TR-209 TECHNICAL REPORT AN EVALUATION OF A COMPUTERIZED NUMERICAL WAVE PREDICTION MODEL FOR THE NORTH ATLANTIC OCEAN JULY 1970 NAVAL OCEANOGRAPHIC OFFICE WASHINGTON, D. C. 20390 a, 1? - OY Price 80 cents ABSTRACT Procedures used to evaluate a computerized numerical wave prediction program are described. Since the model input is the forecast wind, periods corresponding to known wave growth and decay at Ocean Stations ‘A’, ‘I’, ‘J’, and ‘K’ and at ARGUS IS- LAND Tower were selected to evaluate its response. Wave fore- casts for 12, 30, and 36-hour periods are compared statistically to Tucker meter and the ARGUS ISLAND wave staff measure- ments. Comparisons of results using both U. S. Weather Bureau and Fleet Numerical Weather Central wind fields as input data are shown. The evaluation indicates that the forecasts are with- in a reasonable degree of accuracy for forecast intervals up to 36 hours. This basic model which represents another step for- ward in the state-of-the-art is expected to offer a considerable improvement in wave forecasts during the next decade. DONALD C. BUNTING LIONEL I. MOSKOWITZ Ocean Dynamics Branch Exploratory Oceanography Division Research and Development Department FOREWORD The Navy and combat forces have had continuing requirements for improved forecasts of ocean surface waves. The needs expressed during World War II provided a strong motivating force which led to the development of wave forecast techniques which produced consistent results. Techniques developed by Sverdrup and Munk and published in 1947 as H. O. Publication No. 601 provided forecasts of significant wave heights and periods. Later adaptation of random processes to the study of ocean waves resulted in the spectral forecast technique described in H. O. Publication No. 603. A recent outgrowth of the spectral technique has been the development of numerical computer models to forecast the directional wave spectrum. This new capability offers an opportunity to develop shallow water wave forecast models and to automate ship routing techniques. The purpose of this study is to evaluate this recent application of numerical techniques to computerized wave forecasts. ce iliuinn | TT Wes gushes 4 fi iw Baap! se ieogan . Beet ie dic, lou: HRUM Ban apitinave’ Karat meet Th ni iz ac 2 no : : mie é edt fi at antes ong ; eee eo ez alee: ? see gee ania, ate tie ean oe is ‘ t a | TABLE OF CONTENTS Page INMRODUCIION Domenic Cre toute. an muir. rte re tace lg 1 MEO DIS Be sae SEN a REN Vis A vin Ghieyeed = 2 WAVEIRECORDS opus gciap it sl fon caki vole, Cota pe Bes cmattnmancen A DANTAP EMRE ree Ty etc Mt cee eet carte etemee oe 5 RESULTS AND DISCUSSION. ........- es Wl ceaha emt cy che 9 eu ate hs 5 CONCLUSIONS AND RECOMMENDATIONS.......-----++¢°% 8 BIBLIOGRAPHY «2 a a.- 2 6 es BEA oil alae es Ms i dh RR 6] APPENDIX - SYNOPTIC WEATHER CONDITIONS. ......----- A-1 TABLES . OBSERVED WEATHER AND WAVE HEIGHTS AT VARIOUS STATIONS . . 9 . NUMBER OF PROGNOSTIC WAVE - SPECTRA SAMPLES... .... 10 HAWAVIE RECORD! DATA’ SUMMARY) 0th 0 ts see se oe Wl . SIGNIFICANT WAVE HEIGHTS - MEASURED VS. PREDICTED ..... 12 . SCHEDULE OF FORECAST INTERVALS FOR VARIOUS SYNOPTIC TUNIS Se qao Lie kolo tberpoheiien re eno erate Space exec eReg doy ee 15 SURFACE WINDS - MEASURED VS, PREDICTED. ..........-- 16 STATISTICAL ANALYSIS SUMMARY - USWB INPUT. .....-..-.- 19 STATISTICAL ANALYSIS SUMMARY - FNWE INPUT. ........- 20 FIGURES Page - Map of North Atlantic Gridpoints, OWS Stations and Argus Island... 23 2. Comparison of Observed (Measured) and Predicted Wave Heights at Sichionr A (WSWBIWind: Input)it.t. fs tsen iets Messe. ls es eee eee 24 3. Comparison of Observed (Measured) and Predicted Wave Heights at Sienions lp Je IS (WISE Winellie) 6 oo obo 666 bone ob ooo 25 4. Comparison of Observed (Measured) and Predicted Wave Heights at Stotionuk (WSWBaWind! Input) iat. %. 