45-Year High-Resolution Wave Hindcast Reveals Fiji’s Climate, Spectra, and Long-Term Trends

Fiji’s islands are being shaped by a hidden force that arrives from thousands of kilometres away: ocean waves generated by distant storms. A new 45-year reconstruction of the archipelago’s wave climate shows that energy is distributed unevenly across the country, funneled through narrow corridors, blocked by island groups and reefs, and gradually increasing offshore. The study provides the most detailed long-term picture yet of how waves move around Fiji—and offers a warning that even modest changes in offshore conditions could add pressure to coastlines already facing sea-level rise, erosion and flooding.

The research team reconstructed wave conditions from 1979 through 2023 using a high-resolution numerical model known as SWAN, or Simulating WAves Nearshore. Unlike a conventional grid with uniform spacing, the model used an unstructured mesh that varied from about 25 kilometres offshore to 500 metres near the coast. That resolution allowed the simulation to represent island shadowing, reef-fringed shorelines and narrow passages that can redirect or weaken incoming swell. The model covered the entire Fijian archipelago and calculated hourly wave height, peak period and direction, as well as the distribution of energy across different frequencies and directions.

Wave height alone does not tell the whole story. The researchers therefore analysed the directional wave spectrum, a technical description of how wave energy is divided among frequencies and arrival directions. Short-period waves, typically lasting one to five seconds, are usually generated locally by nearby winds. Waves with periods of five to 15 seconds generally reflect more developed wind seas or regional swell, while waves exceeding 15 seconds are characteristic of long-travelled swell produced by powerful storms in distant mid-latitude oceans. By separating these components, the team could identify not only where Fiji’s waves are strongest, but also which weather systems and climate patterns are responsible.

The model revealed a striking south-to-north energy gradient. Offshore areas south of Fiji experience annual mean significant wave heights above 2.3 metres, largely because they face the Southern Ocean, where strong extratropical storms generate long-period swell. Within the island chain, wave heights decline sharply as energy encounters land, shallow reefs and complex bathymetry. Sheltered waters, including the Bligh Waters between Viti Levu and Vanua Levu, typically experience annual mean significant wave heights of only 0.5 to 1.5 metres. Yet the islands do not simply form a barrier. Some channels act as persistent wave corridors, allowing swell to travel northward through gaps between Kadavu, Totoya and Gau and reach waters near Taveuni.

The contrast is especially clear between southern Kadavu and northern Labasa. At Kadavu, the annual spectrum is dominated by energy arriving from the southeast through southwest, with equivalent spectral wave heights reaching about 71 centimetres in the five-to-15-second bands. This pattern reflects direct exposure to swell from the Southern Ocean and the Tasman Sea. Labasa, in northern Fiji, is much more sheltered from that southern source. Its principal energy peak is smaller and concentrated in five-to-10-second waves arriving from north-northeast and north-northwest, with a secondary peak from the northeast in the 10-to-15-second range. The northern spectrum reflects a mixture of trade-wind seas and swell arriving from the North Pacific.

Seasonal changes intensify this geographical pattern. During the austral autumn and winter, from March through August, stronger storm activity in the Southern Ocean and Tasman Sea sends larger, more southerly swells toward Fiji. Offshore significant wave heights reach about 2.5 to 2.6 metres, while wave power exceeds 20 kilowatts per metre in exposed southern waters. Some of that energy travels through the Kadavu–Totoya corridor and across the Koro Sea, extending the influence of Southern Ocean swell far into the archipelago. In summer and spring, offshore wave heights are generally lower, although southern coasts can still receive mean southerly waves around 1.5 metres. Wave power falls to roughly 13–15 kilowatts per metre offshore during summer and rises again during the transition into autumn.

The study also connected Fiji’s changing wave patterns to large-scale climate oscillations. In the north, the Oceanic Niño Index, a measure of El Niño–La Niña conditions, was the strongest interannual influence. Positive ONI values were associated with enhanced energy across many directional and frequency bands at Labasa, including short-period waves and longer swell arriving from northern sectors. El Niño can shift storm tracks, alter tropical cyclone activity and reposition the South Pacific Convergence Zone, a broad band of unsettled weather that influences wind patterns across the region. The Pacific Decadal Oscillation produced a broadly similar but weaker signal in northern Fiji.

