Statistical wave climate projections for coastal impact assessments. Issue 9 (14th September 2017)

• Record Type: Journal Article
• Authors: Camus, P.; Losada, I. J.; Izaguirre, C.; Espejo, A.; Menéndez, M.; Pérez, J.
• Is Part Of: Earth's future. Volume 5:Issue 9(2017)
• Page Start: 918
• Page End: 933
• Publication Date: 2017-09-14
• Subjects: Multivariate wave climate -- Multi‐model ensemble projections -- Statistical downscaling -- Climate change coastal impacts; Environmental sciences -- Periodicals; Environmental sciences; Periodicals; 550
• DOI: 10.1002/2017EF000609
• Languages: English
• ISSNs: 2328-4277

Abstract

Abstract: Global multimodel wave climate projections are obtained at 1.0° × 1.0° scale from 30 Coupled Model Intercomparison Project Phase 5 (CMIP5) global circulation model (GCM) realizations. A semi‐supervised weather‐typing approach based on a characterization of the ocean wave generation areas and the historical wave information from the recent GOW2 database are used to train the statistical model. This framework is also applied to obtain high resolution projections of coastal wave climate and coastal impacts as port operability and coastal flooding. Regional projections are estimated using the collection of weather types at spacing of 1.0°. This assumption is feasible because the predictor is defined based on the wave generation area and the classification is guided by the local wave climate. The assessment of future changes in coastal impacts is based on direct downscaling of indicators defined by empirical formulations (total water level for coastal flooding and number of hours per year with overtopping for port operability). Global multimodel projections of the significant wave height and peak period are consistent with changes obtained in previous studies. Statistical confidence of expected changes is obtained due to the large number of GCMs to construct the ensemble. The proposed methodology is proved to be flexible to project wave climate at different spatial scales. Regional changes of additional variables as wave direction or other statistics can be estimated from the future empirical distribution with extreme values restricted to high percentiles (i.e., 95th, 99th percentiles). The statistical framework can also be applied to evaluate regional coastal impacts integrating changes in storminess and sea level rise. Key Points: Statistical global multimodel wave climate projections under different representative concentration pathways The proposed global statistical downscaling framework allows an easy regional projection of wave climate Climate change coastal impact assessment based on downscaling impact indicators using the global statistical framework