1. tat tate eee scone es Se ae 26 5. Comparison of Observed (Measured) and Predicted Wave Heights at Ares [seme] (USING Wintel Tatty) o 6G ccocc one ocos ous 5 Pf 6. Comparison of Observed (Measured) and Predicted Wave Heights at StaionyAn(ENWEaWind"InpUr)o.. 2 20s ars, net mt oserer tues oe Gua 28 7. Comparison of Observed (Measured) and Predicted Wave Heights at Suetiond) (FNM Winch Instn) geatore Bene o 6 Geo oo Bee 6 6 ao 29 8. Comparison of Observed (Measured) and Predicted Wave Heights at Station We (EINWE Wind: Input)... oe «8s ¢ caer eels whee woe 30 9. Comparison of Observed (Measured) and Predicted Wave Heights at Argus) islandg(GINWiE Wind. Input) 2 5 sasicemcrs uses cpr eachecie ees © 31 10. Scatter Diagram for Wave Heights (USWB Input). ...........-. 32 11. Scatter Diagram for Wave Heights (FNWF Input). ...........-. 33 12. Scatter Diagram for Wind Directions (USWB Input). ........... 34 13. Scatter Diagram for Wind Directions (FNWF Input). .......... 35 14. Scatter Diagram for Wind Speeds (USWB Input). ..........2.- 36 15. Scatter Diagram for Wind Speeds (FNWF Input). ............- 37 16. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Gonrelations(StationvA)! : > spe-e-cecn dae is tice) eee eee Sun eee es 38 vi Page 17. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Gonrelationy(Station lis) wv. B ceo sits Bree sin, Pease ee tetern as, 39 18. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Correlation (Stations) soot oboesrarces desley AERP) Gotininebe? ¢ 40 19. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Gorrelation) (Station, Kit sac toide eae Ot ee Oa. 4) 20. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Gorrelationi(ArgusiIsland)! s.c..)yeste casement sae <7 Se Osea Rese. 42 21. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Correlation - 6-Hour Forecast Interval. ........-26+22+00- 43 22. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Correlation - 12-Hour Forecast Interval... 2... 1.2.22 ec ee eee 44 23. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Correlation - 18-Hour Forecast Interval... .... 2... 22 ee eee 45 24. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Correlation - 24-Hour Forecast Interval... 2... 2... ee ee eee 46 25. Wave Spectra - Observed (Measured) Versus Predicted - Greatest (Gorrelation)= 30-Hour Forecast) Inferval 205) 2.) a ss 47 26. Wave Spectra - Observed (Measured) Versus Predicted - Greatest Correlation - 36-Hour Forecast Interval... 1.1... 2 2 ee ee eee 48 27. Wave Spectra - Observed (Measured) Versus Predicted - Least (Getnralientciny (Suemiel/N) 6 6 6 4 6 6.4 b Glo 610 5.0 0.086 PEC RIN. te 49 28. Wave Spectra - Observed (Measured) Versus Predicted - Least Gofrelationd(Stationnl) mswaaweren aecces. 6.8 Blois AR ee ee Da i 50 29. Wave Spectra - Observed (Measured) Versus Predicted - Least Correlationn(Stationgs) tm sae yma aes wrdbetuRes. 5 Bese rire ea coe 51 FIGURES (CONTINUED) FIGURES (CONTINUED) 30. Wave Spectra - Observed (Measured) Versus Predicted - Least Gorrelationn(Statioma)l sc. t..icct eu-cas we. Shee SMe. etn een te 31. Wave Spectra - Observed (Measured) Versus Predicted - Least Garrelationn(Argustlsland)\aicck ce ws ots rahe neces tee ec lets ete 32. Wave Spectra - Observed (Measured) Versus Predicted - Least Gorrelationi—.6-hour korecash Intervals 62. sna ee ee ee ene 33. Wave Spectra - Observed (Measured) Versus Predicted - Least Correlation - 12-Hour Forecast Interval. ........:---2.--- 34. Wave Spectra - Observed (Measured) Versus Predicted - Least Correlation - 18-Hour Forecast Interval... ......-222cecee 35. Wave Spectra - Observed (Measured) Versus Predicted - Least Correlation - 24-Hour Forecast Interval... ......---2+-00-. 36. Wave Spectra - Observed (Measured) Versus Predicted - Least Correlation - 30-Hour Forecast Interval. ...........0-2+2.