We document the development and simulation characteristics of the next generation modeling system for seasonal to decadal prediction and projection at the Geophysical Fluid Dynamics Laboratory (GFDL). SPEAR (Seamless System for Prediction and EArth System Research) is built from component models recently developed at GFDL ‐ the AM4 atmosphere model, MOM6 ocean code, LM4 land model and SIS2 sea ice model. The SPEAR models are specifically designed with attributes needed for a prediction model for seasonal to decadal time scales, including the ability to run large ensembles of simulations with available computational resources. For computational speed SPEAR uses a coarse ocean resolution of approximately 1.0o (with tropical refinement). SPEAR can use differing atmospheric horizontal resolutions ranging from 1o to 0.25o. The higher atmospheric resolution facilitates improved simulation of regional climate and extremes. SPEAR is built from the same components as the GFDL CM4 and ESM 4 models, but with design choices geared toward seasonal to multidecadal physical climate prediction and projection. We document simulation characteristics for the time‐mean climate, aspects of internal variability, and the response to both idealized and realistic radiative forcing change. We describe in greater detail one focus of the model development process that was motivated by the importance of the Southern Ocean to the global climate system. We present sensitivity tests that document the influence of the Antarctic surface heat budget on Southern Ocean ventilation and deep global ocean circulation. These findings were also useful in the development processes for the GFDL CM4 and ESM 4 models.
Seasonal predictions of Arctic sea ice on regional spatial scales are a pressing need for a broad group of stakeholders, however, most assessments of predictability and forecast skill to date have focused on pan-Arctic sea–ice extent (SIE). In this work, we present the first direct comparison of perfect model (PM) and operational (OP) seasonal prediction skill for regional Arctic SIE within a common dynamical prediction system. This assessment is based on two complementary suites of seasonal prediction ensemble experiments performed with a global coupled climate model. First, we present a suite of PM predictability experiments with start dates spanning the calendar year, which are used to quantify the potential regional SIE prediction skill of this system. Second, we assess the system’s OP prediction skill for detrended regional SIE using a suite of retrospective initialized seasonal forecasts spanning 1981–2016. In nearly all Arctic regions and for all target months, we find a substantial skill gap between PM and OP predictions of regional SIE. The PM experiments reveal that regional winter SIE is potentially predictable at lead times beyond 12 months, substantially longer than the skill of their OP counterparts. Both the OP and PM predictions display a spring prediction skill barrier for regional summer SIE forecasts, indicating a fundamental predictability limit for summer regional predictions. We find that a similar barrier exists for pan-Arctic sea–ice volume predictions, but is not present for predictions of pan-Arctic SIE. The skill gap identified in this work indicates a promising potential for future improvements in regional SIE predictions.
Dynamical prediction systems have shown potential to meet the emerging need for seasonal forecasts of regional Arctic sea ice. Observationally constrained initial conditions are a key source of skill for these predictions, but the direct influence of different observation types on prediction skill has not yet been systematically investigated. In this work, we perform a hierarchy of Observing System Experiments with a coupled global data assimilation and prediction system to assess the value of different classes of oceanic and atmospheric observations for seasonal sea-ice predictions in the Barents Sea. We find notable skill improvements due to the inclusion of both sea-surface temperature (SST) satellite observations and subsurface conductivity-temperature-depth (CTD) measurements. The SST data is found to provide the crucial source of interannual variability, whereas the CTD data primarily provide climatological and trend improvements. Analysis of the Barents Sea ocean heat budget suggests that ocean heat content anomalies in this region are driven by surface heat fluxes on seasonal timescales.
The Caribbean low-level jet (CLLJ) is an important component of the atmospheric circulation over the Intra-Americas Sea (IAS) which impacts the weather and climate both locally and remotely. It influences the rainfall variability in the Caribbean, Central America, northern South America, the tropical Pacific and the continental Unites States through the transport of moisture. We make use of high-resolution coupled and uncoupled models from the Geophysical Fluid Dynamics Laboratory (GFDL) to investigate the simulation of the CLLJ and its teleconnections and further compare with low-resolution models. The high-resolution coupled model FLOR shows improvements in the simulation of the CLLJ and its teleconnections with rainfall and SST over the IAS compared to the low-resolution coupled model CM2.1. The CLLJ is better represented in uncoupled models (AM2.1 and AM2.5) forced with observed sea-surface temperatures (SSTs), emphasizing the role of SSTs in the simulation of the CLLJ. Further, we determine the forecast skill for observed rainfall using both high- and low-resolution predictions of rainfall and SSTs for the July–August–September season. We determine the role of statistical correction of model biases, coupling and horizontal resolution on the forecast skill. Statistical correction dramatically improves area-averaged forecast skill. But the analysis of spatial distribution in skill indicates that the improvement in skill after statistical correction is region dependent. Forecast skill is sensitive to coupling in parts of the Caribbean, Central and northern South America, and it is mostly insensitive over North America. Comparison of forecast skill between high and low-resolution coupled models does not show any dramatic difference. However, uncoupled models show improvement in the area-averaged skill in the high-resolution atmospheric model compared to lower resolution model. Understanding and improving the forecast skill over the IAS has important implications for highly vulnerable nations in the region.
Scaife, Adam A., L Ferranti, Oscar Alves, Panos Athanasiadis, J Baehr, M Dequé, T Dippe, Nick Dunstone, D Fereday, Richard G Gudgel, R J Greatbatch, Leon Hermanson, Yukiko Imada, S Jain, Arun Kumar, C MacLachlan, William J Merryfield, Wolfgang A Müller, Hong-Li Ren, Doug Smith, Yuhei Takaya, Gabriel A Vecchi, and Xiaosong Yang, February 2019: Tropical rainfall predictions from multiple seasonal forecast systems. International Journal of Climatology, 39(2), DOI:10.1002/joc.5855. Abstract
We quantify seasonal prediction skill of tropical winter rainfall in 14 climate forecast systems. High levels of seasonal prediction skill exist for year‐to‐year rainfall variability in all tropical ocean basins. The tropical East Pacific is the most skilful region, with very high correlation scores, and the tropical West Pacific is also highly skilful. Predictions of tropical Atlantic and Indian Ocean rainfall show lower but statistically significant scores.
We compare prediction skill (measured against observed variability) with model predictability (using single forecasts as surrogate observations). Model predictability matches prediction skill in some regions but it is generally greater, especially over the Indian Ocean. We also find significant inter‐basin connections in both observed and predicted rainfall. Teleconnections between basins due to El Niño–Southern Oscillation (ENSO) appear to be reproduced in multi‐model predictions and are responsible for much of the prediction skill. They also explain the relative magnitude of inter‐annual variability, the relative magnitude of predictable rainfall signals and the ranking of prediction skill across different basins.
These seasonal tropical rainfall predictions exhibit a severe wet bias, often in excess of 20% of mean rainfall. However, we find little direct relationship between bias and prediction skill. Our results suggest that future prediction systems would be best improved through better model representation of inter‐basin rainfall connections as these are strongly related to prediction skill, particularly in the Indian and West Pacific regions. Finally, we show that predictions of tropical rainfall alone can generate highly skilful forecasts of the main modes of extratropical circulation via linear relationships that might provide a useful tool to interpret real‐time forecasts.
Responses of tropical cyclones (TCs) to CO2 doubling are explored using coupled global climate models (GCMs) with increasingly refined atmospheric/land horizontal grids (~ 200 km, ~ 50 km and ~ 25 km). The three models exhibit similar changes in background climate fields thought to regulate TC activity, such as relative sea surface temperature (SST), potential intensity, and wind shear. However, global TC frequency decreases substantially in the 50 km model, while the 25 km model shows no significant change. The ~ 25 km model also has a substantial and spatially-ubiquitous increase of Category 3–4–5 hurricanes. Idealized perturbation experiments are performed to understand the TC response. Each model’s transient fully-coupled 2 × CO2 TC activity response is largely recovered by “time-slice” experiments using time-invariant SST perturbations added to each model’s own SST climatology. The TC response to SST forcing depends on each model’s background climatological SST biases: removing these biases leads to a global TC intensity increase in the ~ 50 km model, and a global TC frequency increase in the ~ 25 km model, in response to CO2-induced warming patterns and CO2 doubling. Isolated CO2 doubling leads to a significant TC frequency decrease, while isolated uniform SST warming leads to a significant global TC frequency increase; the ~ 25 km model has a greater tendency for frequency increase. Global TC frequency responds to both (1) changes in TC “seeds”, which increase due to warming (more so in the ~ 25 km model) and decrease due to higher CO2 concentrations, and (2) less efficient development of these“seeds” into TCs, largely due to the nonlinear relation between temperature and saturation specific humidity.
