Hindcast Data Research Articles

Improving the Spring Air Temperature Forecast Skills of BCC_CSM1.1 (m) by Spatial Disaggregation and Bias Correction: Importance of Trend Correction

In this study, an improved method named spatial disaggregation and detrended bias correction (SDDBC) based on spatial disaggregation and bias correction (SDBC) combined with trend correction was proposed. Using data from meteorological stations over China from 1991 to 2020 and the seasonal hindcast data from the Beijing Climate Center Climate System Model (BCC_CSM1.1 (m)), the performances of the model, SDBC, and SDDBC in spring temperature forecasts were evaluated. The results showed that the observed spring temperature exhibits a significant increasing trend in most of China, but the warming trend simulated by the model was obviously smaller. SDBC performed poorly in temperature trend correction. With SDDBC, the model’s deviation in temperature trend was corrected, and consequently, the temporal correlation between the model’s simulation and the observation as well as the forecasting skill on the phase of temperature were improved, thus improving the MSSS and the ACC. From the perspective of probabilistic prediction, the relative operating characteristic skill score (ROCSS) and the Brier skill score (BSS) of the SDDBC for three categorical forecasts were higher than those of the model and SDBC. The SDDBC’s BSS increased as the effect of the increasing resolution component was greater than that of the decreasing reliability component. Therefore, it is necessary to correct the predicted temperature trend in post-processing for the output of numerical prediction models.

Combined Effect of ENSO and AO on Winter Temperatures of the Korean Peninsula on Subseasonal Time Scales

AbstractThe combined effect of the El Niño–Southern Oscillation (ENSO) and Arctic Oscillation (AO) on the variability of boreal winter (December–February) temperature over South Korea is examined at the subseasonal time scale using subseasonal-to-seasonal (S2S) hindcast data. Daily hindcast data from the European Centre for Medium-Range Weather Forecasts (ECMWF) database is used. We selected the following six composite cases using a threshold of ±0.5 for each index: El Niño and positive AO (EP), El Niño and negative AO (EN), La Niña and positive AO (LP), La Niña and negative AO (LN), positive AO only (PA), and negative AO only (NA). Results from reanalysis data suggest the possibility of using these two climate factors as predictors for 1-month prediction of South Korea up to 4 weeks in advance. Thus, we confirmed that the ENSO plays a statistically significant role in strengthening (weakening) the AO influences on the temperature anomalies in the in phase (out of phase). For example, there is a significant increase (decrease) in mean temperature anomalies through positive (negative) geopotential height (GPH) anomalies and warm (cold) temperature advection over South Korea in the EP (LN) case. The ECMWF S2S hindcast demonstrated an acceptable ability to reproduce circulation patterns over East Asia up to 3 weeks in advance, and sufficiently predicted weekly mean temperature anomalies over South Korea in EP, LN, and PA cases.

An application of multiple‐point statistics downscaling approach over North‐West Himalayas in avalanche‐prone areas

AbstractWeather data at fine spatial resolution in the Himalayan region are important for assessing various hydrological events such as avalanches. In recent years, national agencies have been exploring the use of Weather Research and Forecasting (WRF) model in this region. Setting up a WRF model with fine spatial grids in the Himalayan region is challenging due to large topographical variations and associated computational cost. In this study, we use High Asia Refined analysis (HAR) dataset to set up a Multiple‐Point Statistics (MPS) based downscaling approach using the hindcast data of years 2001–2012 at both coarse and fine spatial grids. A total of 11 covariates are used in the study belonging to various environmental factors such as weather, location, topography and vegetation. We generate multiple realizations of precipitation and temperature for the year 2013. The efficacy of the MPS model is tested in a gridbox consisting of 61 × 61 grids, in 10 districts and at sites prone to avalanches and extreme precipitation events. Results show that the downscaled precipitation and temperature have reasonable accuracy. Mean monthly root mean square error and mean absolute error values for precipitation range between 2.87–8.0 mm and 1.45–3.27 mm, respectively. For temperature, both the error statistics range between 0.65–1.13 K and 0.51–0.83 K, respectively. Overall, the study shows that MPS is an effective tool for generating reliable fine‐resolution weather information over multiple geographical locations, overcoming the diverse challenges of the Northwest Himalayas. Furthermore, it is capable of providing consistent fine‐resolution weather data at point locations. Such data may be highly sensitive for hydrological studies, such as avalanche forecasting in the Himalayas.

