Abstract
Bad weather and rough seas continue to be a major cause for ship losses and is thus a significant contributor to the risk to maritime transportation. This stresses the importance of taking severe sea state conditions adequately into account in ship design and operation. Hence, there is a need for appropriate stochastic models describing the variability of sea states, taking into account long-term trends related to climate change. Various stochastic models of significant wave height are reported in the literature, but most are based on point measurements without considering spatial variations. As far as the authors are aware, no model of significant wave height to date exploits the flexible framework of Bayesian hierarchical space-time models. This framework allows modelling of complex dependence structures in space and time and incorporation of physical features and prior knowledge, yet at the same time remains intuitive and easily interpreted. This paper presents a Bayesian hierarchical space-time model for significant wave height. The model has been fitted by significant wave height data for an area in the North Atlantic ocean. The different components of the model will be outlined, and the results from applying the model to monthly and daily data will be discussed. Different model alternatives have been tried and long-term trends in the data have been identified for all model alternatives. Overall, these trends are in reasonable agreement and also agree fairly well with previous studies. Furthermore, a discussion of possible extensions to the model, e.g. incorporating regression terms with relevant meteorological data will be presented.
Similar content being viewed by others
Notes
Data available from URL: http://data-portal.ecmwf.int/data/d/era40_daily/.
Private communication with Dr. Andreas Sterl, KNMI.
The following parametrization of the inverse gamma distribution will be used:
$$ X \sim IG(\alpha, \beta) \Rightarrow f(x) = \frac{\beta^\alpha}{\Upgamma(\alpha)}\left(\frac{1}{x}\right)^{\alpha + 1}e^{-\beta / x} \quad \,\hbox {for } x > 0 $$Note that the credible bands in Vanem et al. (2011) were wrongly calculated, but not the mean.
References
Aggoun L (2002) Kalman filtering of a space-time markov random field. Math Comput Model 36:1193–1209
Akaike H (1974) A new look at the statistical model identification. IEEE Trans Automat Control 19:716–723
Babanin AV, Chalikov D, Young I, Savelyev I (2010) Numerical and laboratory investigation of breaking of steep two-dimensional waves in deep water. J Fluid Mech 644:433–463
Banerjee S, Carlin BP, Gelfand AE (2004) Hierarchical modeling and analysis for spatial data. Chapman & Hall, Boca Raton
Besag J, Kooperberg C (1995) On conditional and intrinsic autoregression. Biometrika 82:733–746
Bitner-Gregersen EM, Hagen Ø (1990) Uncertainties in data for the offshore environment. Struct Saf 7:11–34
Bitner-Gregersen EM, de Valk C (2008) Quality control issues in estimating wave climate from hindcast and satellite data. In: Proceedings of the 27th international conference on offshore mechanics and arctic engineering (OMAE 2008). American Society of Mechanical Engineers (ASME), New York, USA
Boer GJ, Flato G, Reader MC, Ramsden D (2000) A transient climate change simulation with greenhouse gas and aerosol forcing: experimental design and comparison with the instrumental record for the twentieth century. Clim Dyn 16:405–425
Caires S, Sterl A (2005a) 100-year return value estimates for ocean wind speed and significant wave height from the ERA-40 data. J Clim 18:1032–1048
Caires S, Sterl A (2005b) A new nonparametric method to correct model data: application to significant wave height from ERA-40 re-analysis. J Atmos Ocean Technol 22:443–459
Caires S, Swail V (2004) Global wave climate trend and variability analysis. In: Preprints of 8th international workshop on wave hindcasting and forecasting
Caires S, Swail VR, Wang XL (2006) Projection and analysis of extreme wave climate. J Clim 19:5581–5605
Cressie N, Wikle CK (2011) Statistics for spatio-temporal data. Wiley, Chichester
Debernard JB, Røed LP (2008) Future wind, wave and storm surge climate in the Northern Seas: a revisit. Tellus 60:427–438