Improving the seasonal prediction of tropical cyclone (TC) activity demands a robust analysis of the prediction skill and the inherent predictability of TC activity. Using the resampling technique, this study analyzes a state‐of‐the‐art prediction system and offers a robust assessment of when and where the seasonal prediction of TC activity is skillful. We found that uncertainties of initial conditions affect the predictions and the skill evaluation significantly. The sensitivity of predictions to initial conditions also suggests that landfall and high‐latitude activity are inherently harder to predict. The lower predictability is consistent with the relatively low prediction skill in these regions. Additionally, the lower predictability is largely related to the atmospheric environment rather than the sea surface temperature, at least for the predictions initialized shortly before the hurricane season. These findings suggest the potential for improving the seasonal TC prediction and will help the development of the next‐generation prediction systems.
Mountain snowpack in the western United States provides a natural reservoir for cold season precipitation; variations in snowpack influence warm season water supply, wildfire risk, ecology, and industries like agriculture dependent on snow and downstream water availability. Efforts to understand snowpack variability have predominantly been focused on either weekly (weather) or decadal to centennial (climate variability and change) timescales. We focus on a timescale between these ranges by demonstrating that a global climate model suite can provide snowpack predictions 8 months in advance. The predictions from climate models outperform statistical methods from observations alone. Our results show that seasonal hydroclimate predictions are possible and highlight areas for future prediction system improvements.
Extratropical transition can extend the threat of tropical cyclones into the mid‐latitudes, and modify it through expansion of rainfall and wind fields. Despite the scientific and socioeconomic interest, the seasonal forecast of extratropical transition has received little attention. The GFDL High‐Resolution Forecast‐Oriented Low Ocean Resolution (FLOR) model (HiFLOR) shows skill in seasonal forecasts of tropical cyclone frequency as well as major hurricanes. A July‐initialized twelve‐member ensemble retrospective seasonal forecast experiment with HiFLOR in the North Atlantic is conducted, representing one of the very first attempts to predict the extratropical transition activity months in advance. HiFLOR exhibits retrospective skill in seasonal forecasts of basin‐wide and regional ET activity relative to best track and reanalysis data. In contrast, the skill of HiFLOR in predictions of non‐ET activity is limited. Future work targeted at improved predictions of non‐ET storms provides a path for enhanced TC activity forecasting.
We explore factors potentially linked to the enhanced major hurricane activity in the Atlantic during 2017. Using a suite of high-resolution model experiments, we show that the increase in 2017 major hurricanes was not primarily caused by La Niña conditions in the Pacific Ocean, but mainly by pronounced warm sea surface conditions in the tropical North Atlantic. It is further shown that, in the future, a similar pattern of North Atlantic surface warming, superimposed upon long-term increasing sea surface temperature from increases in greenhouse gas concentrations and decreases in aerosols, will likely lead to even higher numbers of major hurricanes. The key factor controlling Atlantic major hurricane activity appears to be how much the tropical Atlantic warms relative to the rest of the global ocean.
A “typical” El Niño leads to wet (dry) wintertime anomalies over the southern (northern) half of the Western United States (WUS). However, during the strong El Niño of 2015/16, the WUS winter precipitation pattern was roughly opposite to this canonical (average of the record) anomaly pattern. To understand why this happened, and whether it was predictable, we use a suite of high-resolution seasonal prediction experiments with coupled climate models. We find that the unusual 2015/16 precipitation pattern was predictable at zero-lead time horizon when the ocean/atmosphere/land components were initialized with observations. However, when the ocean alone is initialized the coupled model fails to predict the 2015/16 pattern, although ocean initial conditions alone can reproduce the observed WUS precipitation during the 1997/98 strong El Niño. Further observational analysis shows that the amplitudes of the El Niño induced tropical circulation anomalies during 2015/16 were weakened by about 50% relative to those of 1997/98. This was caused by relative cold (warm) anomalies in the eastern (western) tropical Pacific suppressing (enhancing) deep convection anomalies in the eastern (western) tropical Pacific during 2015/16. The reduced El Niño teleconnection led to a weakening of the subtropical westerly jet over the southeast North Pacific and southern WUS, resulting in the unusual 2015/16 winter precipitation pattern over the WUS. This study highlights the importance of initial conditions not only in the ocean, but in the land and atmosphere as well, for predicting the unusual El Niño teleconnection and its influence on the winter WUS precipitation anomalies during 2015/16.
Due to its persistence on seasonal timescales, Arctic sea-ice thickness (SIT) is a potential source of predictability for summer sea-ice extent (SIE). New satellite observations of SIT represent an opportunity to harness this potential predictability via improved thickness initialization in seasonal forecast systems. In this work, the evolution of Arctic sea-ice volume anomalies is studied using a 700-year control integration and a suite of initialized ensemble forecasts from a fully-coupled global climate model. Our analysis is focused on the September sea-ice zone, as this is the region where thickness anomalies have the potential to impact the SIE minimum. The primary finding of this paper is that, in addition to a general decay with time, sea-ice volume anomalies display a summer enhancement, in which anomalies tend to grow between the months of May and July. This summer enhancement is relatively symmetric for positive and negative volume anomalies and peaks in July regardless of the initial month. Analysis of the surface energy budget reveals that the summer volume anomaly enhancement is driven by a positive feedback between the SIT state and the surface albedo. The SIT state affects surface albedo through changes in the sea-ice concentration field, melt-onset date, snow coverage, and ice-thickness distribution, yielding an anomaly in the total absorbed shortwave radiation between May and August, which enhances the existing SIT anomaly. This phenomenon highlights the crucial importance of accurate SIT initialization and representation of ice-albedo feedback processes in seasonal forecast systems.
Recent Arctic sea ice seasonal prediction efforts and forecast skill assessments have primarily focused on pan-Arctic sea-ice extent (SIE). In this work, we move towards stakeholder-relevant spatial scales, investigating the regional forecast skill of Arctic sea ice in a Geophysical Fluid Dynamics Laboratory (GFDL) seasonal prediction system. Using a suite of retrospective initialized forecasts spanning 1981–2015 made with a coupled atmosphere-ocean-sea ice-land model, we show that predictions of detrended regional SIE are skillful at lead times up to 11 months. Regional prediction skill is highly region and target month dependent, and generically exceeds the skill of an anomaly persistence forecast. We show for the first time that initializing the ocean subsurface in a seasonal prediction system can yield significant regional skill for winter SIE. Similarly, as suggested by previous work, we find that sea-ice thickness initial conditions provide a crucial source of skill for regional summer SIE.
This study explores the role of the stratosphere as a source of seasonal predictability of surface climate over Northern Hemisphere extra-tropics both in the observations and climate model predictions. A suite of numerical experiments, including climate simulations and retrospective forecasts, are set up to isolate the role of the stratosphere in seasonal predictive skill of extra-tropical near surface land temperature. We show that most of the lead-0 month spring predictive skill of land temperature over extra-tropics, particularly over northern Eurasia, stems from stratospheric initialization. We further reveal that this predictive skill of extra-tropical land temperature arises from skillful prediction of the Arctic Oscillation (AO). The dynamical connection between the stratosphere and troposphere is also demonstrated by the significant correlation between the stratospheric polar vortex and sea level pressure anomalies, as well as the migration of the stratospheric zonal wind anomalies to the lower troposphere.
The 2015 hurricane season in the Eastern and Central Pacific Oceans (EPO and CPO), particularly around Hawaii, was extremely active – including a record number of tropical cyclones (TCs) and the first instance of three simultaneous Category 4 hurricanes in the EPO and CPO. A strong El Niño developed during the 2015 boreal summer season, and was attributed by some to be the cause of the extreme number of TCs. However, according to a suite of targeted high-resolution model experiments, the extreme 2015 EPO and CPO hurricane season was not primarily induced by the 2015 El Niño’s tropical Pacific warming, but by warming in the subtropical Pacific Ocean. This warming is not typical of El Niño, but rather the “Pacific Meridional Mode (PMM)” superimposed on long-term anthropogenic warming. Although the likelihood of such an extreme year depends on the phase of natural variability, the coupled GCM projects an increase in the frequency of such extremely active TC years over the next few decades for the EPO, CPO, and Hawaii due enhanced subtropical Pacific warming from anthropogenic greenhouse forcing.
The TAO/TRITON array is the cornerstone of the tropical Pacific and ENSO observing system. Motivated by the recent rapid decline of the TAO/TRITON array, the potential utility of TAO/TRITON was assessed for ENSO monitoring and prediction. The analysis focused on the period when observations from Argo floats were also available. We coordinated observing system experiments (OSEs) using the global ocean data assimilation system (GODAS) from the National Centers for Environmental Prediction and the ensemble coupled data assimilation (ECDA) from the Geophysical Fluid Dynamics Laboratory for the period 2004–2011. Four OSE simulations were conducted with inclusion of different subsets of in situ profiles: all profiles (XBT, moorings, Argo), all except the moorings, all except the Argo and no profiles. For evaluation of the OSE simulations, we examined the mean bias, standard deviation difference, root-mean-square difference (RMSD) and anomaly correlation against observations and objective analyses. Without assimilation of in situ observations, both GODAS and ECDA had large mean biases and RMSD in all variables. Assimilation of all in situ data significantly reduced mean biases and RMSD in all variables except zonal current at the equator. For GODAS, the mooring data is critical in constraining temperature in the eastern and northwestern tropical Pacific, while for ECDA both the mooring and Argo data is needed in constraining temperature in the western tropical Pacific. The Argo data is critical in constraining temperature in off-equatorial regions for both GODAS and ECDA. For constraining salinity, sea surface height and surface current analysis, the influence of Argo data was more pronounced. In addition, the salinity data from the TRITON buoys played an important role in constraining salinity in the western Pacific. GODAS was more sensitive to withholding Argo data in off-equatorial regions than ECDA because it relied on local observations to correct model biases and there were few XBT profiles in those regions. The results suggest that multiple ocean data assimilation systems should be used to assess sensitivity of ocean analyses to changes in the distribution of ocean observations to get more robust results that can guide the design of future tropical Pacific observing systems.