North Atlantic Oscillation (NAO) Climate Index Hidden in Ocean Generated Secondary Microseisms

AbstractWith the role of surface winds in the generation of ocean waves, secondary microseisms induced by ocean wave‐wave interactions represent a unique interconnection between the solid Earth, the ocean and global atmospheric circulation patterns. In this study, temporal changes in the ocean generated seismic wavefield in Northeast Atlantic are monitored offshore West of Ireland using ocean bottom seismometers located on top of a thick sedimentary basin. Comparisons with numerical seismic simulations and ocean wave model hindcast data suggest those variations are correlated with changing patterns in ocean wave‐wave interactions closely linked to secondary microseism generation areas. Here we show how those changes, accentuated by the specific structure of the Irish offshore margin, reveal the signature of the North Atlantic Oscillation (NAO) in the Earth’s background seismic wavefield. In the North Atlantic, secondary microseism sources likely fluctuate with the changing storm track in response to variations in the NAO.

Drivers of extreme water levels in a large, urban, high-energy coastal estuary – A case study of the San Francisco Bay

Reliable and long-term hindcast data of water levels are essential in quantifying return period and values of extreme water levels. In order to inform design decisions on a local flood control district level, process-based numerical modeling has proven an essential tool to provide the needed temporal and spatial coverage for different extreme value analysis methods. To determine the importance of different physical processes to the extreme water levels we developed a process-based numerical model (Delft3D Flexible Mesh) and applied it to simulate a large, urban, high-energy coastal estuary (the San Francisco Bay). The unstructured grid with 1D/2DH model elements, allows for efficient model simulations and therefore it was possible to simulate over 70 years between 1950 and 2019. Results show significant skill in reproducing observations for the entire modeled time period with an average root-mean-square error of 8.0 cm. A process-based modeling approach allows for the explicit in- and exclusion of different physical processes to quantify their importance to the extremes. For the 100-year still water level (SWL), tide (70%) and non-tidal residual (NTR) (25%) explain the majority of the simulated high water levels in the Bay relative to Mean Higher High Water (MHHW). However, closer to the Delta, local fluvial inflow increases in importance. For longer return periods, the importance of tide decreases and the importance of remote NTRs and fluvial inflow increases.

A new CAM6 + DART reanalysis with surface forcing from CAM6 to other CESM models

An ensemble Kalman filter reanalysis has been archived in the Research Data Archive at the National Center for Atmospheric Research. It used a CAM6 configuration of the Community Earth System Model (CESM), several million observations per day, and the Data Assimilation Research Testbed (DART). The data saved from this global, sim 1^circ resolution, 80 member ensemble span 2011–2019. They include ensembles of: sub-daily, real world, atmospheric forcing for use by all of the nonatmospheric models of CESM; weekly, CAM6, restart file sets; 6 hourly, prior hindcast estimates of the assimilated observations; 6 hourly, land model, plant growth variables, and 6 hourly, ensemble mean, gridded, atmospheric analyses. This data can be used for hindcast studies and data assimilation using component models of CESM; CAM6, CLM5, CICE5, POP2. MOM6, MOSART, and CISM; and non-CESM Earth system models. This large dataset (~ 120 Tb) has a unique combination of a large ensemble, high frequency, and multiyear time span, which provides opportunities for robust statistical analysis and use as a machine learning training dataset.