Debernard J, Sætra Ø, Røed LP (2002) Future wind, wave and storm surge climate in the northern North Atlantic. Clim Res 23:39–49
Gelman A (2006) Prior distributions for variance parameters in hierarchical models. Bayesian Anal 1:515–533
Grabemann I, Weisse R (2008) Climate change impact on extreme wave conditions in the North Sea: an ensemble study. Ocean Dyn 58:199–212
Guedes Soares C, Bitner-Gregersen E, Antão P (2001) Analysis of the frequency of ship accidents under severe North Atlantic weather conditions. In: Proceedings of design and operations in abnormal conditions II. Royal Institution of Naval Architects (RINA), London
IPCC (2007a) Climate change 2007: synthesis report. Tech rep, Intergovernmental Panel on Climate Change
IPCC (2007b) Climate change 2007: the physical sciences basis. In: Contribution of working group I to the fourth assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge
Kalman R (1960) A new approach to linear filtering and prediction problems. J Basic Eng 82:35–45
Kamphuis JW (2010) Introduction to coastal engineering and management, 2nd edn. World Scientific Publishing Co, Singapore
Konig T, Lehner S, Schulz-Stellenfleth J (2007) Global analysis of a 2 year ERS-2 wavemode dataset over the oceans. In: Proceedings of geoscience and remote sensing symposium 2007 (IGARSS 2007). IEEE International
Kushnir Y, Cardone V, Greenwood J, Cane M (1997) The recent increase in North Atlantic wave heights. J Clim 10:2107–2113
Lim J, Wang X, Sherman M (2007) An adjustment for edge effects using an augmented neighborhood model in the spatial auto-logistic model. Comput Stat Data Anal 51:3679–3688
Meinhold RJ, Singpurwalla ND (1983) Understanding the Kalman filter. Am Stat 37:123–127
Nakićenović N, Alcamo J, Davis G, de Vries B, Fenhann J, Gaffin S, Gregory K, Grügler A, Jung TY, Kram T, La Rovere EL, Michaelis L, Mori S, Morita T, Pepper W, Pitcher H, Price L, Riahi K, Roehrl A, Rogner HH, Sankovski A, Schlesinger M, Shukla P, Smith S, Swart R, van Rooijen S, Victor N, Dadi Z (2000) Emissions scenarios. Cambridge University Press, Cambridge
Natvig B, Tvete IF (2007) Bayesian hierarchical space-time modeling of earthquake data. Methodol Comput Appl Probab 9:89–114
Reeve D, Chen Y, Pan S, Magar V, Simmonds D, Zacharioudaki A (2011) An investigation of the impacts of climate change on wave energy generation: the wave Hub, Cornwall, UK. Renew Energy 36:2404–2413
Richardson K, Steffen W, Schellnhuber HJ, Alcamo J, Barker T, Kammen DM, Leemans R, Liverman D, Munasinghe M, Osman-Elasha B, Stern N, Wæver O (2009) International scientific congress climate change: global risks, challenges & decisions—synthesis report. Tech. rep. International Alliance of Research Universities
Robert CP, Casella G (2004) Monte Carlo Statistical Methods, 2nd edn. Springer, New York
Rue H, Martino S, Chopin N (2009) Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations. J R Stat Soc B 71:319–392
Schwarz G (1978) Estimating the dimension of a model. The Annal Stat 6:461–464
Smith RL (2001) Environmental statistics. University of North Carolina. http://www.stat.unc.edu/postscript/rs/envnotes.pdf. Accessed 21 Oct 2010
Spiegelhalter DJ, Best NG, Carlin BP, van der Linde A (2002) Bayesian measures of model complexity and fit. J R Stat Soc B 64:583–639
Sterl A, Caires S (2005) Climatology, variability and extrema of ocean waves: the web-based KNMI/ERA-40 wave atlas. Int J Climatol 25:963–977
Toffoli A, Babanin A, Onorato M, Waseda T (2010) Maximum steepness of oceanic waves: Field and laboratory experiments. Geophys Res Lett 37:L05603. doi:10.1029/2009GL041771
Tvete IF, Natvig B (2002) A comparison of an analytical approach and a standard simulation approach in Bayesian forecasting applied to monthly data from insurance of companies. Methodol Comput Appl Probab 4:95–113
Uppala SM, Kållberg PW, Simmons AJ, Andrae U, DaCosta Bechtold V, Fiorino M, Gibson JK, Haseler J, Hernandez A, Kelly GA, Li X, Onogi K, Saarinen S, Sokka N, Allan RP, Andersson E, Arpe K, Balmaseda MA, Beljaars ACM, Vande Berg L, Bidlot J, Bormann N, Caires S, Chevallier F, Dethof A, Dragosavac M, Fisher M, Fuentes M, Hagemann S, HólmBJ E Hoskins, Isaksen L, Janssen PAEM, Jenne R, McNally AP, Mahfouf JF, Morcrette JJ, Rayner NA, Saunders RW, Simon P, Sterl A, Trenberth KE, Untch A, Vasiljevic D, Vitebro P, Woolen J (2005) The ERA-40 re-analysis. Q J R Meteorol Soc 131:2961–3012