Xue, Y, C. Wen, Arun Kumar, Magdalena Alonso Balmaseda, Yosuke Fujii, Oscar Alves, M Martin, Xiaosong Yang, G Vernieres, C Desportes, Tong Lee, I Ascione, Richard G Gudgel, and I Ishikawa, December 2017: A real-time ocean reanalyses intercomparison project in the context of tropical pacific observing system and ENSO monitoring. Climate Dynamics, 49(11-12), DOI:10.1007/s00382-017-3535-y. Abstract
An ensemble of nine operational ocean reanalyses (ORAs) is now routinely collected, and is used to monitor the consistency across the tropical Pacific temperature analyses in real-time in support of ENSO monitoring, diagnostics, and prediction. The ensemble approach allows a more reliable estimate of the signal as well as an estimation of the noise among analyses. The real-time estimation of signal-to-noise ratio assists the prediction of ENSO. The ensemble approach also enables us to estimate the impact of the Tropical Pacific Observing System (TPOS) on the estimation of ENSO-related oceanic indicators. The ensemble mean is shown to have a better accuracy than individual ORAs, suggesting the ensemble approach is an effective tool to reduce uncertainties in temperature analysis for ENSO. The ensemble spread, as a measure of uncertainties in ORAs, is shown to be partially linked to the data counts of in situ observations. Despite the constraints by TPOS data, uncertainties in ORAs are still large in the northwestern tropical Pacific, in the SPCZ region, as well as in the central and northeastern tropical Pacific. The uncertainties in total temperature reduced significantly in 2015 due to the recovery of the TAO/TRITON array to approach the value before the TAO crisis in 2012. However, the uncertainties in anomalous temperature remained much higher than the pre-2012 value, probably due to uncertainties in the reference climatology. This highlights the importance of the long-term stability of the observing system for anomaly monitoring. The current data assimilation systems tend to constrain the solution very locally near the buoy sites, potentially damaging the larger-scale dynamical consistency. So there is an urgent need to improve data assimilation systems so that they can optimize the observation information from TPOS and contribute to improved ENSO prediction.
This study attempts to improve the prediction of western North Pacific (WNP) and East Asia (EA) landfalling tropical cyclones (TCs) using modes of large-scale climate variability [e.g., the Pacific meridional mode (PMM), the Atlantic meridional mode (AMM), and North Atlantic sea surface temperature anomalies (NASST)] as predictors in a hybrid statistical–dynamical scheme, based on dynamical model forecasts with the GFDL Forecast-Oriented Low Ocean Resolution version of CM2.5 with flux adjustments (FLOR-FA). Overall, the predictive skill of the hybrid model for the WNP TC frequency increases from lead month 5 (initialized in January) to lead month 0 (initialized in June) in terms of correlation coefficient and root-mean-square error (RMSE). The hybrid model outperforms FLOR-FA in predicting WNP TC frequency for all lead months. The predictive skill of the hybrid model improves as the forecast lead time decreases, with values of the correlation coefficient increasing from 0.56 for forecasts initialized in January to 0.69 in June. The hybrid models for landfalling TCs over the entire East Asian (EEA) coast and its three subregions [i.e., southern EA (SEA), middle EA (MEA), and northern EA (NEA)] dramatically outperform FLOR-FA. The correlation coefficient between predicted and observed TC landfall over SEA increases from 0.52 for forecasts initialized in January to 0.64 in June. The hybrid models substantially reduce the RMSE of landfalling TCs over SEA and EEA compared with FLOR-FA. This study suggests that the PMM and NASST/AMM can be used to improve statistical/hybrid forecast models for the frequencies of WNP or East Asia landfalling TCs.
This study explores the potential predictability of the Southern Ocean (SO) climate on decadal timescales as represented in the GFDL CM2.1 model using prognostic methods. We conduct perfect model predictability experiments starting from ten different initial states, and show potentially predictable variations of Antarctic bottom water formation (AABW) rates on time scales as long as twenty years. The associated Weddell Sea (WS) subsurface temperatures and Antarctic sea ice have comparable potential predictability as the AABW cell. The predictability of sea surface temperature (SST) variations over the WS and the SO is somewhat smaller, with predictable scales out to a decade. This reduced predictability is likely associated with stronger damping from air-sea interaction. As a complement to our perfect predictability study, we also make hindcasts of SO decadal variability using the GFDL CM2.1 decadal prediction system. Significant predictive skill for SO SST on multi-year time scales is found in the hindcast system. The success of the hindcasts, especially in reproducing observed surface cooling trends, is largely due to initializing the state of the AABW cell. A weak state of the AABW cell leads to cooler surface conditions and more extensive sea ice. Although there are considerable uncertainties regarding the observational data used to initialize the hindcasts, the consistency between the perfect model experiments and the decadal hindcasts at least gives us some indication as to where and to what extent skillful decadal SO forecasts might be possible.
This study investigates the roles of radiative forcing, sea surface temperatures (SSTs), and atmospheric and land initial conditions in the summer warming episodes of the United States. The summer warming episodes are defined as the significantly above normal (1983-2012) June-August 2-m temperature anomalies, and are referred to as heat waves in this study. Two contrasting cases, the summers of 2006 and 2012, are explored in detail to illustrate the distinct roles of SSTs, direct radiative forcing, and atmospheric and land initial conditions in driving U.S. summer heat waves. For 2012, simulations with the GFDL atmospheric general circulation model reveal that SSTs play a critical role. Further sensitivity experiments reveal the contributions of uniform global SST warming, SSTs in individual ocean basins and direct radiative forcing to the geographic distribution and magnitudes of warm temperature anomalies. In contrast, for 2006, the atmospheric and land initial conditions are key drivers. The atmospheric (land) initial conditions play a major (minor) role in the central and northwestern (eastern) U.S.. Due to changes in radiative forcing, the probability of areal-averaged summer temperature anomalies over U.S. exceeding the observed 2012 anomaly increases with time over the early 21st century. La Niña (El Niño) events tend to increase (reduce) the occurrence rate of heat waves. The temperatures over the central U.S. are mostly influenced by El Niño/La Niña, with the central tropical Pacific playing a more important role than the eastern tropical Pacific. Thus, atmospheric and land initial conditions, SSTs and radiative forcing are all important drivers of, and sources of predictability for U.S. summer heat waves.
Retrospective seasonal forecasts of North Atlantic tropical cyclone (TC) activity over the period 1980–2014 are conducted using a GFDL high-resolution coupled climate model [Forecast-Oriented Low Ocean Resolution (FLOR)]. The focus is on basin-total TC and U.S. landfall frequency. The correlations between observed and model predicted basin-total TC counts range from 0.4 to 0.6 depending on the month of the initial forecast. The correlation values for U.S. landfalling activity based on individual TCs tracked from the model are smaller and between 0.1 and 0.4. Given the limited skill from the model, statistical methods are used to complement the dynamical seasonal TC prediction from the FLOR model. Observed and predicted TC tracks were classified into four groups using fuzzy c-mean clustering to evaluate the model’s predictability in observed classification of TC tracks. Analyses revealed that the FLOR model has the highest skill in predicting TC frequency for the cluster of TCs that tracks through the Caribbean and the Gulf of Mexico.
New hybrid models are developed to improve the prediction of observed basin-total TC and landfall TC frequencies. These models use large-scale climate predictors from the FLOR model as predictors for generalized linear models. The hybrid models show considerable improvements in the skill in predicting the basin-total TC frequencies relative to the dynamical model. The new hybrid model shows correlation coefficients as high as 0.75 for basinwide TC counts from the first two lead months and retains values around 0.50 even at the 6-month lead forecast. The hybrid model also shows comparable or higher skill in forecasting U.S. landfalling TCs relative to the dynamical predictions. The correlation coefficient is about 0.5 for the 2–5-month lead times.