Hindcasts of Sea Surface Wind around the Korean Peninsula Using the WRF Model: Added Value Evaluation and Estimation of Extreme Wind Speeds

Sea surface wind plays an essential role in the simulating and predicting ocean phenomena. However, it is difficult to obtain accurate data with uniform spatiotemporal scale. A high-resolution (10 km) sea surface wind hindcast around the Korean Peninsula (KP) is presented using the weather research and forecasting model focusing on wind speed. The hindcast data for 39 years (1979–2017) are obtained by performing a three-dimensional variational analysis data assimilation, using ERA-Interim as initial and boundary conditions. To evaluate the added value of the hindcasts, the ASCAT-L2 satellite-based gridded data (DASCAT) is employed and regarded as “True” during 2008–2017. Hindcast and DASCAT data are verified using buoy observations from 1997–2017. The added value of the hindcast compared to ERA-Interim is evaluated using a modified Brier skill score method and analyzed for seasonality and wind intensity. Hindcast data primarily adds value to the coastal areas of the KP, particularly over the Yellow Sea in the summer, the East Sea in the winter, and the Korean Strait in all seasons. In case of strong winds (10–25 m·s−1), the hindcast performed better in the East Sea area. The estimation of extreme wind speeds is performed based on the added value and 50-year and 100-year return periods are estimated using a Weibull distribution. The results of this study can provide a reference dataset for climate perspective storm surge and wave simulation studies.

Comparative study on two deployment methods for large subsea spools

The demand for subsea spool deployment is increasing with the expansion of offshore projects. For a project to install multiple spools, different deployment methods can be used. The choice of method may influence the safety and the total cost of the project. Thus, it is important to evaluate different deployment methods in the planning phase. This study addresses weather window analysis of two deployment methods for large subsea spools. The purpose is to compare the efficiency of the two methods in terms of total installation time for projects with different numbers of spools. Numerical modeling and time-domain simulations of the critical activities are carried out. The simulations together with the operational criteria provide the allowable sea states, which are the key input for weather window analysis. Hindcast data from a site in the Barents Sea are used for weather window analysis. The total installation time is compared for various months, different total numbers of spools and transportation durations. The influence of the possible increase of the allowable sea states for the critical activity on the total installation time is also evaluated. Through the comparative studies, recommendations to select the proper deployment method for different situations are provided.

A semi‐empirical wind set‐up forecasting model for Lake Champlain

AbstractThe precision of Lake Champlain's water level estimation is a key component in the flood forecasting process for the Richelieu River. Hydrological models do not typically take into consideration the effects of the wind on the water level (also known as the wind set‐up). The objective of this study is to create an empirical wind set‐up forecast model for Lake Champlain during high wind events. The proposed model uses wind speed and direction across the Lake, as well as wind gusts as inputs. The model is calibrated to a subset of observations and evaluated on an independent sample, considering four wind speed bins. It is tested and compared to a variant of the Zuider Zee equation on 20 wind set‐up events that occurred between 2017 and 2019 using hindcast data from five different numerical weather prediction systems (GDPS, RDPS, HRDPS, NOAA and ECMWF). A quantile mapping‐based forecast calibration scheme is implemented for each of the forecast products to correct their biases. Results show that events are successfully predicted by the proposed model at least 72 h in advance. These results are better than the other comparative models found in the literature and tested herein. Overall, significant improvements are obtained by including wind speed and wind gusts from different weather stations.

Long-term spatio-temporal trends in extreme wave events in the Niger delta coastlines.

Wave climate in the marginal seas and estuaries provide baseline information, risk assessments and essential insight into coastal design, structure and engineering projects. This paper investigates the Spatio-temporal trends in extreme Significant Wave Height (SWH) in the Niger Delta coastlines. For the study domain, European Centre for Medium-Range Weather Forecasts (ECMWF) global atmospheric reanalysis ERA-Interim and wave hindcast data covering 37 years (January 1, 1980–December 31, 2016), derived from WAVEWATCH III (WW3) model were used in the analysis. The largest 99th percentile approach was used to analyze extreme SWH values. Results showed near neutral and negative trends over the region. The trends are generally insignificant for the 100-year return period. The sharp variation in extreme SWH between 1998 and 2000 resulted from a strong El Nino and La Nina phenomenon, which is associated with the weakening and strengthening of the West-African Monsoon.