- 37. Wave Spectra - Observed (Measured) Versus Predicted - Least Gorrelation — 36=Hour Forecast: Intervall (m0) SY.) ee A-1. USWB Synoptic Surface Weather Chart, 6 December 1966, O000Z. . A-2. USWB Synoptic Surface Weather Chart, 28 February 1967, 0000Z. . A-3. USWB Synoptic Surface Weather Chart, 6 March 1967, O000Z ... . vill 56 57 58 SY A-3 A-4 A-5 INTRODUCTION In 1922, Lewis F. Richardson, an Englishman, published his book entitled Weather Prediction by Numerical Process. His manual procedures for numerical weather predicting were impractical, however, because of the length of time required to produce the desired results. Some twenty years later, the advent of high speed electronic data processors made numerical methods much more practical, After many trials and tribulations, somewhat reliable numerical weather predictions on a limited basis and for the upper atmosphere became available by the mid-1950's. During the past decade rapid advances have been made, both in numerical theory and in the sophistication of electronic data processing equipment. Presently, numerical weather predictions are routine over vast areas of the globe with constant improvements being made as researchers attack the various problems. Following a somewhat analogous pattern, the problem of ocean wave prediction has also been attacked with considerable success during the past two decades. Ocean wave predictions received serious con- sideration beginning in the early 1940's after certain relationships among wind speeds, wind directions, and fetch lengths on water surfaces had been empirically determined. During the 1950's prognostic wave charts, giving wave heights and directions over the oceans, were being produced manually at certain weather centers for ship routing and operations. It was soon realized, however, that the subjectivity and tediousness of these manual forecasts required a more practical method for obtaining them. By the early 1960's raw weather data from land and ship obser- vations were being fed into electronic computers to produce synoptic and prognostic charts of surface pressure. From these the winds over the oceans were computed and used as computer inputs to produce synoptic and prognostic charts of wave heights and directions. At first, these forecasts were limited in their application because they yielded only the wave heights and directions instead of the more sophisticated and realistic ocean wave directional spectra. The necessity for the spectral approach to wave forecasting soon became evident. The Naval Oceanographic Office was given the task of preparing a wave spectra climatology for the North Atlantic Ocean in 1961. It was immediately recognized that the immensity of this task required the use of electronic computers and numerical procedures. Accordingly, con- tracts were let with private research groups qualified and experienced in the use of computers and ocean wave fields to produce realistic wind fields over the oceans from synoptic weather data, and from these to compute wave spectral data with the proper growth, propagation, and decay. The Travelers Research Company developed a numerical technique to convert pressure fields to wind fields and New York University developed the numerical wave fore- casting model. The research performed under these contracts has been reported by Bunting (1966). From the initial effort leading to the preparation of a wave spectra clima- tology of the North Atlantic Ocean it was observed that the two basic require- ments for providing realistic ocean wave data were: (1) an accurate representation of the low level wind fields and (2) a spectral wave forecasting model which properly represented the wave growth, propagation, and decay with varying wind speeds, durations, and fetch length. This wave model which in its pres ent form uses the Pierson=Moskowitz spectrum has more recently been modified to include wave growth functions developed by Inoue. Undoubtedly, there will be many future improvements made in both these requirements as further experience is gained. Even as this report is being written, such changes are being included in the development of a model for the North Pacific Ocean. These include improvements in both the wind field analyses and the spectral wave forecasting models. The improvements in the wave model are essentially the inclusion of the Inoue wave growth functions and extensive refinements in the grid system. A