDL extended-range forecasting system In Short- and Medium-Range Numerical Weather Prediction, Collection of papers presented at the WMO/IUGG NWP Symposium, World Meteorological Organization, 359-363. Abstract
An assessment is made of the areas of focus for improving extended-range forecasting. Two topics currently being researched involve the reduction of systematic error by improving a GCM's accuracy and the refinement of the transition between the data assimilation phase and the forecasting phase.
Subgrid-scale orographic parameterizations have been the subject of recent model improvement activities. Results are shown for an envelope orography with an N48L9 gridpoint model and using a mountain gravity wave drag scheme with an R42L18 spectral model. In both cases there is an encouraging reduction in the systematic errors.
Proper initialization of tropical features, i.e., 40-50 day waves, may be crucial for extended-range predictions in the extra-tropics as well as the tropics. Using a continuous data assimilation scheme the 40-50 day oscillations in the tropics appear to be well maintained from the assimilation to the forecast phase. However, the assimilation system underestimates precipitation and evaporation rates.
Daley, R, Jeff J Ploshay, and Kikuro Miyakoda, et al., 1985: Objective analysis and assimilation techniques used for the production of FGGE IIIb analyses. Bulletin of the American Meteorological Society, 66(5), 532-538. Abstract
A set of tables has been prepared which allows side-by-side comparison of the characteristics of six data assimilation systems (ECMWF, GFDL, GLAS, NMC, FSU, and NEPRF) used to produce FGGE IIIb analyses.
Miyakoda, Kikuro, Joseph J Sirutis, and Jeff J Ploshay, 1985: Monthly forecast experiment: preliminary report In Numerical Long-Range Forecast Evaluation Numerical Long-Range Forecasting Errors Monthly Forecasts, Washington, DC, National Academy Press, 292-296. Abstract
An experiment on monthly forecasts with eight winter cases was conducted by using a 1980 general circulation model that incorporates a set of subgrid-scale physics characterized by Mellor-Yamada turbulence closure (hierarchy level 2.5), the Monin-Obukhov parameterization for the layer next to the ground surface, Manabe's cumulus parameterization, and the soil heat conduction. The cases are for January from 1977 to 1983; they include the extraordinarily severe winter of 1977 and the most pronounced El Niño year of 1983. Graphs show correlation coefficients of 500-mb geopotential height anomalies (the deviation from climatology) and of the 1000-mb geopotential height anomalies between the predictions and observations for the Northern Hemispheric domain (90-25 degrees N). The study indicates that the 10- or 20- day height prognoses resemble the observations well in the first 10 days and then rapidly lose the similarity; yet there is some recognized skill, although marginal in the last 10 or 20 days of the month. The skill scores for the 1000-mb level are consistently better than those for the 500-mb level. This feature appears opposite to that for the daily weather forecasts and may suggest how forecast errors propagate in the vertical.
This is a progress report and follow-up of "Three cases of one-month GCM forecasts" (Caverly, et al., 1981). Each case includes an ensemble of three individual integrations, starting with three different initial conditions produced at GFDL, NMC, and ECMWF and prescribing the climatological sea surface temperature as the lower boundary condition. Monthly forecasts with the N48L9-F model and examples of verification statistics are presented. The experiment with four winter cases indicates that there is some skill in the mean height prognosis.
Ploshay, Jeff J., Robert K White, and Kikuro Miyakoda, 1983: FGGE Level III-B Daily Global Analyses Part I (Dec 1978 - Feb 1979), NOAA Data Report ERL GFDL-1, Rockville, MD: NOAA, 278 pp.
Ploshay, Jeff J., Robert K White, and Kikuro Miyakoda, 1983: FGGE Level III-B Daily Global Analyses Part II (Mar 1979 - May 1979), NOAA Data Report ERL GFDL-2, Rockville, MD: NOAA, 285 pp.
Ploshay, Jeff J., Robert K White, and Kikuro Miyakoda, 1983: FGGE Level III-B Daily Global Analyses Part III (Jun 1979 - Aug 1979), NOAA Data Report ERL GFDL-3, Rockville, MD: NOAA, 285 pp.
Ploshay, Jeff J., Robert K White, and Kikuro Miyakoda, 1983: FGGE Level III-B Daily Global Analyses Part IV (Sept 1979 - Nov 1979), NOAA Data Report ERL GFDL-4, Rockville, MD: NOAA, 282 pp.
Krishnamurti, T N., Hua Lu Pan, Chia Bo Chang, and Jeff J Ploshay, et al., October 1979: Numerical weather prediction for GATE. Quarterly Journal of the Royal Meteorological Society, 105(446), DOI:10.1002/qj.49710544617. Abstract
The results of a few numerical weather prediction experiments that utilize GATE data are presented. Aside from some simple forecasts based on single‐level models, experiments with a multilevel model examine the roles of some of the physical processes. They include the influences of diurnally varying radiative effects, effects of cloud feedback on shortwave and longwave radiation, the heat balance of land surfaces, and the influence of West African orography in the 3‐ to 4‐day range of numerical weather prediction. Data sources for these experiments include surface and upper air observations from the World Weather Watch, GATE ships, upper winds from commercial aircraft, low‐ and high‐level cloud winds from geostationary satellites, winds from GATE research aircraft, and special surface observations from a ship data collection. The data analysis is based on a successive correction method where a subjective interface is also incorporated. The initial data are further subjected to a static as well as a dynamic initialization process.
The major results are that single‐level forecasts are very promising with this data set if they are carried out at 700 mb. The multilevel adiabatic experiment produces very poor predictions due to lack of adequate vertical coupling between the lower and the upper troposphere and due to its inability to describe the broad‐scale monsoons. The diurnally varying heat balance of the earth's surface and cloudiness in radiative calculations appear to be important in the 3‐ to 4‐day range of prediction of African disturbances. In this paper we show some interesting vertical structure diagrams of African waves based on the results of numerical weather prediction, which exhibit a close similarity to the structures of composite GATE disturbances constructed by Professor R. J. Reed and his associates.
