Abstract

The prediction skill of the Australian Bureau of Meteorology dynamical seasonal forecast model Predictive Ocean Atmosphere Model for Australia (POAMA) is assessed for probabilistic forecasts of spring season rainfall in Australia and the feasibility of increasing forecast skill through statistical postprocessing is examined. Two statistical postprocessing techniques are explored: calibrating POAMA prediction of rainfall anomaly against observations and using dynamically predicted mean sea level pressure to infer regional rainfall anomaly over Australia (referred to as “bridging”). A “homogeneous” multimodel ensemble prediction method (HMME) is also introduced that consists of the combination of POAMA’s direct prediction of rainfall anomaly together with the two statistically postprocessed predictions.

Using hindcasts for the period 1981–2006, the direct forecasts from POAMA exhibit skill relative to a climatological forecast over broad areas of eastern and southern Australia, where El Niño and the Indian Ocean dipole (whose behavior POAMA can skillfully predict at short lead times) are known to exert a strong influence in austral spring. The calibrated and bridged forecasts, while potentially offering improvement over the direct forecasts because of POAMA’s ability to predict the main drivers of springtime rainfall (e.g., El Niño and the Southern Oscillation), show only limited areas of improvement, mainly because strict cross-validation limits the ability to capitalize on relatively modest predictive signals with short record lengths. However, when POAMA and the two statistical–dynamical rainfall forecasts are combined in the HMME, higher deterministic and probabilistic skill is achieved over any of the single models, which suggests the HMME is another useful method to calibrate dynamical model forecasts.

Abstract

Seasonal forecasts are subject to various types of errors: amplification of errors in oceanic initial conditions, errors due to the unpredictable nature of the synoptic atmospheric variability, and coupled model error. Ensemble forecasting is usually used in an attempt to sample some or all of these various sources of error. How to build an ensemble forecasting system in the seasonal range remains a largely unexplored area. In this paper, various ensemble generation methodologies for the European Centre for Medium-Range Weather Forecasts (ECMWF) seasonal forecasting system are compared. A series of experiments using wind perturbations (applied when generating the oceanic initial conditions), sea surface temperature (SST) perturbations to those initial conditions, and random perturbation to the atmosphere during the forecast, individually and collectively, is presented and compared with the more usual lagged-average approach. SST perturbations are important during the first 2 months of the forecast to ensure a spread at least equal to the uncertainty level on the SST measure. From month 3 onward, all methods give a similar spread. This spread is significantly smaller than the rms error of the forecasts. There is also no clear link between the spread of the ensemble and the ensemble mean forecast error. These two facts suggest that factors not presently sampled in the ensemble, such as model error, act to limit the forecast skill. Methods that allow sampling of model error, such as multimodel ensembles, should be beneficial to seasonal forecasting.

Abstract

Tropical instability waves (TIWs) appear as monthly oscillations of the currents, sea level, and sea surface temperature of the eastern equatorial Pacific. They are understood as unstable waves feeding on the kinetic and potential energy of the mean currents. A general circulation model is shown to reproduce the main features associated with TIWs. It is then used to investigate the dynamical regime of TIWs, by assessing their sensitivity to oceanic initial conditions. Locally in space and time, small perturbations can grow enough to modify significantly the phase of the TIW field, suggesting some chaotic behavior. When considered over the whole active TIW region, however, the phases of the perturbed and unperturbed experiments remain in agreement. This suggests that TIW activity in this model is more consistent with a limit cycle behavior than with fully developed turbulence and that irregular behavior of TIWs mostly stems from external forcing by the wind. A stronger result is that TIWs in experiments starting from very different initial conditions come back into phase after a few years. This is consistent with the suggestion that TIWs might be phase-locked to the wind forcing. Quiescent periods of the TIW's cycle play an important role in the decay of TIW's phase disagreement but cannot explain the phase-locking mechanism entirely. The implications of these results and their sensitivity to the forcing are discussed.

Abstract

In this paper the structure and dynamics of the optimal perturbations of tropical low-frequency coupled ocean–atmosphere oscillations relevant to El Niño–Southern Oscillation (ENSO) are explored. These optimal perturbations yield information about potential precursors for ENSO events, and about the fundamental dynamical processes that may control perturbation growth and limit the predictability of interannual variability. The present study uses a hierarchy of hybrid coupled models. Each model is configured for the tropical Pacific Ocean and shares a common ocean general circulation model. Three different atmospheric models are used: a statistical model, a dynamical model, and a combination of a dynamical model and boundary layer model. Each coupled model possesses a coupled ocean–atmosphere eigenmode oscillation with a period of the order of several years. The properties of these various eigenmodes and their corresponding adjoint eigenmodes are explored.