On the selection of time-varying scenarios of wind and ocean waves: Methodologies and applications in the North Tyrrhenian Sea

This paper analyses a methodology to identify sub-series of waves parameters and wind speed able to explain the overall variability of the input dataset. To this end, the K-means clustering technique is applied to a 40-year long time series of hindcast data off the Genoa coastline (NW Italy). K-means aims to group the data in a reduced number of clusters, represented by as many modes of variability (or “model scenarios”). This work reviews and discusses a methodology to select time-varying model scenarios and assess the performances of K-means according to two indexes and increasing number of clusters. These indexes are used to compute the number of clusters best suited for the application at hand, testing different conditions as concerns the variables involved in the analysis and their temporal resolution. Results show that the indexes may not be consistent with each other, and that the number of scenarios to be reasonably employed strongly depends on how data are initially assembled. Finally, some of the model scenarios selected in front of Genoa are analysed and discussed in the framework of the local wind-wave climatology.

Effect of environmental modelling and inspection strategy on the optimal design of floating wind turbines

In order to reduce design conservatism and consequently the cost of energy, appropriate and cost-optimal safety factors should be derived, in light of environmental load uncertainties and lifetime costs. In the present work, a linearized dynamic model has been used together with Monte Carlo simulations and a numerical design optimization procedure to evaluate the impact of the description of wind and wave loads on the fatigue reliability and optimal design of a 10 MW spar floating wind turbine. Trade-offs between design costs and inspection costs with different design fatigue factors (DFFs) have also been assessed. The analyses have been performed for a realistic wind park site, where an environmental model has been developed based on hindcast data. Considering stochastic turbulence intensity, wind-wave misalignment, wind directional distribution, and a two-peak wave spectrum reduced the long-term fatigue damage by approximately two-thirds along the fatigue-critical part of the support structure compared to the base model. Re-designing the tower and platform with the full environmental model resulted in 11% reduction in CAPEX. However, due to the applied design optimization procedure, consistent reliability levels were achieved along the tower length, which resulted in important system side effects for the total structural reliability. Trade-offs between CAPEX and OPEX were derived based on a probabilistic fracture mechanics model and reliability updating through inspections. The necessary inspection intervals to achieve the same accumulated reliability after 20 years of operation were identified with different DFFs, and cost-optimal safety factors were computed with different OPEX costs and interest rates.

Analysis of the New Zealand’s Taranaki regional wave climate using high-resolution modelling

Understanding the regional wave climate is essential for engineering applications. The last two decades have not been included in assessments of the wave climate of New Zealand’s Taranaki region, where the country exploits offshore oil and gas. To make up for this lack of assessment, we carried out a high-resolution long-term (1979–2018) wave hindcast and examined several aspects of the climate in the region. The hindcast data validation against buoy and satellite data showed good agreement and suitability for wave climate analysis. Seasonal wave distributions reveal a bi-modal pattern, composed of (1) more energetic and long period (13–15 s) westerly–southwesterly swells; and (2) shorter-period (∼8 s) southwesterly wind-seas. These signatures result from the region’s high exposure to swells generated by persistent strong winds blowing over long fetches in the Southern Ocean, and to local storm-associated wind-seas, respectively. The largest waves are found offshore, with mean Hs value reaching 2.83 m, 90th percentile 4.3 m, 99th percentile 6.1 m, and maximum 10.8 m. Storm wave monthly climatologies show that storm peaks are largest in the austral autumn, especially in May, while the number of events is the largest in July. Trends in Hs statistics reveal an increase over the past 40 years, with higher rates at higher percentiles. Storm peaks have also increased, by up to 8 cm/decade, whereas the number of storm events has decreased. In agreement with previous work, we found relationships between Hs and climate patterns. Waves get larger in Taranaki waters during El Nino events, as a result of stronger southwesterly winds, and during negative Antarctic Oscillation phases, as westerly winds displace northward.

