INTERNATIONAL JOURNAL OF CLIMATOLOGY Int. J. Climatol. 25: 963–977 (2005) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/joc.1175 CLIMATOLOGY, VARIABILITY AND EXTREMA OF OCEAN WAVES: THE WEB-BASED KNMI/ERA-40 WAVE ATLAS ANDREAS STERL a, * and SOFIA CAIRES a,b a Royal Netherlands Meteorological Institute (KNMI), De Bilt, Netherlands b Meteorological Service of Canada, Climate Research Branch, Downsview, Ontario, Canada Received 9 April 2004 Revised 17 August 2004 Accepted 18 August 2004 ABSTRACT The European Centre for Medium-Range Weather Forecasts (ECMWF) has recently finished ERA-40, a reanalysis covering the period September 1957 to August 2002. One of the products of ERA-40 consists of six-hourly global fields of wave parameters, like significant wave height and wave period. These data have been generated with the centre’s WAM wave model. From these results we have derived climatologies of important wave parameters, including significant wave height, mean wave period, and extreme significant wave heights. Particular emphasis is on the variability of these parameters, both in space and time. Besides being important for scientists studying climate change, these results are also important for engineers who have to design maritime constructions. This paper describes the ERA-40 data and gives an overview of the results derived. The results are available on a global 1.5 ° × 1.5 ° grid. They are accessible from the Web-based KNMI/ERA-40 wave atlas at http://www.knmi.nl/waveatlas. Copyright  2005 Royal Meteorological Society. KEY WORDS: waves; climatology; extremes; variability; reanalysis 1. INTRODUCTION The European Centre for Medium-Range Weather Forecasts (ECMWF) has recently completed the computa- tions of the ERA-40 dataset, a reanalysis of global meteorological variables, among which are ocean surface wind waves, from September 1957 to August 2002 (45 years). The reanalysis was produced by ECMWF’s Integrated Forecasting System (IFS) that uses variational data assimilation. In terms of sea-state data, this reanalysis is the first in which an ocean wind wave model is coupled to the atmosphere. Moreover, its final product consists of the longest and most complete global wave dataset available. It is given on a 1.5 ° × 1.5 ° latitude–longitude grid covering the whole globe. The continuous 45 year length of the ERA-40 datasets makes it especially suitable for studying climate variability and for estimating extreme values of certain wave parameters, e.g. the 100 year return wave height. As part of the ERA-40 project, we have extensively assessed the quality of the wave-related parameters and used them to build the Web-based KNMI/ERA-40 wave atlas describing the global wave climate. This paper gives an overview of the verification work done and highlights the main features of the atlas. The atlas contains some explanatory text, a basic description of the wind and wave climates in terms of means and variability, and wave statistics that are important in ocean engineering and naval architecture. The objective is twofold. On the one hand, the atlas aims at providing a global description of the ocean climate by means of simple statistical measures. On the other hand, it aims at revealing the existence of decadal variability in the wave climate and showing the extent to which this variability affects the estimates * Correspondence to: Andreas Sterl, Royal Netherlands Meteorological Institute (KNMI), De Bilt, The Netherlands; e-mail: sterl@knmi.nl Copyright  2005 Royal Meteorological Society