report on the results of these changes will be made following the completion of the work. ‘METHOD Until the wave hindcasts for the North Atlantic Ocean were completed in 1965, to our knowledge, no attempt had been made to use wind analyses to specify ocean wave spectra over oceanic scales through the use of computers. These hindcasts which constitute a wave climatology were based on synoptic weather data continuously updated at six hourly intervals using ship wind reporis. Since evaluations indicated that these hindcasts yielded relatively satisfactory wave spectra (Moskowitz, 1967), the next step forward was to attempt forecasting wave spectra from prognostic sea-level pressure fields and computed winds from these fields. After considerable planning, it was decided to use the "real time" synoptic and prognostic meteorological data as supplied by the Fleet Numerical Weather Facility (FNWF) at Monterey, California and by the United States Weather Bureau (USWB) at Suitland, Maryland as the basis of a prognostic wave~spectra evaluation program. Although this represented a duplication of data, it provided a comparison between two different prognostic techniques as well as providing fill-ins for missing data in either of the two sets of data. The duplication procedure was found to be desirable in the operational test program since there were several occasions of missing data in one or the other sets. The "real time" operational test program which was started in the summer of 1966 (17 August 1966) continued until 11 March 1967, a total of almost seven months. Data transmissions of sea surface pressures and observed ship reports for six-hourly synoptic times and for six-hourly prognostic times were made by a dataphone link from Monterey, California and Suitland, Maryland to New York City. The transmitted data were used at New York University as the input for data processors to give six-hourly synoptic and prognostic directional wave spectra for periods up to 36 hours. Preliminary detailed reports have been given by Moskowitz (1966, 1967). For each six-hourly synoptic time, a first-guess wind field analysis was made based on either the FNWF or the USWB meterological surface pressure field. In both sets of data the "surface" wind was taken to be the wind at the 19.5 meter level since the wave forecasting model was formulated from the recorded wind observations of the Ocean Weather Ships (OWS) which have anemometers at this elevation. The first-guess wind fields were then modified by using all the available ship wind measurements corrected to 19.5 meters, so that the wind analyses could be made as accurately as possible. The shipboard wind estimates were used when measurements were not available. The data, however, were weighted according to a predetermined priority scheme. A relatively small error in the wind speed can make fairly large errors in the forecasted wave spectra, hence great care must be taken in constructing the wind field. For the prognostic analyses, of course, no ship observations were used. The wave spectra output began with the computations of several days of hindcasts up to the 0000Z analysis for the day on which the prognostic spec- tra were to be made. This up-dating procedure was found to be essential in order to allow the model time to build up realistic wave spectra. The six- hourly prognostic data were made on a daily basis beginning at 0000Z for a total of 30 hours or for five separate forecasts for the FNWF input and for 36 hours or six separate forecasts for the USWB input. Each wave spectrum included the following: (1) date, (2) time, (3) gridpoint number, (4) wind speed, (5) wind direction, (6) 180 spectral components in 30-degree intervals of the compass and 15 intervals of frequency, (7) the sums of each of the 15 frequency intervals for all directions, and (8) the computed significant wave height. Spectra were computed for all 519 gridpoints over the North Atlantic Ocean at locations shown in Figure 1. The gridpoint numbers and locations for which complete