The Heat Is Online

Microseismological evidence for a changing wave climate in the Northeast Atlantic Ocean

Nature 408, 349 - 352 (2000) © Macmillan Publishers Ltd. (Nov. 16, 2000)

Microseismological evidence for a changing wave climate in the northeast Atlantic Ocean


* Department of Earth Sciences, University of Bremen, Klagenfurter Str., Bldg. GEO, 28359 Bremen, Germany
† Geophysical Observatory, University of Hamburg, Kuhtrift 18, 21075 Hamburg, Germany
‡ Institute of Oceanography, University of Hamburg, Troplowitzstr. 7, 22529 Hamburg, Germany

Correspondence and requests for materials should be addressed to I.G. (e-mail:

One possible consequence of a change in climate over the past several decades is an increase in wave heights, potentially threatening coastal areas as well as the marine industry1-4. But the difficulties in observing wave heights exacerbates a general problem of climate-change detection: inhomogeneities in long-term observational records owing to changes in the instruments or techniques used, which may cause artificial trends5, 6. Ground movements with periods of 4–16 seconds, known as microseisms, are associated with ocean waves and coastal surf 7-10, and have been recorded continuously since the early days of seismology. Here we use such a 40-year record of wintertime microseisms from Hamburg, Germany, to reconstruct the wave climate in the northeast Atlantic Ocean. For the period 1954–77, we detect an average of seven days per month with strong microseismic activity, without a significant trend. This number increases significantly in the second half of the record, reaching approximately 14 days of strong microseisms per month. The implied increase in northeast Atlantic wave height over the past 20 years parallels increased surface air temperatures11 and storminess12 in this region, suggesting a common forcing.

Observations of the Earth's near-surface temperature show a global increase since 1901, occurring from 1925–1944 and 1978–1997. Over these periods global temperature rose by 0.37 and 0.32 K, respectively11. The temperature change over the past decades is unlikely to be entirely due to internal climate variability13-15 and has been attributed to changes in the concentration of greenhouse gases caused by human activity16, 17. There are, however, basic physical relationships between temperature, air pressure and wind fields17, 18 (and hence wave fields4, 19). Consequently, many people are concerned about the possibility of an intensification of extratropical storms20, 21. However, because of inhomogeneities in historical data sets, the impact of climate changes on the temporal evolution of the storm and wave climate is difficult to reveal.

In climatology, homogeneity and therefore quality of data is essential5, 6. A climatological time series is termed homogeneous22 if the variations exhibited by the series are solely the result of the vagaries of the weather and climate. Therefore, the methodological challenge with the analysis of historical data sets is the discrimination between signals that reflect real changes and signals that reflect changes that are attributable to improved instrumental accuracies, altered environmental conditions, observational practices, data coverage and analysis routines. In terms of wave height, data are available from reports of visual assessments from ships of opportunity and lighthouses, from river buoys and shipborne wave recorders at ocean weather stations; wave height maps have also been constructed for the purpose of ship routing from wind analyses. Analyses of these data have revealed a substantial worsening of the wave climate in the north Atlantic Ocean1-3. However, these data are sparse and suffer from various inhomogeneities23.

In the late nineteenth century seismologists started to record movements of the ground continuously. In addition to ordinary earthquakes seismographs almost always record small movements in the Earth's crust, called microseisms. It has rapidly been recognized that several parts of microseisms (periods of 4 to 16 s) are strongly correlated with oceanographic forcing and atmospheric disturbances. In 1904 Wiechert7 considered the surf to be the main cause of microseisms produced by extratropical low-pressure areas.

Many authors confirmed this idea by experimental data8, 24, 25. However, microseismic energy could also be generated in the oceans away from coastal areas9, 26, 27 and Gutenberg9 showed that microseisms could be used in weather forecasting, especially in locating tropical disturbances. Later a mathematical theory was established10, 27 that explains the generation of microseisms by standing ocean waves and coastal surf. Although this permitted ocean wave heights to be calculated from seismometer data28, most seismologists considered microseisms primarily to be noise in seismic recordings. Today, however, these data can be re-examined to assess the wave climate of the twentieth century.

We have shown that microseisms at the seismological station HAM (Hamburg, Germany) are related to the northeast Atlantic wave climate24. Large-amplitude microseismic ground motions occurring at HAM are primarily related to secondary microseisms10, 26, 27 with periods of 6–8 s.

