Patterns and cycles in the Climate Forecast System Reanalysis wind and wave data
Highlights
► We analyze CFSR winds and hindcast waves to describe the global patterns. ► Quarterly averages and percentiles reveal the seasonal variability. ► The inter-annual variability reveals the areas related to cycles longer than one year. ► The dataset shows strong correlation with the AO, AAO, ENSO, and MJO. ► The correlation reveals the regions strongly influenced by the oscillations.
Introduction
Wave climate has assumed greater importance in the daily utilization of ocean resources for commerce and recreation as well as long-term planning for coastal land-use and hazard mitigation. Winds transfer momentum to the ocean providing the energy source for seas and swells, and as the atmospheric climate evolves so does the wave climate. It is therefore necessary to analyze surface winds and waves simultaneously to understand the physical and climatic processes. Datasets of wide coverage and sufficient duration are essential for interpretation of the global phenomena at numerous temporal and spatial scales.
Wind and wave data is available from buoys, voluntary observing ships, satellite altimetry, and numerical models. Buoys provide ground truth for validation of observational and model data. Despite the lack of spatial coverage, buoy data represents a critical source of information for assessment of climate variability (e.g., Bromirski et al., 2005, Menendez et al., 2008, Genmrich et al., 2011). Ship visual observations were shown by Gulev and Grigorieva (2006) as an accurate source of information, but have limited coverage in the Southern Hemisphere and in extreme conditions (Gulev et al., 2003). Numerous studies have used satellite altimetry to describe the wind and wave climate as well as extreme events (e.g., Young, 1999, Woolf et al., 2002, Chen et al., 2002, Hemer et al., 2010, Young et al., 2011, Izaguirre et al., 2011, Vinoth and Young, 2011). Third generation phase-averaged wave models such as WAve Model (WAM) described by WAMDIG (1988) and WAVEWATCH III (WW3) of Tolman et al. (2002) have greatly complemented observational data in terms of resolution and coverage providing new insights into the global wave climate as well as smaller scale processes.
The quality of the wave data from numerical models depends on the wind forcing. Reanalysis datasets, which are based on the best available models and data assimilation techniques, are most suitable for production of wave datasets. The National Centers for Environmental Predictions (NCEP) produced the global Reanalysis I and II (R1 and R2) datasets covering respectively from 1948 and 1979 to the present time at 1.9° resolution every 6 h (Kalnay et al., 1996, Kistler et al., 2001). The European Centre for Medium-Range Weather Forecasts (ECMWF) created the ERA-15 and ERA-40 for 1979–1993 and 1957–2001 at 1.125° available every 6 h (Sterl et al., 1998, Uppala et al., 2005). ERA-40 is a coupled atmosphere-wave model utilizing WAM to produce wave heights, periods, and directions at 1.5° resolution. Sterl and Caires (2005) validated the wave heights and statistically corrected the dataset with measurements to produce a wave atlas. Studies of hindcast wave data have identified relationships between the wave field and atmospheric cycles in the north Atlantic (Wang and Swail, 2001, Dodet et al., 2010), the Pacific Basin (Graham and Diaz, 2001, Tsai et al., 2012), the Southern Ocean (Hemer et al., 2010, Bosserelle et al., 2011), and globally (Semedo et al., 2011, Reguero et al., 2012, Fan et al., 2012).
Newly available reanalysis datasets, such as NCEP’s Climate Forecast System Reanalysis (CFSR) for 1979–2009 at 0.34° available every hour and ECMWF’s ERA-Interim for 1979-present at ∼0.7° available every 3 h, have updated physics and improved assimilation (Saha et al., 2010, Dee et al., 2011). The CFSR and ERA-Interim datasets perform superior to earlier reanalysis datasets developed by NCEP and ECMWF. These datasets provide an opportunity to re-evaluate the global wind and wave climate at much higher fidelity and resolution than before. In particular, the main improvements in CFSR are coupling between ocean, atmosphere, and land surface processes; an interactive sea ice model; assimilation of satellite radiances to provide temperature and moisture profiles; and increased horizontal and vertical resolution in the atmosphere model. NCEP has utilized the CFSR surface winds to create a hindcast wave dataset at 0.5° resolution using WW3 v3.14 for the period 1979–2009 (Chawla et al., 2013). Using the two-way nesting algorithm of Tolman (2008), higher resolution grids as fine as 1/15° are nested into the global wave model to cover semi-enclosed seas and nearshore regions.
The present study provides a systematic analysis of the newly released CFSR wind and WW3 generated wave datasets from NCEP to complement the numerous studies at regional and global scales. The high-resolution data around the globe allows examination of large and small-scales features as well as their relationships with climate cycles. In this paper, we summarize atmospheric climate cycles and illustrate representative signals from the literature in Section 2. This provides the background for the subsequent analysis of the wind and wave climate. Section 3 provides a summary of the methodologies and validation of the CFSR data products from Saha et al. (2010) and Chawla et al. (2013). Data anomalies are pointed out and their effects on the climate analysis are discussed. Section 4 illustrates the seasonal pattern and statistical properties, while Section 5 discusses annual and inter-annual variability of the wind and wave datasets. Section 6 analyzes the variability of the datasets against published indices from atmospheric oscillations. The results from Sections 4 Seasonal patterns and statistics, 5 Annual and inter-annual variability, 6 Correlation to climate cycles are highlighted for discussion in terms of their implications to ocean modeling and the connections to climate cycles. Section 7 gives a summary of the major findings pertinent to climate research and wave modeling communities.