Skillful seasonal forecasting of tropical cyclone (TC; wind speed ≥17.5 m s−1) activity is challenging, even more so when the focus is on major hurricanes (wind speed ≥49.4 m s−1), the most intense hurricanes (Category 4–5; wind speed ≥58.1 m s−1), and landfalling TCs. Here we show that a 25-km resolution global coupled model (HiFLOR) developed at the Geophysical Fluid Dynamics Laboratory (GFDL) has improved skill in predicting the frequencies of major hurricanes and Category 4–5 hurricanes in the North Atlantic, and landfalling TCs over the United States and Caribbean Islands a few months in advance, relative to its 50-km resolution predecessor climate model (FLOR). HiFLOR also shows significant skill in predicting Category 4–5 hurricanes in the western North Pacific and eastern North Pacific, while both models show comparable skills in predicting basin-total and landfalling TC frequency in the basins. The improved skillful forecasts of basin-total TCs, major hurricanes, and Category 4–5 hurricane activity in the North Atlantic by HiFLOR are obtained mainly by improved representation of the TCs and their response to climate from the increased horizontal resolution, rather than improvements in large-scale parameters.
This study aims to assess whether, and the extent to which, an increase in atmospheric resolution in versions of the Geophysical Fluid Dynamics Laboratory (GFDL) High-Resolution Forecast-oriented Low Ocean Resolution Version of CM2.5 (FLOR) with 50 km and HiFLOR with 25 km improves the simulation of the El Niño Southern Oscillation-tropical cyclone (ENSO-TC) connections in the western North Pacific (WNP). HiFLOR simulates better ENSO-TC connections in the WNP including TC track density, genesis and landfall than FLOR in both long-term control experiments and sea surface temperature (SST)- and sea surface salinity (SSS)-restoring historical runs (1971-2012). Restoring experiments are performed with SSS and SST restored to observational estimates of climatological SSS and interannually-varying monthly SST. In the control experiments of HiFLOR, an improved simulation of the Walker circulation arising from more realistic SST and precipitation is largely responsible for its better performance in simulating ENSO-TC connections in the WNP. In the SST-restoring experiments of HiFLOR, more realistic Walker circulation and steering flow during El Niño/La Niña are responsible for the improved simulation of ENSO-TC connections in the WNP. The improved simulation of ENSO-TC connections with HiFLOR arises from a better representation of SST and better responses of environmental large-scale circulation to SST anomalies associated with El Niño/La Niña. A better representation of ENSO-TC connections in HiFLOR can benefit the seasonal forecasting of TC genesis, track and landfall, improve our understanding of the interannual variation of TC activity, and provide better projection of TC activity under climate change.
Zhang, Wei, Gabriele Villarini, Gabriel A Vecchi, Hiroyuki Murakami, and Richard G Gudgel, June 2016: Statistical-dynamical seasonal forecast of western North Pacific and East Asia landfalling tropical cyclones using the high-resolution GFDL FLOR coupled model. Journal of Advances in Modeling Earth Systems, 8(2), DOI:10.1002/2015MS000607. Abstract
This study examines the seasonal prediction of western North Pacific [WNP) and East Asia landfalling tropical cyclones (TCs) using the Geophysical Fluid Dynamics Laboratory(GFDL) Forecast-oriented Low Ocean Resolution version of CM2.5 with Flux Adjustment (FLOR-FA) and finite-mixture-model (FMM)-based statistical cluster analysis. Using the FMM-based cluster analysis, seven clusters are identified from the historical and FLOR-FA-predicted TC tracks for the period 1980–2013. FLOR-FA has significant skill in predicting year-to-year variations in the frequency of TCs within clusters 1 (recurving TCs) and 5 (straight-moving TCs). By building Poisson regression models for each cluster using key predictors (i.e., sea surface temperature, 500 hPa geopotential height, and zonal vertical wind shear), the predictive skill for almost all the clusters at all initialization months improves with respect to the dynamic prediction. The prediction of total WNP TC frequency made by combining hybrid predictions for each of the seven clusters in the hybrid model shows skill higher than what achieved using the TC frequency directly from FLOR-FA initialized from March to July. However, the hybrid predictions for total WNP TC frequency initialized from January to February exhibit lower skill than FLOR-FA. The prediction of TC landfall over East Asia made by combining the hybrid models of TC frequency in each cluster and its landfall rate over East Asia also outperforms FLOR-FA for all initialization months January through July.
This study aims to assess the connections between the El Niño Southern Oscillation (ENSO) and tropical cyclones near Guam (GuamTC) using the state-of-the-art Geophysical Fluid Dynamics Laboratory (GFDL) Forecast-oriented Low Ocean Resolution Version of CM2.5 (FLOR). In observations, more (less) GuamTCs occur in El Niño (La Niña) years and the ENSO-GuamTC connections arise from TC genesis locations in ENSO phases. The observed ENSO-GuamTC connections are realistically simulated in the two control experiments that use two versions of FLOR, the standard version and another with flux adjustments (FLOR-FA). The ENSO-GuamTC connections in FLOR-FA are closer to observations than those in FLOR because of a better representation of TC genesis during ENSO phases. The physical mechanisms underlying the observed ENSO-GuamTC connections are further supported in the long-term control experiments with FLOR/FLOR-FA. The ENSO-GuamTC connections in sea surface temperature (SST)- and sea surface salinity (SSS)-restoring experiments with FLOR 1990 strongly resemble the observations, suggesting the ENSO-GuamTC connections arise substantially from the forcing of SST. The prediction skill of FLOR-FA for GuamTC frequency is quite promising in terms of correlation and root mean square error and is higher than that of FLOR for the period 1980-2014. This study shows the capability of global climate models (FLOR/FLOR-FA) in simulating the linkage between ENSO and TC activity near a highly localized region (i.e., Guam) and in predicting the frequency of TCs at the sub-basin scale.
This study demonstrates skillful seasonal prediction of 2m air temperature and precipitation over land in a new high-resolution climate model developed by Geophysical Fluid Dynamics Laboratory, and explores the possible sources of the skill. We employ a statistical optimization approach to identify the most predictable components of seasonal mean temperature and precipitation over land, and demonstrate the predictive skill of these components. First, we show improved skill of the high-resolution model over the previous lower-resolution model in seasonal prediction of NINO3.4 index and other aspects of interest. Then we measure the skill of temperature and precipitation in the high-resolution model for boreal winter and summer, and diagnose the sources of the skill. Lastly, we reconstruct predictions using a few most predictable components to yield more skillful predictions than the raw model predictions. Over three decades of hindcasts, we find that the two most predictable components of temperature are characterized by a component that is likely due to changes in external radiative forcing in boreal winter and summer, and an ENSO-related pattern in boreal winter. The most predictable components of precipitation in both seasons are very likely ENSO-related. These components of temperature and precipitation can be predicted with significant correlation skill at least 9 months in advance. The reconstructed predictions using only the first few predictable components from the model show considerably better skill relative to observations than raw model predictions. This study shows that the use of refined statistical analysis and a high-resolution dynamical model leads to significant skill in seasonal predictions of 2m air temperature and precipitation over land.
A new high-resolution Geophysical Fluid Dynamics Laboratory (GFDL) coupled model (HiFLOR) has been developed and used to investigate potential skill in simulation and prediction of tropical cyclone (TC) activity. HiFLOR comprises of high-resolution (~25-km mesh) atmosphere and land components and a more moderate-resolution (~100-km mesh) sea ice and ocean components. HiFLOR was developed from the Forecast Oriented Low Resolution Ocean model (FLOR) by decreasing the horizontal grid spacing of the atmospheric component from 50-km to 25-km, while leaving most of the sub-gridscale physical parameterizations unchanged. Compared with FLOR, HiFLOR yields a more realistic simulation of the structure, global distribution, and seasonal and interannual variations of TCs, and a comparable simulation of storm-induced cold wakes and TC-genesis modulation induced by the Madden Julian Oscillation (MJO). Moreover, HiFLOR is able to simulate and predict extremely intense TCs (categories 4 and 5) and their interannual variations, which represents the first time a global coupled model has been able to simulate such extremely intense TCs in a multi-century simulation, sea surface temperature restoring simulations, and retrospective seasonal predictions.
Sea surface temperature (SST) anomalies are often both leading indicators and important drivers of marine resource fluctuations. Assessment of the skill of SST anomaly forecasts within coastal ecosystems accounting for the majority of global fish yields, however, has been minimal. This reflects coarse global forecast system resolution and past emphasis on the predictability of ocean basin-scale SST variations. This paper assesses monthly to inter-annual SST anomaly predictions in coastal “Large Marine Ecosystems” (LMEs). We begin with an analysis of 7 well-observed LMEs adjacent to the United States and then examine how mechanisms responsible for prediction skill in these systems are reflected in predictions for LMEs globally. Historical SST anomaly estimates from the 1/4o daily Optimal Interpolation Sea Surface Temperature reanalysis (OISST.v2) were first found to be highly consistent with in-situ measurements for 6 of the 7 U.S. LMEs. Thirty years of retrospective forecasts from climate forecast systems developed at NOAA’s Geophysical Fluid Dynamics Laboratory (CM2.5-FLOR) and the National Center for Environmental Prediction (CFSv2) were then assessed against OISST.v2. Forecast skill varied widely by LME, initialization month, and lead but there were many cases of high skill that also exceeded that of a persistence forecast, some at leads greater than 6 months. Mechanisms underlying skill above persistence included accurate simulation of a) seasonal transitions between less predictable locally generated and more predictable basin-scale SST variability; b) seasonal transitions between different basin-scale influences; c) propagation of SST anomalies across seasons through sea ice; and d) re-emergence of previous anomalies upon the breakdown of summer stratification. Globally, significant skill above persistence across many tropical systems arises via mechanisms a) and b). Combinations of all four mechanisms contribute to less prevalent but nonetheless significant skill in extratropical systems. While continued refinement of global climate forecast systems and observations are needed to improve coastal SST anomaly prediction and extend predictions to other ecosystem relevant variables (e.g., salinity), present skill warrants close examination of forecasts for marine resource applications.