Here, we present historical microseismic data recorded between 1954 and 1998. This period covers the time where other climatological data sets detected major changes in, for example, the surface air temperature11, the wintertime north Atlantic oscillation (NAO) index18 and the northwestern European storm climate4, 12. The seismic records used are those from a Wiechert vertical seismograph (1,300 kg pendulum) and a Sprengnether long-period vertical seismometer. The instruments were operated from 1954–1975 and 1974–1998, respectively. To yield absolute ground motion from different seismometers the recordings have to be convolved with an instrument-dependent response function. Historical seismological data, however, are analogue recordings. In the analysis of microseisms it is not feasible to digitize 40 years of continuously recorded data. The standard approach to obtaining daily statistics of microseisms from analogue data is to read the amplitude and frequency of the strongest microseismic ground motions over a fixed time window of 1–2 hours per day29. Using digital data from the wintertime microseisms of the years 1992 to 1998 (ref. 24) we were able to show that this technique provided reliable averages. In addition, we introduce a measure called the microseismic index that should be independent of the instrumental response within the narrow band of periods considered in our analysis. To characterize days with strong microseisms, we define a threshold amplitude of ground movements which is significantly above the noise level and above the average microseisms recorded during summer seasons. If this threshold is reached, we consider the day to be affected by microseismic activity. Thus, the microseismic index defines the number of days per month affected by strong microseisms and is therefore more a qualitative measure indicating overall changes of the wave climate rather than a quantitative measure yielding absolute changes of wave height.

Figure 2 shows the annual averages of the microseismic index. Like other climatological time series it displays a considerable degree of year-to-year variability. There is a noticeable increase (estimated by a linear trend calculated using least squares) in the frequency of microseismic storms, that is, the number of days per month affected by microseisms increases by 0.26 per year. However, the data clearly indicate two different periods; from 1954–1977 the number of microseisms that occur increases only slightly, settling at a steady rate of 7 days per month, but increases significantly from 1978–1998 to 14 days per month. This fact is visible even in the data obtained from the Sprengnether seismometer alone; operated since 1974, it provided the lowest values in its first years. Thereafter, values double within a few years.

The same trend could be observed at the Seven Stones Light Vessel off Land's End1 (Fig. 2). It was the first site in the world where a wave recorder was installed; annual mean values of significant wave height were used1 to examine whether the northeast Atlantic has become rougher in recent years. The measurements published cover the years 1962 to 1985. We note that, just as for the microseismic index, annual means obtained before 1978 are always lower than those obtained thereafter. Unfortunately, there is no homogeneous data set available for comparison. However, problems of inhomogeneities were overcome by feeding a numerical wave model (WAM)19 with historical surface pressure and wind distribution data. This simulation19 suggested, for the Norwegian Sea at the weather station OWS Mike, an increase of significant wave heights over the period 1955–1994 of the order of 8 cm yr-1 for the annual maxima. The increase in wave height was even more significant in the second part of the simulated period, 1975–1994. During that time the maxima of significant wave height increased by more than 17 cm yr-1. A formal correlation of microseisms and wave hindcast data gave a regression coefficient of r = 0.4 which increases to 0.61 when using data smoothed with a five-year gaussian filter. Even the 99 and 90 percentiles indicate similar trends. In general, however, a two-step evolution is not evident in the simulations of the WASA project. Nonetheless, the model output suggests that statistics of significant wave height in the northern North Sea and the Norwegian Sea have undergone a steady increase since 1954 (refs 4, 19). Thus, in addition to previous studies, microseisms at HAM present a new and homogeneous data set that suggests and supports a worsening of the northeast Atlantic wave climate over the last two decades.

If we compare microseisms with the Northern Hemisphere (or global) temperatures (Fig. 2) we observe a similar trend; temperatures remaining nearly at a constant level between 1954 and 1977 and a warming of 0.32 K from 1978–1997 (ref. 11). Similar trends are also revealed by the NAO index and the northwestern European storm climate12. A two-phase evolution, however, is not evident; but in terms of an overall increase the correlation between the time series is statistically significant. The similarities between the different climatological records strongly suggest a common forcing. Recent simulations and analyses of the Earth's temperature pattern exclude purely natural forcing and attribute it largely to changes in the concentration of greenhouse gases and aerosol loading due to human activity13, 17. Therefore, it seems reasonable to propose that greenhouse forcing affects the ocean's wave climate and hence coastal surf and storm surges along northern Europe's coastlines, which in turn produced the observed increase of microseisms. However, large ocean waves and hence significant microseisms are generally related to the very high wind speed of storms. The storm climate itself seems to be comparable with that at the beginning of the twentieth century4, 12, though. Work is now required to backtrack microseisms and hence the wave climate into the early twentieth and late nineteenth century. Such a time series will allow us to understand the interaction between the NAO index, surface air temperature, storm frequency and intensity and the north Atlantic wave climate on a longer timescale.

Received 23 February 2000; accepted 24 August 2000



Carter, D. J. T. & Draper, L. Has the north-east Atlantic become rougher? Nature 332, 494 (1988).


Bouws, E., Jannick, D. & Komen, G. J. On increasing wave height in the North Atlantic Ocean. Bull. Am. Meteorol. Soc. 77, 2275-2277 (1996).