Vanem E (2010) Stochastic models for long-term prediction of extreme waves: a literature survey. In: Proceedings of the 29th international conference on offshore mechanics and arctic engineering (OMAE 2010). American Society of Mechanical Engineers (ASME), New York, USA
Vanem E (2011) Long-term time-dependent stochastic modelling of extreme waves. Stoch Environ Res Risk Assess 25:185–209
Vanem E, Huseby AB, Natvig B (2011) A Bayesian-hierarchical space-time model for significant wave height data. In: Proceedings of 30th international conference on offshore mechanics and arctic engineering (OMAE 2011). American Society of Mechanical Engineers (ASME), New York, USA
Wang XJ, Zwiers FW, Swail VR (2004) North Atlantic ocean wave climate change scenarios for the twenty-first century. J Clim 17:2368–2383
Wang XL, Swail VR (2006a) Climate change signal and uncertainty in projections of ocean wave heights. Clim Dyn 26:109–126
Wang XL, Swail VR (2006b) Historical and possible future changes of wave heights in northern hemisphere oceans. In: Perrie W (ed) Atmosphere-ocean interactions, advances in fluid mechanics, vol 39, chap 8. WIT Press, pp 185–218
West M, Harrison J (1997) Bayesian forecasting and dynamic models, 2nd edn. Springer, New York
Wikle CK (2003) Hierarchical models in environmental science. Int Stat Rev 71:181–199
Wikle CK, Berliner LM, Cressie N (1998) Hierarchical Bayesian space-time models. Environ Ecol Stat 5:117–154
Wikle CK, Milliff RF, Nychka D, Berliner LM (2001) Spatiotemporal hierarchical Bayesian modeling: tropical ocean surface winds. J Am Stat Assoc 96:382–397
Acknowledgements
The authors want to express their thanks to Dr. Andreas Sterl at KNMI for kindly providing the data used in this analysis and for clarifying some issues discovered when investigating the data. The simulations for the 6-hourly data, which were very computational intensive and time consuming, were performed on the Titan Cluster, owned by the University of Oslo and the Norwegian metacenter for High Performance Computing (NOTUR), and operated by the Research Computing Services group at USIT, the University of Oslo IT-department.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
1.1 A Derivation of the full conditional distributions
In the following, the derivation of the full conditional distributions, needed for the Gibbs sampler, will be outlined. Vectors will be written in bold, and the notation \(P(V | \cdot )\) will be used to represent the conditional distribution of V conditioned on all other random quantities, i.e. the full conditional distribution. The values of the hyperparameters in Table 1 applies for all the specified prior distributions in the following and will not be referred in each case.
Spatial data will be treated as a vector of size X = 153 × 1, so that for example \({\user2{Z}}_{t}, {\varvec{\theta}}_{t}, {\varvec{\mu}}\) are all column vectors of size X. For the temporal components without spatial description, e.g. M t and T t , a vector of corresponding size may be obtained by multiplication by a vector of ones, \(\user2{1},\) of size X. In the following, e.g. M t is assumed to be interpreted as \(\mathbf{1} M_{t}\) in the instances where the \(\user2{1}\) is not explicitly written. The transpose of a matrix or vector \(\user2{V}\) will be denoted \({\user2{V}}^{\prime}.\)
The derivation of the full conditionals is similar to the derivation in Natvig and Tvete (2007), but whereas Natvig and Tvete (2007) sampled most parameters individually, here some sets of parameters are sampled jointly where this is natural. Presumably, this may speed up convergence of the Gibbs sampler. It is noted that for model alternatives 4 and 5, with only one temporal noise term, all relevant parameters c, d, γ and η were sampled jointly. The derivation of these joint full conditionals are not outlined here, but follow the same general approach.