spectral data were printed out at each forecast time were as follows: number 20, corres- ponding most nearly to the location of OWS A; number 36, corresponding to OWS |; number 72, OWS J; number 142, OWS K; and numbers 229, 230, 259, and 260 corresponding to the four nearest gridpoints to Argus Island (located approximately 25 miles southwest of Bermuda). The locations of all these gridpoints, together with the locations of the OWS stations and Argus Island, used in the evaluation, are also shown in Figure 1. To evaluate the New York University wave-spectra data obtained by automated methods, wave meter records were acquired for four OWS locations, Weather Adviser at stations | and J, Weather Reporter at stations J and K, and FRANCE II at station A and K, plus the wave staff records from Argus Island. The wave-spectra data chosen for evaluation were those for three days in December 1966, for two days in February 1967, and for three days in March 1967. These three periods of data were chosen because an examina- tion of the synoptic surface weather charts showed considerable storm activity which should be representative of adverse wave conditions in the eastern North Atlantic. A description of the synoptic weather conditions for each of the periods is given in the Appendix. Figures A-1, A-2, and A-3 of the Appendix show the surface synoptic weather charts as analyzed by the National Meteorological Center for 0000Z 6 December 1966, O0000Z 28 February 1967, and 0000Z 6 March 1967, which correspond to the times of maximum wave heights during each of the three periods of data. Table I shows the observed weather and the wave heights at the various OWS locations and at Argus Island. WAVE RECORDS Each of the wave meter records and the Argus Island wave staff records were digitized manually to a time series which contained 800 to 1440 points at time intervals of one or one and a half seconds depending on the scale of the wave record. The manually read amplitudes were punched on cards and computer processed to yield wave spectra containing 60 to 90 lag points over the frequency range from zero to 3.14 radians per second. The method followed for obtaining the spectral estimates was the same as that followed by Moskowitz, Pierson, and Mehr (1962). The computed spectral estimates from the wave records were next plotted on graph paper for lag numbers from 4 to 32, corresponding to radian frequencies from 0.14 to 1.1117 radians per second. The units of energy density were shown in feet squared-seconds. The time-corresponding forecasted wave spectra of New York University were then plotted in histogram form on the same graphs for lag numbers 6.5 to 29.5 for 14 different frequency bands. Thus, visual comparisons could be readily made between the wave-record estimated spectra and the computed prognostic spectra obtained from the meteorological data. Comparisons were also made of the wave-record and prognostic significant wave heights as well as the observed and computed prognostic wind directions and speeds. By these comparisons a reliable evaluation is believed to be possible. DATA Two sets of data were used: (1) the standard of comparison obtained from the wave records and meteorological observations of the weather ships and Argus Island; (2) the data to be evaluated which were generated on the New York University CDC 1604 computer in the form of two sets (USWB and FNWF) of prognostic wave spectra and wind conditions. The significant wave heights for both the standard and the prognostic data were computed from the respective wave spectra. Although the forecast wave spectra were given in the directional form to twelve 30-degree ranges, no use was made of these since the wave records were all one-dimensional allowing no comparisons of the directional spectra as prognosticated, Therefore, only the non-directional total energy versus frequency spectra were used in this evaluation. A total of 75 different wave records were used for making evaluation comparisons with 190 prognostic wave spectra at 6, 