The seasonal predictability of extratropical storm tracks in Geophysical Fluid Dynamics Laboratory (GFDL)’s high-resolution climate model has been investigated using an average predictability time analysis. The leading predictable components of extratropical storm tracks are ENSO-related spatial pattern for both boreal winter and summer, and the second predictable components are mostly due to changes in external radiative forcing and multidecadal oceanic variability. These two predictable components for both seasons show significant correlation skill for all leads from 0 to 9 months, while the skill of predicting the boreal winter storm track is consistently higher than that of the austral winter. The predictable components of extratropical storm tracks are dynamically consistent with the predictable components of the upper troposphere jet flow for both seasons. Over the region with strong storm track signals in North America, the model is able to predict the changes in statistics of extremes connected to storm track changes (e.g., extreme low and high sea level pressure and extreme 2m air temperature) in response to different ENSO phases. These results point towards the possibility of providing skillful seasonal predictions of the statistics of extratropical extremes over land using high-resolution coupled models.
Decadal prediction experiments were conducted as part of CMIP5 using the GFDL-CM2.1 forecast system. The abrupt warming of the North Atlantic subpolar gyre (SPG) that was observed in the mid 1990s is considered as a case study to evaluate our forecast capabilities and better understand the reasons for the observed changes. Initializing the CM2.1 coupled system produces high skill in retrospectively predicting the mid-90s shift, which is not captured by the uninitialized forecasts. All the hindcasts initialized in the early 90s show a warming of the SPG, however, only the ensemble mean hindcasts initialized in 1995 and 1996 are able to reproduce the observed abrupt warming and the associated decrease and contraction of the SPG. Examination of the physical mechanisms responsible for the successful retrospective predictions indicates that initializing the ocean is key to predict the mid 90s warming. The successful initialized forecasts show an increased Atlantic Meridional Overturning Circulation and North Atlantic current transport, which drive an increased advection of warm saline subtropical waters northward, leading to a westward shift of the subpolar front and subsequently a warming and spin down of the SPG. Significant seasonal climate impacts are predicted as the SPG warms, including a reduced sea-ice concentration over the Arctic, an enhanced warming over central US during summer and fall, and a northward shift of the mean ITCZ. These climate anomalies are similar to those observed during a warm phase of the Atlantic Multidecadal Oscillation, which is encouraging for future predictions of North Atlantic climate.
We present seasonal predictions of Arctic sea ice extent (SIE) over the 1982-2013 period using two suites of retrospective forecasts initialized from a fully coupled ocean-atmosphere-sea ice assimilation system. High skill scores are found in predicting year-to-year fluctuations of Arctic SIE, with significant correlations up to 7 month ahead for September detrended anomalies. Predictions over the recent era, which coincides with an improved observational coverage, outperform the earlier period for most target months. We find, however, a degradation of skill in September during the last decade, a period of sea ice thinning in observations. The two prediction models, CM2.1 and FLOR, share very similar ocean and ice component and initialization but differ by their atmospheric component. FLOR has improved climatological atmospheric circulation and sea ice mean state but its skill is overall similar to CM2.1 for most seasons, which suggests a key role for initial conditions in predicting seasonal SIE fluctuations.
In our original paper (Vecchi et al., 2013, hereafter V13) we stated “the skill in the initialized forecasts comes in large part from the persistence of the mid-1990s shift by the initialized forecasts, rather than from predicting its evolution”. Smith et al (2013, hereafter S13) challenge that assertion, contending that DePreSys was able to make a successful retrospective forecast of that shift. We stand by our original assertion, and present additional analyses using output from DePreSys retrospective forecasts to support our assessment.
Tropical cyclones (TCs) are a hazard to life and property and a prominent element of the global climate system, therefore understanding and predicting TC location, intensity and frequency is of both societal and scientific significance. Methodologies exist to predict basin-wide, seasonally-aggregated TC activity months, seasons and even years in advance. We show that a newly developed high-resolution global climate model can produce skillful forecasts of seasonal TC activity on spatial scales finer than basin-wide, from months and seasons in advance of the TC season. The climate model used here is targeted at predicting regional climate and the statistics of weather extremes on seasonal to decadal timescales, and is comprised of high-resolution (50km×50km) atmosphere and land components, and more moderate resolution (~100km) sea ice and ocean components. The simulation of TC climatology and interannual variations in this climate model is substantially improved by correcting systematic ocean biases through “flux-adjustment.” We perform a suite of 12-month duration retrospective forecasts over the 1981-2012 period, after initializing the climate model to observationally-constrained conditions at the start of each forecast period – using both the standard and flux-adjusted versions of the model. The standard and flux-adjusted forecasts exhibit equivalent skill at predicting Northern Hemisphere TC season sea surface temperature, but the flux-adjusted model exhibits substantially improved basin-wide and regional TC activity forecasts, highlighting the role of systematic biases in limiting the quality of TC forecasts. These results suggest that dynamical forecasts of seasonally-aggregated regional TC activity months in advance are feasible.
Retrospective predictions of multi-year North Atlantic hurricane frequency are explored, by applying a hybrid statistical-dynamical forecast system to initialized and non-initialized multi-year forecasts of tropical Atlantic and tropical mean sea surface temperatures (SSTs) from two global climate model forecast systems. By accounting for impacts of initialization and radiative forcing, retrospective predictions of five-year mean and nine-year mean tropical Atlantic hurricane frequency show significant correlation relative to a null hypothesis of zero correlation. The retrospective correlations are increased in a two-model average forecast and by using a lagged-ensemble approach, with the two-model ensemble decadal forecasts hurricane frequency over 1961-2011 yielding correlation coefficients that approach 0.9.
These encouraging retrospective multi-year hurricane predictions, however, should be interpreted with care: although initialized forecasts have higher nominal skill than uninitialized ones, the relatively short record and large autocorrelation of the time series limits our confidence in distinguishing between the skill due to external forcing and that added by initialization. The nominal increase in correlation in the initialized forecasts relative to the uninitialized experiments is due to improved representation of the multi-year tropical Atlantic SST anomalies. The skill in the initialized forecasts comes in large part from the persistence of a mid-1990s shift by the initialized forecasts, rather than from predicting its evolution. Predicting shifts like that observed in 1994-1995 remains a critical issue for the success of multi-year forecasts of Atlantic hurricane frequency. The retrospective forecasts highlight the possibility that changes in observing system impact forecast performance.
The decadal predictability of sea surface temperature (SST) and 2m air temperature (T2m) in Geophysical Fluid Dynamics Laboratory (GFDL)'s decadal hindcasts, which are part of the Fifth Coupled Model Intercomparison Project experiments, has been investigated using an average predictability time (APT) analysis. Comparison of retrospective forecasts initialized using the GFDL's Ensemble Coupled Data Assimilation system with uninitialized historical forcing simulations using the same model, allows identification of internal multidecadal pattern (IMP) for SST and T2m. The IMP of SST is characterized by an inter-hemisphere dipole, with warm anomalies centered in the North Atlantic subpolar gyre region and North Pacific subpolar gyre region, and cold anomalies centered in the Antarctic Circumpolar Current region. The IMP of T2m is characterized by a general bi-polar seesaw, with warm anomalies centered in Greenland, and cold anomalies centered in Antarctica. The retrospective prediction skill of the initialized system, verified against independent observations, indicates that the IMP of SST may be predictable up to 4 (10) year lead time at 95% (90%) significance level, and the IMP of T2m may be predictable up to 2 (10) years at 95% (90%) significance level. The initialization of multidecadal variations of northward oceanic heat transport in the North Atlantic significantly improves the predictive skill of the IMP. The dominant roles of oceanic internal dynamics in decadal prediction are further elucidated by fixed-forcing experiments, in which radiative forcing is returned to 1961 values. These results point towards the possibility of meaningful decadal climate outlooks using dynamical coupled models, if they are appropriately initialized from a sustained climate observing system.
Skillfully predicting North Atlantic hurricane activity months in advance is of potential
societal significance and a useful test of our understanding of the factors controlling
hurricane activity. We describe a statistical-dynamical hurricane forecasting system,
based on a statistical hurricane model, with explicit uncertainty estimates, built from a
suite of high-resolution global atmospheric dynamical model integrations spanning a
broad range of climate states. The statistical model uses two climate predictors: the sea
surface temperature (SST) in the tropical North Atlantic and SST averaged over the
global tropics. The choice of predictors is motivated by physical considerations, results of
high-resolution hurricane modeling and of statistical modeling of the observed record.