Southern Fiji responded to a different climate control. At Kadavu, the Antarctic Oscillation—also called the Southern Annular Mode—was strongly linked to long-period swell arriving from the south and southwest. Positive phases of the index are associated with changes in the position and strength of the Southern Hemisphere westerly winds. When those winds intensify or shift poleward, storms in the Southern Ocean can generate more powerful, longer-period swell that propagates north toward Fiji. The contrasting influence of the Antarctic Oscillation in the south and the Oceanic Niño Index in the north reveals a hemispheric transition in Fiji’s wave climate: Southern Ocean storminess dominates the exposed southern islands, while North Pacific and tropical atmospheric variability becomes more important farther north.

After testing the reconstruction against observations, the researchers found that it reproduced ordinary wave conditions with considerable skill. Comparisons with more than 1.5 million quality-controlled satellite altimetry measurements from 2018 to 2023 produced a correlation coefficient of 0.93 for significant wave height. The model’s average wave height was 1.91 metres, compared with 1.84 metres in the satellite observations, and its root-mean-square error was 0.26 metres. Four deep-water wave buoys south of Fiji provided an independent check. At those sites, correlations reached 0.93 and errors in wave height remained below 0.27 metres. The agreement was weaker in sheltered coastal waters, where radar altimeters can be contaminated by nearby land and the model’s bathymetry cannot fully resolve intricate reef systems.

The reconstruction does, however, have an important limitation: it is not designed to capture Fiji’s most extreme cyclone waves. The simulation was forced with winds from the ERA5 atmospheric reanalysis, whose approximately 31-kilometre resolution tends to underestimate the peak wind speeds of tropical cyclones and other intense systems. As a result, the model generally underestimated wave heights above four metres, particularly during high-energy events. The authors stress that the dataset is best suited to studying average wave climate, seasonal swell, wave-energy resources and long-term trends—not the full hazard posed by cyclone-generated extremes. A separate probabilistic modelling effort using enhanced cyclone wind fields will be needed to estimate those events more realistically.

Despite that caveat, the long-term trend is clear in the offshore record. Across the model domain, significant wave height increased by about 0.5 millimetres per year, equivalent to a cumulative rise of 2.4 centimetres over 45 years, with an uncertainty of plus or minus 2.8 centimetres. Peak wave period increased by about 0.004 seconds per year, or 0.2 seconds over the study period, with an uncertainty of 0.13 seconds. The statistically significant changes were concentrated offshore, especially in southern and southwestern sectors for wave height and in several offshore sectors for wave period. Longer periods mean waves carry energy farther and can interact with coastlines differently from shorter wind waves, potentially increasing shoreline run-up even when changes in average height are small.

The spectral results indicate that this increase is not uniform. At Labasa, positive trends appeared mainly in waves arriving from the northwest to northeast, including both shorter-period seas and long-period swell. At Kadavu, the strongest increases occurred in long-period swell from the south and southwest, with some directional spectral trends reaching 0.190 centimetres per year. At the same time, short- and intermediate-period waves from easterly and southeasterly directions declined in several seasons. Fiji’s future wave climate may therefore involve a redistribution of energy among directions and periods rather than a simple increase everywhere.

For coastal communities, that distinction matters. Waves transport sediment, reshape beaches and reefs, and can push seawater inland during high tides or storms. Distant swell can produce flooding even when local weather appears calm, because long-period waves carry energy across entire ocean basins and can generate higher run-up when they reach shallow coastal waters. Fiji’s low-lying communities and infrastructure are already exposed to rising seas, shoreline erosion and episodic inundation. The new hindcast gives planners a much sharper baseline for identifying which coastlines are persistently exposed, which are protected by island geometry, and where wave-energy projects might be feasible. It also makes clear that a national coastal strategy cannot rely on a single average wave value: the hazards faced by Kadavu, Labasa and the sheltered waters between Fiji’s major islands are shaped by fundamentally different combinations of ocean physics and climate variability.