1.2 A.1 Conditional distribution for \({\user2{H}}(t)\)
It is noted that since there are no random noise term in Eq. 2, the full conditional distribution of \({\user2{H}}_{t}\) reduces to just a single, fixed value, at each location, x and time, t:
Hence, determining \({\varvec{\mu}}, {\varvec{\theta}}_{t}, \hbox{T}_{t} \hbox { and } M_{t}\) determines \({\user2{H}}_{t} = H(x, t). \)
1.3 A.2 Conditional distribution for \(\varvec{\mu}\)
From the model specification, the following conditional distributions are obtained directly for all the parameters with dependence with \(\varvec{\mu}\)
where the X × X precision matrix \({\user2{A}}_{\mu}\) has elements (see e.g. Smith 2001)
\(\user2{I}\) is the identity matrix and \(\user2{1}\) is a X × 1 vector of ones. It is noted that \({\user2{A}}_{\mu}\) must be positive definite, and sufficient conditions are that both a ϕ and a λ are positive and that \(a_{\phi} + a_{\lambda}\le \frac{1}{2}\) (see e.g. Besag and Kooperberg 1995, p. 734). Similar constraints were used in Natvig and Tvete (2007).
Now, conditioned on the underlying processes, the observations are independent, and the full conditional distribution for \(\varvec{\mu}\) is proportional to a product of the known conditional distributions above
which gives
1.4 A.3 Conditional distribution for \(\varvec{\theta}(t)\)
\(P(\varvec{\theta}_t | \cdot )\) for \(t = 1, \ldots, T-1: \) To derive the conditional distribution of \(\varvec{\theta}_t\) for \(t = 1, \ldots, T-1, \) it is noted that the hierarchical model gives
where the matrix \(\user2{B}\) is an X × X matrix with elements
Now, the full conditional is obtained in a similar way as in A.2 above, arriving at the following multinormal distribution, for \(t = 1, \ldots, T-1 \):
\(P(\varvec{\theta}_T | \cdot ):\) From the model specification:
and parallel to the above, the full conditional for \(\varvec{\theta}_T\) becomes
\(P(\varvec{\theta}_0 | \cdot ){:} \) The model specification and the specification of priors for θ(x, 0), for \(x = 1, \ldots, X\) give
which gives the full conditional for \({\varvec{\theta}}_0, \)
1.5 A.4 Conditional distribution for \({\varvec{\mu}}_0\)
In order to derive the conditional distribution for \({\varvec{\mu}}_0, \) first define the vector \({\varvec{\mu}}_{0L} = \left(\mu_{0,1}, \mu_{0,2}, \mu_{0,3}, \mu_{0,4}, \mu_{0,5}, \mu_{0,6} \right)^{\prime}. \) It is noted that since there are no noise terms in Eq. 4, \({\varvec{\mu}}_0\) is uniquely determined by \({\varvec{\mu}}_{0L}, \) so sampling \({\varvec{\mu}}_{0}\) is equivalent to sampling \({\varvec{\mu}}_{0L}. \) The deterministic relation \({\varvec{\mu}}_0 = {\user2{P}}{\varvec{\mu}}_{0L}, \) with \(\user2{P}\) the X × 6 trend design matrix with rows \(\left(1, m(x), n(x), m(x)^2, n(x)^2, m(x)n(x) \right) \) for \(x = 1, \ldots , X = 153, \) holds (analog to Natvig and Tvete 2007; Wikle et al. 1998) and the model and prior specifications give the conditional distributions
Hence, the full conditional for \({\varvec{\mu}}_0\) becomes
1.6 A.5 Conditional distribution for b 0, b N , b E , b S and b W
The specification of the model and prior distributions give directly, by defining the vector \({\user2{b}}= (b_0, b_N, b_E, b_S, b_W)^{\prime}, \)
Even though it might be easier, implementationwise, to sample the b-parameters individually, it is noted that sampling from the joint conditional distribution for all b-parameters may speed up convergence of the Gibbs sampler. Hence, these parameters will be sampled from the joint full conditional of \(\mathbf{b}{:} \)
In order to rewrite this as an expression of \(\mathbf{b}, \) we need to find an X × 5-matrix \(\mathbf{J}_{\theta_{t-1}}\) so that