The statistical hurricane model is applied to a suite of initialized dynamical global
climate model forecasts of SST to predict North Atlantic hurricane frequency, which
peaks in the August-October season, from different starting dates. Retrospective forecasts
of the 1982-2009 period indicate that skillful predictions can be made from as early as
November of the previous year – that is, skillful forecasts for the coming North Atlantic
hurricane season could be made as the current one is closing. Based on forecasts
initialized between November 2009 and March 2010, the model system predicts that the
upcoming 2010 North Atlantic hurricane season will likely be more active than the 1982-
2009 climatology, with the forecasts initialized in March 2010 predicting an expected
hurricane count of eight and a 50% probability of counts between six (the 1966-2009
median) and nine.
The formulation and simulation characteristics of two new global coupled climate models developed at NOAA's Geophysical Fluid Dynamics Laboratory (GFDL) are described. The models were designed to simulate atmospheric and oceanic climate and variability from the diurnal time scale through multicentury climate change, given our computational constraints. In particular, an important goal was to use the same model for both experimental seasonal to interannual forecasting and the study of multicentury global climate change, and this goal has been achieved.
Two versions of the coupled model are described, called CM2.0 and CM2.1. The versions differ primarily in the dynamical core used in the atmospheric component, along with the cloud tuning and some details of the land and ocean components. For both coupled models, the resolution of the land and atmospheric components is 2° latitude × 2.5° longitude; the atmospheric model has 24 vertical levels. The ocean resolution is 1° in latitude and longitude, with meridional resolution equatorward of 30° becoming progressively finer, such that the meridional resolution is 1/3° at the equator. There are 50 vertical levels in the ocean, with 22 evenly spaced levels within the top 220 m. The ocean component has poles over North America and Eurasia to avoid polar filtering. Neither coupled model employs flux adjustments.
The control simulations have stable, realistic climates when integrated over multiple centuries. Both models have simulations of ENSO that are substantially improved relative to previous GFDL coupled models. The CM2.0 model has been further evaluated as an ENSO forecast model and has good skill (CM2.1 has not been evaluated as an ENSO forecast model). Generally reduced temperature and salinity biases exist in CM2.1 relative to CM2.0. These reductions are associated with 1) improved simulations of surface wind stress in CM2.1 and associated changes in oceanic gyre circulations; 2) changes in cloud tuning and the land model, both of which act to increase the net surface shortwave radiation in CM2.1, thereby reducing an overall cold bias present in CM2.0; and 3) a reduction of ocean lateral viscosity in the extratropics in CM2.1, which reduces sea ice biases in the North Atlantic.
Both models have been used to conduct a suite of climate change simulations for the 2007 Intergovernmental Panel on Climate Change (IPCC) assessment report and are able to simulate the main features of the observed warming of the twentieth century. The climate sensitivities of the CM2.0 and CM2.1 models are 2.9 and 3.4 K, respectively. These sensitivities are defined by coupling the atmospheric components of CM2.0 and CM2.1 to a slab ocean model and allowing the model to come into equilibrium with a doubling of atmospheric CO2. The output from a suite of integrations conducted with these models is freely available online (see http://nomads.gfdl.noaa.gov/).
Manuscript received 8 December 2004, in final form 18 March 2005
The climate response to idealized changes in the atmospheric CO2 concentration by the new GFDL climate model (CM2) is documented. This new model is very different from earlier GFDL models in its parameterizations of subgrid-scale physical processes, numerical algorithms, and resolution. The model was constructed to be useful for both seasonal-to-interannual predictions and climate change research. Unlike previous versions of the global coupled GFDL climate models, CM2 does not use flux adjustments to maintain a stable control climate. Results from two model versions, Climate Model versions 2.0 (CM2.0) and 2.1 (CM2.1), are presented.
Two atmosphere–mixed layer ocean or slab models, Slab Model versions 2.0 (SM2.0) and 2.1 (SM2.1), are constructed corresponding to CM2.0 and CM2.1. Using the SM2 models to estimate the climate sensitivity, it is found that the equilibrium globally averaged surface air temperature increases 2.9 (SM2.0) and 3.4 K (SM2.1) for a doubling of the atmospheric CO2 concentration. When forced by a 1% per year CO2 increase, the surface air temperature difference around the time of CO2 doubling [transient climate response (TCR)] is about 1.6 K for both coupled model versions (CM2.0 and CM2.1). The simulated warming is near the median of the responses documented for the climate models used in the 2001 Intergovernmental Panel on Climate Change (IPCC) Working Group I Third Assessment Report (TAR).
The thermohaline circulation (THC) weakened in response to increasing atmospheric CO2. By the time of CO2 doubling, the weakening in CM2.1 is larger than that found in CM2.0: 7 and 4 Sv (1 Sv 106 m3 s−1), respectively. However, the THC in the control integration of CM2.1 is stronger than in CM2.0, so that the percentage change in the THC between the two versions is more similar. The average THC change for the models presented in the TAR is about 3 or 4 Sv; however, the range across the model results is very large, varying from a slight increase (+2 Sv) to a large decrease (−10 Sv).
Webb, M J., Catherine A Senior, D M H Sexton, W J Ingram, K D Williams, M A Ringer, B McAveney, R Colman, Brian J Soden, Richard G Gudgel, Thomas R Knutson, S Emori, T Ogura, Y Tsushima, N Andronova, B Li, I Musat, Sandrine Bony, and K E Taylor, 2006: On the contribution of local feedback mechanisms to the range of climate sensitivity in two GCM ensembles. Climate Dynamics, 27(1), DOI:10.1007/s00382-006-0111-2. Abstract
Global and local feedback analysis techniques have been applied to two ensembles of mixed layer equilibrium CO2 doubling climate change experiments, from the CFMIP (Cloud Feedback Model Intercomparison Project) and QUMP (Quantifying Uncertainty in Model Predictions) projects. Neither of these new ensembles shows evidence of a statistically significant change in the ensemble mean or variance in global mean climate sensitivity when compared with the results from the mixed layer models quoted in the Third Assessment Report of the IPCC. Global mean feedback analysis of these two ensembles confirms the large contribution made by inter-model differences in cloud feedbacks to those in climate sensitivity in earlier studies; net cloud feedbacks are responsible for 66% of the inter-model variance in the total feedback in the CFMIP ensemble and 85% in the QUMP ensemble. The ensemble mean global feedback components are all statistically indistinguishable between the two ensembles, except for the clear-sky shortwave feedback which is stronger in the CFMIP ensemble. While ensemble variances of the shortwave cloud feedback and both clear-sky feedback terms are larger in CFMIP, there is considerable overlap in the cloud feedback ranges; QUMP spans 80% or more of the CFMIP ranges in longwave and shortwave cloud feedback. We introduce a local cloud feedback classification system which distinguishes different types of cloud feedbacks on the basis of the relative strengths of their longwave and shortwave components, and interpret these in terms of responses of different cloud types diagnosed by the International Satellite Cloud Climatology Project simulator. In the CFMIP ensemble, areas where low-top cloud changes constitute the largest cloud response are responsible for 59% of the contribution from cloud feedback to the variance in the total feedback. A similar figure is found for the QUMP ensemble. Areas of positive low cloud feedback (associated with reductions in low level cloud amount) contribute most to this figure in the CFMIP ensemble, while areas of negative cloud feedback (associated with increases in low level cloud amount and optical thickness) contribute most in QUMP. Classes associated with high-top cloud feedbacks are responsible for 33 and 20% of the cloud feedback contribution in CFMIP and QUMP, respectively, while classes where no particular cloud type stands out are responsible for 8 and 21%.
Williams, K D., M A Ringer, Catherine A Senior, M J Webb, B McAveney, N Andronova, Sandrine Bony, J-L Dufresne, S Emori, Richard G Gudgel, Thomas R Knutson, B Li, K Lo, I Musat, J Wegner, A Slingo, and J F B Mitchell, 2006: Evaluation of a component of the cloud response to climate change in an intercomparison of climate models. Climate Dynamics, 26(2-3), DOI:10.1007/s00382-005-0067-7. Abstract
Most of the uncertainty in the climate sensitivity of contemporary general circulation models (GCMs) is believed to be connected with differences in the simulated radiative feedback from clouds. Traditional methods of evaluating clouds in GCMs compare time–mean geographical cloud fields or aspects of present-day cloud variability, with observational data. In both cases a hypothetical assumption is made that the quantity evaluated is relevant for the mean climate change response. Nine GCMs (atmosphere models coupled to mixed-layer ocean models) from the CFMIP and CMIP model comparison projects are used in this study to demonstrate a common relationship between the mean cloud response to climate change and present-day variability. Although atmosphere–mixed-layer ocean models are used here, the results are found to be equally applicable to transient coupled model simulations. When changes in cloud radiative forcing (CRF) are composited by changes in vertical velocity and saturated lower tropospheric stability, a component of the local mean climate change response can be related to present-day variability in all of the GCMs. This suggests that the relationship is not model specific and might be relevant in the real world. In this case, evaluation within the proposed compositing framework is a direct evaluation of a component of the cloud response to climate change. None of the models studied are found to be clearly superior or deficient when evaluated, but a couple appear to perform well on several relevant metrics. Whilst some broad similarities can be identified between the 60°N–60°S mean change in CRF to increased CO2 and that predicted from present-day variability, the two cannot be quantitatively constrained based on changes in vertical velocity and stability alone. Hence other processes also contribute to the global mean cloud response to climate change.