How deep ocean-land coupling controls the generation of secondary microseism Love waves

Wind driven ocean wave-wave interactions produce continuous Earth vibrations at the seafloor called secondary microseisms. While the origin of associated Rayleigh waves is well understood, there is currently no quantified explanation for the existence of Love waves in the most energetic region of the microseism spectrum (3–10 s). Here, using terrestrial seismic arrays and 3D synthetic acoustic-elastic simulations combined with ocean wave hindcast data, we demonstrate that, observed from land, our general understanding of Rayleigh and Love wave microseism sources is significantly impacted by 3D propagation path effects. We show that while Rayleigh to Love wave conversions occur along the microseism path, Love waves predominantly originate from steep subsurface geological interfaces and bathymetry, directly below the ocean source that couples to the solid Earth. We conclude that, in contrast to Rayleigh waves, microseism Love waves observed on land do not directly relate to the ocean wave climate but are significantly modulated by continental margin morphologies, with a first order effect from sedimentary basins. Hence, they yield rich spatio-temporal information about ocean-land coupling in deep water.

Assessment of APCC models fidelity in simulating the Northeast monsoon rainfall variability over Southern Peninsular India

The fidelity of the eight Asia-Pacific Economic Cooperation (APEC) Climate Center (APCC) models in representing the inter-annual variability and decadal shift in the northeast monsoon (NEM; October–December) rainfall over Southern Peninsular India (SPI) is evaluated. The hindcast data is used for the period of 28 years from 1983 to 2010 based on September initial conditions. The observations showed a clear inter-annual and inter-decadal variability of NEM rainfall during the study period. The analysis suggests that most of the models exhibited poor skill in representing the inter-annual variability. Only APCC model rainfall is in phase with observed SPI rainfall variations on the inter-annual time scale. It is noticed from the observed NEM rainfall time series that the period 1990–1999 (first decade) displays an above-normal rainfall and the period 2000–2010 (second decade) displays a below normal rainfall over the SPI region. It is also evident from the observations that NEM rainfall for most of the years displayed negative anomalies in the second decade including El Nino and La Nina years, while in the first decade positive anomalies are noted, suggesting the presence of decadal variability in the NEM. Rainfall variations in most of the coupled models are in phase with the observations during the second decade but are out of phase during the first decade. As evidenced from the observations that the intensified deep convection over the Indo-western Pacific region results in too far southward movement of the intertropical convergence zone (ITCZ) during the second decade. The southward shift in the strong upper-level divergence associated with lower-level convergence over the south Indian Ocean caused negative rainfall anomalies over the SPI in this decade. Further, the difference between the second and first decade demonstrates that an anomalous anticyclonic circulation over the Indian subcontinent is accountable for the vigorous dry northerly flow towards the SPI region and the resultant decadal shift in the rainfall pattern. Though the southward shift in the rainfall and large-scale circulation patterns are mildly captured by some models from decade to decade, most of the models completely misrepresented it. This study suggests that the coupled models displayed a very limited skill not only in capturing the inter-annual variability but also in representing the decadal variability of NEM rainfall.

A method for the probabilistic assessment of the on-board comfort on a passenger vessel route

The intense maritime traffic of modern days requires that a considerable attention be paid toward the comfort of ships not only as design criteria, but also as a means of assessing the quality of navigation in service conditions. Certain passenger routes and ship dimensions are more vulnerable than others to rough weather and the current climate changes require adaptation of design and navigation warning strategies.The main aim of the paper is to describe a method to derive a probabilistic representation of the comfort of passengers and crew on a selected route. AIS data is used to identify and characterise the chosen route. The seakeeping properties of the principal ships are evaluated employing a strip theory code. Detailed wave hindcast data is then employed to build state-of-the-art wave scatter diagrams along the route which allow the calculation of the ship responses for each sea state and the associated probability. The Motion Sickness Incidence is finally used as a comparative measure of the overall on-board comfort experienced on a variety of passenger vessels.