12, 18, 24, 30, and 36-hourly periods. Table I shows a breakdown of the number of prognostic wave spectra for both the USWB and FNWF inputs in each of the six-hourly intervals at the five locations where comparisons were made. Table III is a summary of the more important factors concerning the wave record data from the wave meters on the three weather ships and from the Argus Island wave staff. RESULTS AND DISCUSSION The results of the data evaluation are presented in graphic and tabular form. Figures 2 through 5 present graphic comparisons of the predicted and observed significant wave heights with time for the USWB input data at the five different locations from which wave observations were obtained. Figures 6 through 9 give the same comparisons of wave heights for the FNWF input data. Note that for both sets of data the observed significant wave heights are graphed but the 5 and 95% confidence levels are also shown for each value by the vertical line. Table IV presents in tabular form the significant wave heights at the various stations for the analyzed wave meter or wave staff data and the automated pre- dicted wave heights at the various corresponding gridpoints. For the wave meter or wave staff data, the columns headed "lower and upper limits" are the wave height values for the 5 and 95% confidence levels shown graphically as vertical lines through the values on Figures 2 through 9. The “long range" columns under the predicted wave heights are the predicted values for the 30 or 36-hourly forecasts at the times of 0600Z and 1200Z, respectively. Table V gives the forecast intervals for each of the 6-hourly times. Figures 10 and 11 show two scatter diagrams for wave height, one for the USWB input and the other for the FNWF input. These figures disclose the bias for the observed to predicted wave heights to be -1.6 feet and the RMS error +6.2 feet for 124 observations using the USWB input; for the FNWF input the bias and RMS error are slightly less, +0.4 feet and +5.1 feet, respectively, for 66 observations. Bias as used in this report is the average of the differences between the observed and predicted values. RMS values were computed by taking the square root of the average of the square of the differences between the observed and predicted values. Note that on Figures 10 and 11 the central diagonal line represents perfect correspondence between predicted and observed wave heights. The vertical distance from this line represents the plus or minus bias for each observation depending on whether the plot lies above or below the central diagonal. The diagonals labeled RMS show the positions on the graphs of the computed RMS errors for each figure. In addition to evaluating the wave height data, a similar statistical analysis was made of the wind directions and speeds. Table V1 shows the observed wind directions and speeds together with the machine predicted ' values for the same times at the same various forecast intervals as shown in Table V. It should be realized that there is a discrepancy between the locations of the observation stations and the gridpoint locations used in the evaluation as indicated on Figures 2 through 9. The effect of this discrepancy on both the wave heights and the winds was not determined in this evaluation. Scatter diagrams for the wind directions, using the two different inputs, are shown in Figures 12 and 13. Figures 14 and 15 are scatter diagrams for the wind speeds, using the two different inputs. For the wind directions, the USWB input produces slightly better results with a bias of +46 degrees and an RMS error of +75 degrees as against a bias of +52 degrees and an RMS error of +78 degrees for the FNWF input. A positive bias indicates that the predicted winds veered from the observed directions. The analysis of the wind speeds discloses very little difference between the two inputs. There is a bias of -5 knots and an RMS error of +10.1 knots for the USWB input as against a bias of -4 knots and an RMS error of +10.5 knots for the FNWF input. Tables