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.
Gudgel, Richard G., Anthony Rosati, and C Tony Gordon, 2001: The sensitivity of a coupled atmospheric-oceanic GCM in prescribed low-level clouds over the ocean and tropical landmasses. Monthly Weather Review, 129(8), 2103-2115. Abstract PDF
The sensitivity of a coupled general circulation model (CGCM) to tropical marine stratocumulus (MSc) clouds and low-level clouds over the tropical land is examined. The hypothesis that low-level clouds play an important role in determining the strength and position of the Walker circulation and also on the strength and phase of the El Niño–Southern Oscillation (ENSO) is studied using a Geophysical Fluid Dynamics Laboratory (GFDL) experimental prediction CGCM. In the Tropics, a GFDL experimental prediction CGCM exhibits a strong bias in the western Pacific where an eastward shift in the ascending branch of the Walker circulation diminishes the strength and expanse of the sea surface temperature (SST) warm pool, thereby reducing the east–west SST gradient, and effectively weakening the trade winds. These model features are evidence of a poorly simulated Walker circulation, one that mirrors a “perpetual El Niño” state. One possible factor contributing to this bias is a poor simulation of MSc clouds in the eastern equatorial Pacific (which are essential to a proper SST annual cycle). Another possible contributing factor might be radiative heating biases over the land in the Tropics, which could, in turn, have a significant impact on the preferred locations of maximum convection in the Tropics. As a means of studying the sensitivity of a CGCM to both MSc clouds and to varied radiative forcing over the land in the Tropics, low-level clouds obtained from the International Satellite Cloud Climatology Project (ISCCP) are prescribed. The experiment sets consist of one where clouds are fully predicted, another where ISCCP low-level clouds are prescribed over the oceans alone, and a third where ISCCP low-level clouds are prescribed both over the global oceans and over the tropical landmasses. A set of ten 12-month hindcasts is performed for each experiment.
The results show that the combined prescription of interannually varying global ocean and climatological tropical land low-level clouds into the CGCM results in a much improved simulation of the Walker circulation over the Pacific Ocean. The improvement to the tropical circulation was also notable over the Indian and Atlantic basins as well. These improvements in circulation led to a considerable increase in ENSO hindcast skill in the first year by the CGCM. These enhancements were a function of both the presence of MSc clouds over the tropical oceans and were also due to the more realistic positioning of the regions of maximum convection in the Tropics. This latter model feature was essentially a response to the change in radiative forcing over tropical landmasses associated with a reduction in low cloud fraction and optical depth when ISCCP low-level clouds were prescribed there. These results not only underscore the importance of a reasonable representation of MSc clouds but also point out the considerable impact that radiative forcing over the tropical landmasses has on the simulated position of the Walker circulation and also on ENSO forecasting.
The seasonal cycle of SST observed in the eastern equatorial Pacific is poorly simulated by many ocean-atmosphere coupled GCMs. This deficiency may be partly due to an incorrect prediction of tropical marine stratocumulus (MSc). To explore this hypothesis, two basic multiyear simulations have been performed using a coupled GCM with seasonally varying solar radiation. The model's cloud prediction scheme, which under-predicts tropical marine stratocumulus, is used for all clouds in the control run. In contrast, in the "ISCCP" run, the climatological monthly mean low cloud fraction is specified over the open ocean, utilizing C2 data from the International Satellite Cloud Climatology Project (ISCCP). In this manner, the treatment of MSc clouds, including the annual cycle, is more realistic than in previous sensitivity studies.
Robust surface and subsurface thermodynamical and dynamical responses to the specified MSc are found in the Tropics, especially near the equator. In the annual mean, the equatorial cold tongue extends farther west and intensifies, while the east-west SST gradient is enhanced. A double SST maximum flanking the cold tongue becomes asymmetric about the equator. The SST annual cycle in the eastern equatorial Pacific strengthens, and the equatorial SST seasonal anomalies migrate farther westward. MSc-induced local shortwave radiative cooling enhances dynamical cooling associated with the southeast trades. The surface meridional wind stress in the extreme eastern equatorial Pacific remains southerly all year, while the surface zonal wind stress and equatorial upwelling intensify, as does the seasonal cycle of evaporation, in better agreement with observation. Within the ocean, the thermocline steepens and the Equatorial Undercurrent intensifies. When the low clouds are entirely removed, the SST warms by about 5.5 K in the western and central tropical Pacific, relative to "ISCCP," and the model's SST bias there reverses sign.
ENSO-like interannual variability with a characteristic timescale of 3-5 yr is found in all simulations, though its amplitude varies. The "ISCCP" equatorial cold tongue inhibits the eastward progression of ENSO-like warm events east of the date line. When the specified low cloud fraction in "ISCCP' is reduced by 20%, the interannual variability amplifies somewhat and the coupled model responds more like a delayed oscillator. The apparent sensitivity in the equatorial Pacific to a 20% relative change in low cloud fraction may have some cautionary implications for seasonal prediction by coupled GCMs.
Gordon, C T., Anthony Rosati, and Richard G Gudgel, 1999: Tropical interannual variability response of a coupled model to specified low clouds In Proceedings of the Twenty-Fourth Annual Climate Diagnostics and Prediction Workshop, Springfield, VA, NTIS, 331-334.
Gudgel, Richard G., Anthony Rosati, and C Tony Gordon, 1999: The impact of prescribed tropical land and ocean clouds on the Walker circulation in the GFDL coupled ocean-atmosphere GCM In Proceedings of the Twenty-Fourth Annual Climate Diagnostics and Prediction Workshop, Springfield, VA, NTIS, 307-310. Abstract
The role that tropical land and ocean clouds play in the GFDL coupled ocean-atmosphere GCM is studied through a series of 10 one-year model runs. In the tropics, a strong bias in the GFDL coupled GCM is evident in the western Pacific where excessive convection erodes the SST warm pool, reducing the SST pacific gradient, and effectively weakening the trade winds. This bias is exacerbated by the poor simulation of eastern equatorial Pacific marine stratus clouds which are essential to a proper seasonal cycle (annual as opposed to biannual) of the trade winds and the SST's. As a means to better understand the importance of ocean-only versus ocean and land tropical cloud prediction, low-level ISCCP clouds are used to study the effects on the GFDL atmosphere-only and coupled model simulation. The prescription of tropical low-level ocean and land clouds into the GCM resulted in a better simulation of the Walker circulation in both coupled and uncoupled modes. This climatological improvement to the Walker circulation corresponded with an improvement in the ability of the coupled GCM to simulate ENSO (El Niño/Southern Oscillation). The more reasonably represented land surface heating in the tropics led to more well-defined and positioned regions of convergence and divergence both at the surface and aloft. The GCM appears to be quite sensitive to the pronounced horizontal and vertical topographical structure in the Indonesian Archipelago and in tropical South America. This is most notable in the sensitivity of the model to small cloud fraction changes over these regions. This emphasizes the importance of reasonably representing the land surface heating in these regions. Whereas this sensitivity is evident in both the coupled and uncoupled simulations not only in terms of the model's climate but also in term of the model's ability to simulate ENSO, it underscores the importance of producing reasonable heating profiles over the land regions in the tropics.
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.
Gordon, C T., Anthony Rosati, and Richard G Gudgel, 1998: Tropical sensitivity to specified ISCCP low clouds in a coupled model In Proceedings of the Twenty-Second Annual Climate Diagnostics and Prediction Workshop, Springfield, VA, NTIS, 232-235.
Anderson, Jeffrey L., and Richard G Gudgel, 1997: Impact of atmospheric initial conditions on seasonal predictions with a coupled ocean-atmosphere model In Proceedings of the Twenty-First Annual Climate Diagnostics and Prediction Workshop, Springfield, VA, NTIS, 61-66.
Anderson, Jeffrey L., Anthony Rosati, and Richard G Gudgel, 1997: Potential predictability in an ensemble of coupled atmosphere-ocean general circulation model seasonal forecasts In Proceedings of the Twenty-First Annual Climate Diagnostics and Prediction Workshop, Springfield, VA, NTIS, 18-21.