Since, \({\user2{B}} {\varvec{\theta}}_{t-1} = b_0{\varvec{\theta}}_{t-1} + b_N{\varvec{\theta}}_{t-1}^N + b_E{\varvec{\theta}}_{t-1}^E + b_S{\varvec{\theta}}_{t-1}^S + b_W{\varvec{\theta}}_{t-i}^W, \) it is easily seen that, if \({\user2{J}}_{\theta_{t-1}}\) is the matrix with columns \({\varvec{\theta}}_{t-1}, {\varvec\theta}_{t-1}^N, \) \({\varvec{\theta}}_{t-1}^E, {\varvec\theta}_{t-1}^S\) and \({\varvec{\theta}}_{t-1}^W\) respectively, the equation above holds, and by defining the \({\user2{J}}_D\)-matrices, for D = N, E, S, W, to be the X × X matrices with ones along the diagonal corresponding to the b D coefficients in \({\user2{B}}\) and zeros elsewhere then \({\varvec{\theta}}_{t-1}^{D} = {\user2 {J}}_{D}{\varvec{\theta}}_{t-1}\) and this can be written as
where now each of the elements in the vector above is actually an X-dimensional standing vector, defining the rows in the X × 5 dimensional matrix. Hence, completely parallel to the derivations above, the full joint conditional for \(\user2{b}\) becomes
1.7 A.6 Conditional distribution for a ϕ and a λ
The following is given by the model specification
which gives the full conditional joint distribution of a ϕ and a λ
Now, due to the appearance of the determinant \(|{\user2 {A}}_\mu | , \) which contains a ϕ and a λ, this distribution is difficult to sample from. Hence, parallel to what was done in Wikle et al. (2001) and Natvig and Tvete (2007), a Metropolis-Hastings step is introduced to sample from this distribution. As proposal distribution for the Metropolis-Hastings step, the pseudo conditional distribution for \(\varvec{\mu}\) is introduced:
where now the μ(x i )’s, conditional on all other μ(x j ) are univariate normally distributed as specified by the model. Now, as above, one can obtain the following pseudo distribution \(Q(a_\phi, a_\lambda | \cdot), \) where we denote \(\Updelta\mu(x) = \mu(x)-\mu_0(x) {:} \)
When sampling a ϕ and a λ, the constraints on these to ensure positive definiteness of \({\user2 {A}}_\mu\) should be kept in mind, and a somewhat different procedure than in Natvig and Tvete (2007) will be adopted, i.e. samples are drawn jointly from a bivariate pseudo proposal distribution and if the conditions are fulfilled (both are positive and the sum not greater than 0.5) they are used in a Metropolis-Hastings step. It is observed that such proposals often have expectations outside the required area but since any proposal is valid, the pseudo-distribution may be adjusted by a shift that brings the expectation within the required area. Since samples are drawn simultaneously, no asymmetry is introduced by this, and this is one advantage by using the bivariate distribution instead of sampling the parameters sequentially as in Natvig and Tvete (2007).
A bivariate normal distribution can be written on the form, with \({\user2 {x}} = (x, y)\)
where symmetry requires that q xy = q yx . Now, comparing this expression with the expression for the bivariate pseudo-distribution for a ϕ and a λ above, it is seen that this has the following elements in its precision matrix \({\user2 {Q}}_{\phi, \lambda}: \)
The variances and the covariance is found directly by taking the inverse of this matrix, and the variances and correlation becomes.
The expectations may also be found giving the following expectations of the proposal distribution
This bivariate normal distribution can now be used as proposal in the Metropolis-Hastings step, which may be adjusted if the expectation of a ϕ and a λ is outside the area determined by the positive definiteness requirements.