VIi and VIII give a breakdown of the statistical analysis for each forecast interval and for each OWS location and Argus Island. In some examples, the number of observations is not large enough jo be siatistically valid, hence conclusions cannot be drawn from these tables for all categories. Table IX presents a side by side comparison of each station's statistical analysis for the USWB and FNWF inputs, together with various combinations and a final summa- tion of all the observations. The combined statistical analysis, showing the bias for significant wave heights to be -1.2 feet and the RMS error to +5.8 feet, comes within an acceptable value when consideration is given to the fact that a sampling variability in the average observed significant height of about 10% is built into the statistical analysis as noted by Pierson and Tick (1965). The concluding evaluation data are shown in Figures 16 to 37. The machine predicted frequency-amplitude wave-spectra histograms are super- imposed on the computed wave meter or wave staff spectra, as given by the equipment on the OWS station or at Argus Island. If there is good correlation between these two spectral curves, then certainly the predicted wave spectra are acceptable. In choosing samples for this report, the procedure was to select examples by visual inspection of: the greatest correlation for each station loca- tion without regard to the input or the forecast interval (Figures 16 to 20); the greatest correlation for each forecast interval without regard to the input or the station (Figures 21 to 26); and (Figures 27 to 31 and 32 to 37) the least correlation in each of the same two categories. In choosing the greatest and least correlation samples, the significant height correlation was considered with other factors being equal in order to select the most appropriate examples. The chosen representative set of samples has observed wave heights from 3.4 to 36.4 feet with the majority between 10 and 30 feet. The observed and predicted wave heights are shown on each figure as well as the date, time, forecast interval, and location. An inspection of Figures 16 to 20 shows that even the greatest correla- tion samples for station A and Argus Island could be improved considerably; whereas, the correlation samples for stations |, J, and K coincide quite closely. The poor correlations for station A and Argus Island may be the result of the station locations. Both stations are relatively near the boundaries of the gridpoint network with respect to the prevailing winds. The numerical procedures normally would tend to be least accurate in these regions since the propagational effects would not be so accurately projected there as at stations |, J, and K which have a much wider expanse of water in the western semicircle. Furthermore, at station A the location is such that the paths of the low pressure centers are to the south of or possibly very close to the station. The winds would be very much more irregular there than for the other three OWS locations so that a slight discrepancy in the predicted path and speed of the low center would result in large wave-spectra errors. Since the wind speeds at station A have by far the largest bias and RMS error compared with stations I, J, and K (-13 and +17 knots, respectively, versus -3 to -7 and 48.5 to 412.3 knots), the discrepancies in the wave spectra would seem to be caused by mis- calculations in the wind field rather than by any fault with the wave-spectra procedure. This points up the fact that the automated numerical forecasting of wave spectra can be no better than the capability of making accurate wind- field predictions. CONCLUSIONS AND RECOMMENDATIONS 1. The evaluation of the automated numerical wave-spectra prediction program shows that the procedure is feasible and that it produces results which are within a reasonable degree of accuracy with the present state of the art. 2. The evaluation indicates that there is need for improvement in pre- dicting surface wind fields over ocean areas and that the reliability of wave- spectra predictions closely follows the reliability of wind-field forecasts. 