A coupled atmosphere-ocean GCM (general circulation model) has been developed for climate predictions on seasonal to interannual timescales. The atmosphere model is a global spectral GCM T30L18 and the ocean model is global on a 1 degree grid. Initial conditions for the atmosphere were obtained from National Meteorological Center (now known as the National Centers for Environmental Prediction) analyses, while those for the ocean came from three ocean data assimilation (DA) systems. One system is a four-dimensional DA scheme that uses conventional SST observations and vertical temperature profiles inserted into the ocean model and is forced from winds from an operational analysis. The other two initialization schemes are based on the coupled model, both nudging the surface temperature toward observed SSTs and one nudging surface winds from an operational analysis. all three systems were run from 1979 to 1988, saving the state of the ocean every month, thus initial conditions may be obtained for any month during this period. The ocean heat content from the three systems was examined, and it was found that a strong lag correlation between Niño-3 SST anomalies and equatorial thermocline displacements exists. This suggests that, based on subsurface temperature field only, eastern tropical Pacific SST changes are possibly predictable at lead times of a year or more. It is this "memory" that is the physical basis for ENSO predictions.
Using the coupled GCM, 13-month forecasts were made for seven January and seven July cases, focusing on ENSO (El Niño-Southern Oscillation) prediction. The forecasts, whose ocean initial conditions contained subsurface thermal data, were successful in predicting the two El Niño and two La Niña events during the decade, whereas the forecasts that utilized ocean initial conditions from the coupled model that were nudged toward surface wind fields and SST only, failed to predict the events. Despite the coupled model's poor simulation of the annual cycle in the tropical Pacific, the ENSO forecasts from the full DA were remarkably good.
Gordon, C T., Anthony Rosati, and Richard G Gudgel, 1996: The impact of specified ISCCP low clouds in coupled model integrations In 11th Conference on Numerical Weather Prediction, Boston, MA, American Meteorological Society, 8-10.
Gordon, C T., Anthony Rosati, and Richard G Gudgel, 1996: The impact of specified ISCCP low clouds in coupled model integrations In Research Activities in Atmospheric and Oceanic Modelling, CAS/JSC Working Group on Numerical Experimentation, Report No. 23 WMO/TD No. 734, World Meteorological Organization, 9.12-9.13.
Harrison, Matthew J., Anthony Rosati, Richard G Gudgel, and Jeffrey L Anderson, 1996: Initialization of coupled model forecasts using an improved ocean data assimilation system In 11th Conference on Numerical Weather Prediction, Boston, MA, American Meteorological Society, 7.
Rosati, Anthony, Richard G Gudgel, and Kikuro Miyakoda, 1996: Global ocean data assimilation system In Modern Approaches to Data Assimilation in Ocean Modeling, The Netherlands, Elsevier Science Publishers, 181-203. Abstract
A global oceanic four-dimensional data assimilation system has been developed for use in initializing coupled ocean-atmosphere general circulation models and also to study interannual variability. The data inserted into a high resolution global ocean model consists only of conventional sea surface temperature observations and vertical temperature profiles. The data are inserted continuously into the model by updating the model's temperature solution every timestep. This update is created using a statistical interpolation routine applied to all data in a 30-day window for three consecutive timesteps and then the correction is held constant for nine timesteps. Not updating every timestep allows for a more computational efficient system without affecting the quality of the analysis.
The data assimilation system was run over a ten year period from 1979-1988. The resulting analysis product was compared with independent analysis including model derived fields like velocity. The large scale features seem consistent with other products based on observations. Using the mean of the ten-year period as a climatology, the data assimilation system was compared with the Levitus climatological atlas. Looking at the sea surface temperature and the seasonal cycle, as represented by the mixed layer depth, the agreement is quite good, however, some systematic differences do emerge.
Special attention is given to the tropical Pacific examining the El Niño signature. Two other assimilation schemes based on using Newtonian nudging of SST, are compared to the full data assimilation system. The heat content variability in the data assimilation seemed faithful to the observations. Overall, the results are encouraging, demonstrating that the data assimilation system seems to be able to capture many of the large scale general circulation features that are observed, both in a climatological sense and in the temporal variability.
Miyakoda, Kikuro, Joseph J Sirutis, Anthony Rosati, C Tony Gordon, Richard G Gudgel, William F Stern, Jeffrey L Anderson, and A Navarra, 1995: Atmospheric parameterizations in coupled air-sea models used for forecasts of ENSO In Proceedings of the International Scientific Conference on the Tropical Ocean Global Atmosphere (TOGA) Programme, WCRP-91, WMO/TD No. 717, Geneva, Switzerland, World Meteorological Organization, 802-806. Abstract
In order to investigate the feasibility of seasonal forecasts, a prediction system is developed. Here the main theme is the study of atmospheric physics parameterization for coupled air-sea modeling. The oceanic GCM uses 1 degree global grid with a finer resolution in the equatorial belt. The atmospheric GCM has the spectral T30 representation, which includes all of the usual physics parameterizations. Using a first version of the model (Coupled Model I) and a set of appropriate initial conditions, the capability of El Niño and La Niña forecasting with a 13 month lead time was tested, resulting in successful forecasts of the 1982/83 and 1988/89 events (Rosati et al., 1995b). However, longer runs of this system have revealed a sizable systematic error in simulations with a tendency to cool most of the world ocean, particularly the western tropical Pacific, and also without an adequate annual cycle of the SST in the eastern tropical Pacific.
In order to improve some of these features, particularly the ENSO phenomena, various versions of the atmospheric parameterizations and mountain representation are incorporated into the atmospheric GCM, and the model simulations are examined. The experiments are divided into two steps: one is with the uncoupled atmospheric model, and the other is with the coupled model. In the first step, five year simulations are carried out with the observed SST prescribed, and the results are compared with observations, which enables one to make the critical validation of the model. The second step is to couple the atmospheric and oceanic models, and integrate them from a January 1982 initial condition for 7 years, and also for another initial condition, i.e., January, 1988 for 13 months.
Compared with the boundary forced simulation, the coupling process introduces more degree of freedom, with increase of the sensitivity as well as the complexity considerably. In particular, the El Niño simulation is sensitive to any change of physics. For this reason, the objective of the simulation is focused only on the equatorial Pacific process and secondly the Indian monsoon, as opposed to the overall improvement of the general circulation. In other words, the approach is close to that of mechanistic modeling with specific targets rather than that of a GCM with broader objectives. The research is proceeding in two directions. One is: investigating the model's sensitivity for El Niño and La Niña processes to variation in a coupling parameter. The second is: after a number of trial-and-error experiments on various combinations of the parameterizations, the second atmospheric model, i.e., Model II, is selected. It is shown that Coupled Model II performs substantially better in some aspects but worse in other aspects than Coupled Model I. The improvement is found in the SST: warming occurs not only over the equatorial Pacific but also over the whole globe. The SST increase is achieved by the strong effect of the cumulus convection. On the other hand, some deficiencies remain the same in both models, i.e., the large positive errors of the SST in the eastern oceans, the lack of an annual cycle of the SST in the eastern equatorial Pacific, and the failure in forecast of the second El Niño. In summary, the prediction of the Southern Oscillation has been achieved by the two models for a full first cycle but not for the second cycle .
A global oceanic four-dimensional data assimilation system has been developed for use in initializing coupled ocean-atmosphere general circulation models and also to study interannual variability. The data inserted into a high-resolution global ocean model consist of conventional sea surface temperature observations and vertical temperature profiles. The data are inserted continuously into the model by updating the model's temperature solution every time step. This update is created using a statistical interpolation routine applied to all data in a 30-day window for three consecutive time steps and then the correction is held constant for nine time steps. Not updating every time step allows for a more computationally efficient system without affecting the quality of the analysis.
The data assimilation system was run over a 10-yr period from 1979 to 1988. The resulting analysis product was compared with independent analysis including model-derived fields like velocity. The large-scale features seem consistent with other products based on observations. Using the mean of the 10-yr period as a climatology, the data assimilation system was compared with the Levitus climatological atlas. Looking at the sea surface temperature and the seasonal cycle, as represented by the mixed-layer depth, the agreement is quite good, however, some systematic differences do emerge.
Special attention is given to the tropical Pacific examining the El Niño signature. Two other assimilation schemes based on the coupled model using Newtonian nudging of SST and then SST and surface winds are compared to the full data assimilation system. The heat content variability in the data assimilation seemed faithful to the observations. Overall, the results are encouraging, demonstrating that the data assimilation system seems to be able to capture many of the large-scale general circulation features that are observed, both in a climatological sense and in the temporal variability.
Miyakoda, Kikuro, Anthony Rosati, and Richard G Gudgel, 1994: Air-sea coupling experiments: ENSO forecasting. Part I In Proceedings of the 18th Annual Climate Diagnostics Workshop, U. S. Dept. of Commerce/NOAA/NWS, 153-156.
Rosati, Anthony, Kikuro Miyakoda, and Richard G Gudgel, 1994: Air-sea coupling experiments: ENSO forecasting. Part II In Proceedings of the 18th Annual Climate Diagnostics Workshop, U. S. Dept. of Commerce/NOAA/NWS, 358-361.