Tide-gauge-based diagnosis of sea-level-rise-sensitive threshold exceedance and adaptation-relevant elevation bands along Southwestern Taiwan

Infographic of five tide gauges in southwestern Taiwan, 2004-2025: sea level peaks in August at every station, the seasonal range differs by station, and sea-level rise of 0.5 to 1.0 m adds up to 239 to 345 exceedance days per year. Sea-level rise is expected to increase coastal water-level exceedance, particularly in low-lying and subsidence-prone regions. This study develops a tide-gauge-based diagnostic framework to examine sea-level-rise-sensitive threshold exceedance and adaptation-relevant elevation bands along the southwestern coast of Taiwan. Quality-controlled six-minute tide-gauge records from Boziliao, Dongshih, Jiangjun, Kaohsiung, and Donggang during 2004–2025 were aggregated to hourly and monthly tide-gauge water levels. We quantified annual sea-level cycle climatology, annual range changes, non-parametric trends, threshold-response curves under fixed sea-level-rise offsets of 0.10, 0.30, 0.50, and 1.00 m, and station-specific sensitive elevation bands. The five stations showed a coherent seasonal phase, with the annual sea-level cycle maximum occurring in August at all stations, but the range varied from 0.221 m at Donggang to 0.349 m at Dongshih. No station showed a statistically significant monotonic trend in annual range. Fixed sea-level-rise offsets substantially increased annual exceedance duration, expressed as day-equivalents per year, with maximum increases ranging from 79.1 to 239.3 day-equivalents yr−1 under 0.50 m and from 145.2 to 345.4 day-equivalents yr−1 under 1.00 m. Kaohsiung and Donggang showed the largest amplification, whereas Dongshih had the largest seasonal range. This separation shows that seasonal amplitude and threshold-exceedance sensitivity are related but non-equivalent. These results indicate that Kaohsiung and Donggang should be prioritized for follow-up site-specific assessment because their observed water-level distributions show the strongest exceedance-duration response to fixed SLR offsets.