1.8 A.7 Conditional distribution for M t
From the model specification, the following distributions are given for each \(t = 1, \ldots, T\)
and the full conditional becomes, for all t, the following univariate normal distribution
1.9 A.8 Joint conditional distribution for c and d
Defining \({\varvec{\xi}}_{cd} = \left[ \begin{array}{l} \xi_c \\ \xi_d \end{array} \right], {\varvec{\Upsigma}}_{cd} = \left[ \begin{array}{ll} \sigma_c^2 & 0 \\ 0 & \sigma_d^2 \end{array} \right], \) \({\varvec{\beta}}_{cd} = \left[ \begin{array}{l} c \\ d \end{array} \right]\) and \({\varvec{\omega}}_{t} = \left[ \begin{array}{l} \cos(\omega t) \\ \sin(\omega t) \end{array} \right]\) the model specification gives, for \(t = 1, \ldots , T, \)
and it is straightforward to derive the full conditional for (c, d) as the bivariate normal distribution
with \({\user2{Q}}_M = \left[ \begin{array}{ll} \frac{1}{\sigma_c^2} + \frac{1}{\sigma_m^2}\sum_t(\cos( \omega t))^2 & \frac{1}{\sigma_m^2}\sum_t\cos( \omega t) \sin( \omega t) \\ \frac{1}{\sigma_m^2}\sum_t \cos( \omega t) \sin( \omega t) & \frac{1}{\sigma_d^2} + \frac{1}{\sigma_m^2} \sum_t (\sin (\omega t) )^2 \end{array} \right]\)
1.10 A.9 Conditional distribution for T t
From the model specification, the following distributions are given for each \(t = 1, \ldots, T\)
and the full conditional for T becomes, for all t,
1.11 A.10 Joint conditional distribution for γ and η
Defining \({\varvec{\xi}}_{\gamma \eta} = \left[ \begin{array}{l} \xi_\gamma \\ \xi_\eta \end{array} \right], {\varvec{\Upsigma}}_{\gamma\eta} = \left[ \begin{array}{ll} \sigma_\gamma^2 & 0 \\ 0 & \sigma_\eta^2 \end{array} \right], {\varvec{\beta}}_{\gamma\eta} = \left[ \begin{array}{l} \gamma \\ \eta \end{array} \right]\) and \({\varvec{\tau}}_{t} = \left[ \begin{array}{c} t \\ t^2 \end{array} \right]\) the model specification gives
and the full conditional for (γ, η) becomes,
with precision matrix \({\user2{Q}}_{\rm{T}} = \left[ \begin{array}{ll} \frac{1}{\sigma_\gamma^2} + \frac{1}{\sigma_{\rm{T}}^2}\sum_t t^2 & \frac{1}{\sigma_{\rm{T}}^2}\sum_t t^3 \\ \frac{1}{\sigma_{\rm{T}}^2}\sum_t t^3 & \frac{1}{\sigma_\eta^2} + \frac{1}{\sigma_{\rm{T}}^2} \sum_t t^4 \end{array} \right] .\)
1.12 A.11 Conditional distribution for σ 2 Z
From the model and prior specification, the following are given
and parallel to the above, the full conditional distribution becomes another inverse gamma distribution with updated parameters. Denoting \({\varvec{\Updelta}} = \left({\user2{Z}}_t - {\varvec{\mu}} -{\varvec{\theta}}_t - {\user2{1}}(M_t + \hbox{T}_t)\right), \) gives
1.13 A.12 Conditional distribution for σ 2μ
Similarly, the model and prior specification gives
Hence,
1.14 A.13 Conditional distribution for σ 2θ
From the model and prior specification, the following is given for \(t = 1, \ldots, T\)
Hence,
1.15 A.14 Conditional distribution for σ 2 m
The following is specified by the model, for all \(t = 1, \ldots , T,\)
and thus,
1.16 A.15 Conditional distribution for σ 2T
Finally, the model and prior specification, gives the following for \(t = 1, \ldots, T\)
Hence, finally
Rights and permissions
About this article
Cite this article
Vanem, E., Huseby, A.B. & Natvig, B. A Bayesian hierarchical spatio-temporal model for significant wave height in the North Atlantic. Stoch Environ Res Risk Assess 26, 609–632 (2012). https://doi.org/10.1007/s00477-011-0522-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00477-011-0522-4