The optimal perturbations of each coupled model for two different perturbation growth norms are also examined, and their behavior can be understood in terms of the properties of the aforementioned eigenmode oscillations. It is found that the optimal perturbation spectrum of each coupled model is primarily dominated by one member. The dominant optimal perturbation evolves into the most unstable eigenmode of the system. The structure of the optimal perturbations of each model is found to be controlled by the dynamics of the atmospheric model and air–sea interaction processes. For the coupled model with a statistical atmosphere, the optimal perturbation center of action is spread across the entire tropical Pacific in the form of a dipole. For the coupled models that include deep atmospheric convection, the optimal perturbation center of action is primarily confined to the western Pacific warm pool. In addition, the degree of nonnormality of the eigenmodes is controlled by the atmospheric model dynamics. These findings are in general agreement with the results obtained from intermediate coupled models. In particular, the atmospheric models used here have also been used in intermediate coupled models that have been employed extensively in previous studies of the optimal perturbations of El Niño–Southern Oscillation. Thus, a direct comparison of the optimal perturbation behavior of those intermediate models and the optimal perturbations of the hybrid models used here can be made.

Abstract

The optimal forcing patterns for El Niño–Southern Oscillation (ENSO) are examined for a hierarchy of hybrid coupled models using generalized stability theory. Specifically two cases are considered: one where the forcing is stochastic in time, and one where the forcing is time independent. The optimal forcing patterns in these two cases are described by the stochastic optimals and forcing singular vectors, respectively. The spectrum of stochastic optimals for each model was found to be dominated by a single pattern. In addition, the dominant stochastic optimal structure is remarkably similar to the forcing singular vector, and to the dominant singular vectors computed in a previous related study using a subset of the same models. This suggests that irrespective of whether the forcing is in the form of an impulse, is time invariant, or is stochastic in nature, the optimal excitation for the eigenmode that describes ENSO in each model is the same. The optimal forcing pattern, however, does vary from model to model, and depends on air–sea interaction processes.

Estimates of the stochastic component of forcing were obtained from atmospheric analyses and the projection of the dominant optimal forcing pattern from each model onto this component of the forcing was computed. It was found that each of the optimal forcing patterns identified may be present in nature and all are equally likely. The existence of a dominant optimal forcing pattern is explored in terms of the effective dimension of the coupled system using the method of balanced truncation, and was found to be O(1) for the models used here. The implications of this important result for ENSO prediction and predictability are discussed.

Abstract

This paper is an evaluation of the role of salinity in the framework of temperature data assimilation in a global ocean model that is used to initialize seasonal climate forecasts. It is shown that the univariate assimilation of temperature profiles, without attempting to correct salinity, can induce first-order errors in the subsurface temperature and salinity fields. A recently developed scheme by A. Troccoli and K. Haines is used to improve the salinity field. In this scheme, salinity increments are derived from the observed temperature, by using the model temperature and salinity profiles, assuming that the temperature–salinity relationship in the model profiles is preserved. In addition, the temperature and salinity fields are matched below the observed temperature profile by vertically displacing the original model profiles.

Two data assimilation experiments were performed for the 6-yr period 1993–98. These show that the salinity scheme is effective at maintaining the haline and thermal structures at and below thermocline level, especially in tropical regions, by avoiding spurious convection. In addition to improvements in the mean state, the scheme allows more temporal variability than simply controlling the salinity field by relaxation to climatological data. Some comparisons with sparse salinity observations are also made, which suggest that the subsurface salinity variability in the western Pacific is better reproduced in the experiment in which the salinity scheme is used. The salinity analyses might be improved further by use of altimeter sea level or sea surface salinity observations from satellite.

Abstract

The European Centre for Medium-Range Weather Forecasts (ECMWF) has made seasonal forecasts since 1997 with ensembles of a coupled ocean–atmosphere model, System-1 (S1). In January 2002, a new version, System-2 (S2), was introduced. For the calibration of these models, hindcasts have been performed starting in 1987, so that 15 yr of hindcasts and forecasts are now available for verification.

Seasonal predictability is to a large extent due to the El Niño–Southern Oscillation (ENSO) climate oscillations. ENSO predictions of the ECMWF models are compared with those of statistical models, some of which are used operationally. The relative skill depends strongly on the season. The dynamical models are better at forecasting the onset of El Niño or La Niña in boreal spring to summer. The statistical models are comparable at predicting the evolution of an event in boreal fall and winter.