Assessment of operational limits: Effects of uncertainties in sea state description

For operations dominated by waves, operational limits are normally expressed in terms of allowable sea state parameters, such as the significant wave height (Hs) and spectral peak period (Tp). The allowable sea states need to be assessed in the planning phase of the operations. Different sources of uncertainties (including weather forecasts, wave spectral model and numerical models) should be accounted for in the allowable sea states to provide safety margins. This study focuses on assessment of the operational limits (in terms of Hs and Tp) considering the uncertainties associated with the sea state descriptions used in the numerical analysis. The aim is to demonstrate the effects of simultaneous description of wind sea and swell sea accounting for their different directional spreading and directions of propagation. A case study using lifting operation is chosen to address these uncertainties in the allowable sea states. To illustrate the results, two types of floating crane vessels, mono-hull and semi-submersible are employed in the case study. Operability analysis by using these vessels are performed and compared using NORA10 hindcast data for a site in the Barents Sea.

Identification of heatwave hotspots in Seoul using high-resolution population mobility data

This study proposes the methodology to identify heatwave hotspots in Seoul, the metropolis of Korea, using high-resolution data. Resident credit data, population mobility data, and temperature observation data are analyzed to determine vulnerable regions to heatwaves. Potentially vulnerable regions are derived in two ways: static vulnerable regions (SVRs) and dynamic vulnerable regions (DVRs), depending on their characteristics. SVRs are determined by low-income (lower 20% income quantile) residential areas fixed on time. In contrast, DVRs vary with the time and day of the week. DVRs are defined by the place less responsive to heatwaves, where are with low population variability and low correlation with temperature. The final vulnerable regions, so-called hotspots, are determined by the high temperature predicted area where the SVRs and DVRs intersect. We examine how to remove commuting-related displacement signals, which are represented as noise when analyzing population mobility. An example of the hotspots identification result is also shown using temperature hindcast data generated by the Korean Meteorological Administration short-range forecast system. Applying the vulnerability information can improve the quality of disaster planning and decision-making by highlighting the time and area of need for resources in the implementation of short- and long-term disaster response.

Comparative Analysis of Environmental Contour Approaches to Estimating Extreme Waves for Offshore Installations for the Baltic Sea and the North Sea

At the stage of design load analysis for offshore installations such as wave energy systems, a critical step is the determination of environmental cases to be evaluated for the definition of the characteristic design load. A commonly used methodology for load case selection, applied in multiple studies and recommended by technical specifications and guidelines, is the environmental contour approach. Here, 50-year environmental contours were generated for four study sites located in the North Sea, Skagerrak and the Baltic Sea by considering both observations and hindcast (model) data. For the construction of the contours, the well-established inverse first-order method (I-FORM) and a modified version using principal component analysis (PCA) were both examined. Furthermore, a 2-dimensional peaks-over-threshold (2D POT) method was evaluated. It was found that a version of the regular I-FORM was able to produce satisfactory contours which properly accounted for the highest waves. When using PCA, the dependency in the data was not properly captured by the probability functions under consideration. The 2D POT method, where applicable, was found to underestimate the extreme sea states. Comparisons between contours obtained from observations and hindcast data showed that the contours may differ substantially depending on the site and method, and thus care must be exercised when using hindcast data for such purposes.

ROMS Based Hydrodynamic Modelling Focusing on the Belgian Part of the Southern North Sea

The North Sea is a shallow sea that forms a complex physical system. The nonlinear interaction of the astronomical tides, varying wind fields and varying pressure systems requires appropriate approaches to be described accurately. An application based on the advanced numerical model Regional Ocean Modeling System (ROMS) was newly developed by the authors, tailored to simulate these hydrodynamic processes in the North Sea and the Belgian Continental Shelf, which is the area of particular interest in the present study. The purpose of this work is to develop and validate a state-of-the-art three-dimensional numerical model to form the basis of a compound operational and forecasting tool for the Belgian coastal zone. The model was validated with respect to water levels and temperature. Validation for astronomical tides was accomplished through the comparison of the principal constituents between the model results and observations at a number of tidal gauges in Belgium and other countries. A statistical analysis of the results showed that the model behaves as expected throughout the North Sea. The model response to the varying meteorological conditions was also validated using hindcast data for 2011 as input. In this case, the comparison between observed and modelled water levels showed a good agreement with average RMSE in Belgium 9.5 cm. Overall, the added value of this work is the development of an independent model for validation and comparison with other models and which can be used as an efficient tool for operational and forecasting purposes.