Section snippets
Climate cycles
The CFSR provides continuous and consistent coverage of the atmospheric conditions around the globe for 31 years. This allows examination of climate cycles on global and regional scales with periods up to several years. These include the Arctic Oscillation (AO), Antarctic Oscillation (AAO), El Nino Southern Oscillation (ENSO), and Madden-Julian Oscillation (MJO). Longer cycles such as the Pacific Decadal Oscillation (PDO) and the Atlantic Multi-decadal Oscillation (AMO), which have periods of 10
CFSR data products
CFSR comprises a suite of coupled atmosphere, ocean circulation, land surface, and sea ice models for environmental prediction (Saha et al., 2010). The Global Forecast System (GFS) of Yang et al. (2006) constitutes the atmosphere model, which has ∼38 km horizontal resolution and 64 vertical layers extending from the surface to 0.26 hPa. The Geophysical Fluid Dynamic Lab’s Modular Ocean Model (MOM) version 4 describes the ocean circulation at 0.25° resolution in the equatorial region and 0.5°
Seasonal patterns and statistics
Seasonal patterns are described by averaging the wind speed and significant wave height in the four quarters: December–January–February (DJF), March–April–May (MAM), June–July–August (JJA), and September–October–November (SON). Fig. 2, Fig. 3 show the respective seasonal averages from the 31 years of data. The directional arrows for the wind speed are computed from the two orthogonal components, while the directional arrows for the significant wave height present the average peak direction. DJF
Annual and inter-annual variability
The mean annual variability measures the spread of the data to provide an indication of the seasonal extremes. Let x denote the time series of wind speed or wave height over a period of m years with n records each. The mean annual variability (MAV), which is the average of the annual standard deviation normalized by the annual average, provides a measure of the variability within each year aswhere the indices j and k refer to the year and
Correlation to climate cycles
The IAV and MAV hint that some of the variability in the winds and waves may be linked to atmospheric cycles. In this section, we examine the relationship between the wind and wave dataset as well as their correlation with the published indices of the AO, AAO, ENSO, and MJO in Section 2. Let xi and yi denote two time series with n records each. The correlation coefficient is defined aswhere cov(x,y) is the covariance and σx and σy are
Conclusions
The CFSR wind and WW3 generated wave datasets provide a wealth of information for examination and interpretation of climate characteristics. The high resolution of 0.5° reveals large and small scale features over a continuous 31-year period for studies of the seasonal patterns as well as long-term climate cycles. Seasonal averages and percentiles reveal the overall features of the global wind and wave climate with strong zonal structures. The Northern Hemisphere has stronger seasonal patterns,
Acknowledgement
The Department of Energy supported this study through Grant No. DE-FG36-08G018180 via the National Marine Renewable Energy Center. We would like to thank the guest editor Dr. Mark Hemer, Dr. Fernando Mendez, and two anonymous reviewers for the comments and suggestions that have improved this paper. SOEST Contribution Number 8699.
References (50)
Numerical modeling of ocean swell contributions to the global wind-wave climate
Ocean Modelling
(2006)- et al.
Atlas of global wave energy from 10 years of reanalysis and hindcast data
Renewable Energy
(2012) - et al.
Obstruction grids for spectral wave models
Ocean Modelling
(2008) - et al.
Wave climate variability in the North-East Atlantic Ocean over the last six decades
Ocean Modelling
(2010) - et al.
A global ocean wave (GOW) calibrated reanalysis from 1948 onwards
Coastal Engineering
(2012) - et al.
The atmospheric bridge: the influence of ENSO teleconnections on air-sea interaction over the global oceans
Journal of Climate
(2002) - et al.
Wave climate changes in the north Atlantic and north sea
International Journal of Climatology
(1991) - et al.
Inter-annual variability and longer-term changes in the wave climate of Western Australia between 1970 and 2009
Ocean Dynamics
(2011) - Bromirski, P.D., Cayan, D.R., Flick, R.E., 2005. Wave spectral energy variability in the northeast Pacific. Journal of...
- et al.
Intercomparison of different wind-wave reanalysis
Journal of Climate
(2004)
A global view of swell and wind sea climate in the ocean by satellite altimeter and scatterometer
Journal of Atmospheric Oceanic Technology
Global ocean surface wave simulation using a coupled atmosphere-wave model
Journal of Climate
Evidence for intensification of North Pacific winter cyclones since 1948
Bulletin of American Meteorological Society
Observational changes and trends in the northeast Pacific wave records
Geophysical Research Letters
Variability of the winter wind waves and swell in the North Atlantic and North Pacific as revealed by voluntary observing ship data
American Meteorological Society
Variability and trends in the directional wave climate of the Southern Hemisphere
International Journal of Climatology
Extreme precipitation events in the western United States related to tropical forcing
Journal of Climate
Global extreme wave height variability based on satellite data
Geophysical Research Letters
Improving incremental balance in the GSI 3DVAR analysis system
Monthly Weather Review
Cited by (0)
- 1
Tel.: +1 (808) 956 8198; fax: +1 (808) 956 3498.
- 2
Tel.: +1 (301) 763 8000x7253; fax: +1 (301) 763 8545.
- 3
Tel.: +1 (301) 763 8000x7209; fax: +1 (301) 763 8545.