Bijl, W. Impact of wind climate change on the surge in the southern North Sea. Clim. Res. 8, 45-59 (1997).


WASA Group. Changing waves and storms in the Northeast Atlantic? Bull. Am. Meteorol. Soc. 79, 741-760 (1998).


Karl, T. R., Quayle, R. G. & Groisman, P. Y. Detecting climate variations and changes: new challenges for observing and data management systems. J. Clim. 6, 1481-1494 (1993).


Jones, P. D. in Analysis of Climate Variability (eds von Storch, H. & Navarra, A.) 53-76 (Springer, Berlin, 1995).


Wiechert, E. Discussion, Verhandlung der zweiten Internationalen Seismologischen Konferenz, Strasbourg. Gerlands Beitr. Geophysik 2, 41-43 (1904).


Gutenberg, B. Untersuchungen ber die Bodenunruhe mit Perioden von 4-10 Sekunden in Europa. Verff. Zentr. Bur. Int. Seismol. Assoz. 106 (1921).


Gutenberg, B. Microseisms and weather forecasting. J. Meteorol. 4, 21-28 (1947).


Hasselmann, K. Statistical analysis of the generation of microseisms. Rev. Geophys. 1, 177-210 (1963).


Jones, P. D., New, M., Parker, D. E., Martin, S. & Rigor, I. G. Surface air temperature and its changes over the past 150 years. Rev. Geophys. 37, 173-199 (1999).


Alexandersson, H., Smith, T., Iden, K. & Tuomenvirta, H. Long-term trend variations of the storm climate over NW Europe. Glob. Atmos. Ocean Sys. 6, 97-120 (1998).


Stouffer, R. J., Manabe, S. & Vinnikov, K. Y. Model assessment of the role of natural variability in recent global warming. Nature 367, 634-636 (1994).


Santer, B. D. et al. A search for human influences on the thermal structure of the atmosphere. Nature 382, 39-45 (1996).


Mann, M. E., Bradley, R. S. & Huges, M. K. Global-scale temperature patterns and climate forcing over the past six centuries. Nature 392, 779-787 (1998).


Tett, S. F. B., Stott, P. A., Allen, M. R., Ingram, W. J. I. & Mitchell, J. F. B. Causes of twentieth-century temperature change near the Earth's surface. Nature 399, 569-572 (1999).


Shindell, D. T., Miller, R. L., Schmidt, G. A. & Pandolfo, L. Simulation of recent northern winter climate trends by greenhouse-gas forcing. Nature 399, 452-455 (1999).


Rodwell, M. J., Rowell, D. P. & Folland, C. K. Oceanic forcing of the wintertime North Atlantic oscillation and European climate. Nature 399, 320-323 (1999).


Gnther, H. et al. The wave climate of the Northeast Atlantic over the period 1955-1994: the WASA wave hindcast. Glob. Atmos. Ocean Sys. 6, 121-163 (1998).


Berz, G. Global warming and the insurance industry. Interdisciplinary Sci. Rev. 18, 120-125 (1993).


Berz, G. & Conrad, K. Stormy weather: the mounting wind-storm risk and consequences for the insurance industry. Eurodecision 12, 65-69 (1994).


Conrad, V. & Pollak, L. D. Methods in Climatology (Harvard Univ. Press, Cambridge, Massachusetts, 1962).


WASA Group. Comment on Increases in Wave Heights over the North Atlantic: a review of the evidence and some implications for the naval architect by N. Hogben. Trans. R. Inst. Naval Arch. 137, 107-110 (1994).


Essen, H.-H., Klussmann, J., Herber, R. & Grevemeyer, I. Do microseisms in Hamburg (Germany) reflect the wave climate of the North Atlantic? Germ. J. Hydrogr. 51, 33-45 (1999).


Darbyshire, J. Analysis of twenty microseism storms during the winter of 1987-1988 and comparison with wave hindcasts. Phys. Earth Planet. Int. 63, 181-195 (1990).


Longuet-Higgins, M. S. & Ursell, F. Sea waves and microseisms. Nature 162, 700 (1948).


Longuet-Higgins, M. S. A theory of the origin of microseisms. Phil. Trans. R. Soc. Lond. A 243, 1-35 (1950).


Bromirski, P. D., Flick, R. E. & Graham, N. Ocean wave height determined from inland seismometer data: Implications for investigating wave climate changes in the NE Pacific. J. Geophys. Res. 104, 20753-20766 (1999).


Bath, M. An Investigation of the Uppsala Microseisms (Institute of Meteorology, Royal Univ. Uppsala, Report No. 14, Uppsala, 1949).

Acknowledgements. We thank G. Spars for assistance in analysing the historical seismological records. This work benefited from support of the Deutsche Forschungsgemeinschaft for the SFB 512 "Cyclones and the North Atlantic Climate System".