Abstract

Since 1997, the European Centre for Medium-Range Weather Forecasts (ECMWF) has made seasonal forecasts with ensembles of a coupled ocean–atmosphere model, System-1 (S1). In January 2002, a new version, System-2 (S2), was introduced. For the calibration of these models, hindcasts have been performed starting in 1987, so that 15 yr of hindcasts and forecasts are now available for verification.

The main cause of seasonal predictability is El Niño and La Niña perturbing the average weather in many regions and seasons throughout the world. As a baseline to compare the dynamical models with, a set of simple statistical models (STAT) is constructed. These are based on persistence and a lagged regression with the first few EOFs of SST from 1901 to 1986 wherever the correlations are significant. The first EOF corresponds to ENSO, and the second corresponds to decadal ENSO. The temperature model uses one EOF, the sea level pressure (SLP) model uses five EOFs, and the precipitation model uses two EOFs but excludes persistence.

As the number of verification data points is very low (15), the simplest measure of skill is used: the correlation coefficient of the ensemble mean. To further reduce the sampling uncertainties, we restrict ourselves to areas and seasons of known ENSO teleconnections.

The dynamical ECMWF models show better skill in 2-m temperature forecasts over sea and the tropical land areas than STAT, but the modeled ENSO teleconnection pattern to North America is shifted relative to observations, leading to little pointwise skill. Precipitation forecasts of the ECMWF models are very good, better than those of the statistical model, in southeast Asia, the equatorial Pacific, and the Americas in December–February. In March–May the skill is lower. Overall, S1 (S2) shows better skill than STAT at lead time of 2 months in 29 (32) out of 40 regions and seasons of known ENSO teleconnections.

A comprehensive and cohesive aerosol measurement record with consistent, well-understood uncertainties is a prerequisite to understanding aerosol impacts on long-term climate and environmental variability. Objectives to attaining such an understanding include improving upon the current state-of-the-art sensor calibration and developing systematic validation methods for remotely sensed microphysical properties. While advances in active and passive remote sensors will lead to needed improvements in retrieval accuracies and capabilities, ongoing validation is essential so that the changing sensor characteristics do not mask atmospheric trends. Surface-based radiometer, chemical, and lidar networks have critical roles within an integrated observing system, yet they currently undersample key geographic regions, have limitations in certain measurement capabilities, and lack stable funding. In situ aircraft observations of size-resolved aerosol chemical composition are necessary to provide important linkages between active and passive remote sensing. A planned, systematic approach toward a global aerosol observing network, involving multiple sponsoring agencies and surface-based, suborbital, and spaceborne sensors, is required to prioritize trade-offs regarding capabilities and costs. This strategy is a key ingredient of the Progressive Aerosol Retrieval and Assimilation Global Observing Network (PARAGON) framework. A set of recommendations is presented.

Aerosols exert myriad influences on the earth's environment and climate, and on human health. The complexity of aerosol-related processes requires that information gathered to improve our understanding of climate change must originate from multiple sources, and that effective strategies for data integration need to be established. While a vast array of observed and modeled data are becoming available, the aerosol research community currently lacks the necessary tools and infrastructure to reap maximum scientific benefit from these data. Spatial and temporal sampling differences among a diverse set of sensors, nonuniform data qualities, aerosol mesoscale variabilities, and difficulties in separating cloud effects are some of the challenges that need to be addressed. Maximizing the longterm benefit from these data also requires maintaining consistently well-understood accuracies as measurement approaches evolve and improve. Achieving a comprehensive understanding of how aerosol physical, chemical, and radiative processes impact the earth system can be achieved only through a multidisciplinary, interagency, and international initiative capable of dealing with these issues. A systematic approach, capitalizing on modern measurement and modeling techniques, geospatial statistics methodologies, and high-performance information technologies, can provide the necessary machinery to support this objective. We outline a framework for integrating and interpreting observations and models, and establishing an accurate, consistent, and cohesive long-term record, following a strategy whereby information and tools of progressively greater sophistication are incorporated as problems of increasing complexity are tackled. This concept is named the Progressive Aerosol Retrieval and Assimilation Global Observing Network (PARAGON). To encompass the breadth of the effort required, we present a set of recommendations dealing with data interoperability; measurement and model integration; multisensor synergy; data summarization and mining; model evaluation; calibration and validation; augmentation of surface and in situ measurements; advances in passive and active remote sensing; and design of satellite missions. Without an initiative of this nature, the scientific and policy communities will continue to struggle with understanding the quantitative impact of complex aerosol processes on regional and global climate change and air quality.