3. The evaluation discloses that there is no great difference in the accuracy of the prognoses of the two different inputs. If it were necessary to decide between the two, the evaluation suggests that the FNWF might be slightly superior, although this could be due to the maximum forecast interval being only 30 hours for the FNWF input as against 36 hours for the USWB input. 4. The evaluation reveals some remarkably good correlations between the predicted and the observed wave spectra. On the other hand, for station A and Argus Island the entire set of predictions could be considered as practically a "bust". Some reasons for this were listed in the preceding discussion section. 5. It is believed the evaluation gives ample proof that the automated numerical wave-spectra predictions would be valuable for operational use in most areas of the North Atlantic Ocean and that this procedure should now take precedence over the less sophisticated automated and manual methods. 6. As refinements in wind-field forecasting and in wave-spectra models become available, they can be readily adapted into the automated numerical procedures evaluated in the report, resulting in continuous future improvements in the wave-spectra predictions. ONKNRTNNNHNHOMYD BDAornwmnrnrnrowvonnny SLHO13H JAVM Lylol €6lol rade} v°SZzOl L°Szol ¥' 9201 9°¥20l vy ezol 0°020lL S*slol y'slol 0°0z0l > NOILVLS 1°6001 ev lol » NOILVIS aw JYNSSAYd 4S GNV1S!I SNDuv sin tI Ka] SGNIM AON — (se) Nndoonrwxonans NNNNN NMOANnNNOnNTMNM——O —N SLHO13H JAVM 8°S86 c°L66 ¥° 9001 6°S00L ¥° 886 ¥°t66 L-€001 0°9001 0°SOOL 9°y00l 6°2001 1° 866 f NOILVLS O°OLOL 9° 1001 7066 0°786 1°S86 £°086 6° 886 F NOILVES vy" LEO Vv Lol 0° LEOL 6° 10 Z°0€0L 8°6z0L aw JUNSSAed 24S > NOILVLS SNOILVLS SNOIIVA LV SLHOISH AVM CNV UaHLVaM daAddsSdO si yd SGNIM | a1aVvl 19 10 tO MOON NO AnRONDOtTNOOCSO . . nw _- owt 3 dd SIH 13H JAVM Vv NOILV1S S°LL6 cf — OSE €°626 6c — 050 V NOlILvIs 8° 8001 L°SOOL L°000t sly uid JaNSSAad SGNIM 34S | NOILVIs AVYH=9€ A1aH=-08 AVaH=72 AVdH=81 AVAH=Z L ATVSH=9 SATdWVS VULDAdS - JAVM DILSONDONd 4O YIGWNAN I dVavl “SI “dv 8 SdIHS 11V SdIHS 11V 10 JONVA AINANO Aes Hilddd LNAaWNYLSNI YOLOvV4 Viva WOddsd4d “940 JO “ON *XVW woqdaadd “94d 4O “ON “NIW WOdadddd “D4d JO “ON “AV os“ | I] JONVYS YdLyOddy YaSIAGV Y3LYOdaY YSHLVSM YSHIVSM YsSHIV4aM SMO SMO SMO NIVWWNS VLIVG GYODIY JAVM Hl J1aVL GvyOO4a ddd “S€O “ON “AV IVAYALN! AWIL FIdWYVS SNOILV NG qdvyOD34 SqddY¥OD34 YagWAN Silva NOILVILS 11 TABLE IV SIGNIFICANT WAVE HEIGHTS - MEASURED VERSUS PREDICTED GRIDPOINT 20 NOILVIS SMO" Gadnsvaw f NOILVLS SMO YASIAGYV YSHLVIM Il JDNV&d YSSIAGV YSHLVIM 17 (CE EFDS Ulta) JONVY ONO] INdNi 4dMN&I GaddS Id IndNt AMN&A q3adS tI Kal qdaadsS uid 94d JONVY ONO INd NI INd Ni NOILD3uIG aMsn aMsn GNIM 3.151d3ud (EVAR-E}S:[0) 622 LNIOdGldS GNV1SI SNODYV GaLDIGau¥d “SA G3AYISAO - SGNIM JDVAUNS DILdONAS (*LNOD) IA 318VL 8 1 4 H H H +H H L°S6F 0° €6F 0°S6# S“OOLF 8° ZF e°ZLLF 0° 19% G*OLt G 687 0°6SF L Let vl t 68F 8° 29F 4H ANNNIMNAN G2) Ga} Gece a) Se) So 4 4H +H “H +H NITNNMAMNA|S M-CAnn0N VAeor €°89F O°LyF o°9CF 6 8yF 1° 8S+ 8° ZOLF NOILVLS SMO. sdO 11V GNvi1sI snouVv As TIV “SUH 9E “SUH O€ “SUH $2 “SUH BL “SUH ZL "SUH 9 TV “SUH 9E “SUH Of “SUH ¥Z “SUH 81 “SUH ZL “SUH 9 Tv “SUH 9E “SUH OF “SUH $2 “SUH BL “SUH ZL “SUH 9 8°91F 1°S9F f° LF el- 8'0l- ral Tv Ove le 8° €8F LF Ol- LL+ Z'L- y “SUH 9€ €°22F ¥'98F ZF 0z= €z+ 8" LI- y “SUH 2 S*elF 6°S9F 6°6F 6s 9e- $°6- y “SUH ZL SLONY saau93d 1334 SIONX NI | 930 NI 1334 NI 0330S ‘ula 4H913H a33ds ‘IG 4H9I3H s#O NI ONIGNA NOlLv1s ONIM ONIM 3AM ONIM ONIM 3AVM Gaal WIL YOd SMO YOUN “S "WY SvIa Sd33dS GNIM ~ SNOILD3YIG GNIM - SLHDIZH JAVM LAdNI @MSN - AIVWWNS SISAIVNYV TVDILSILVLS HA JTaVi 19 9° YLF C 68F 8° L9F O° LEL+ v'9GF L£°68F epee ee ee €° 001+ CVLLF 0° £01F 8° 78F 9°G6F Z°2O\F $3dd930 1334 SLON» NI 1344 NI “yld 1H9I3H a33dS i LH9ISH NI ONIGNS HORREL °S °\Q) °e) $$$ = S@dadS GNIM - SNOILD3YIG GNIM - SLHDISH JAVM LAdNI SMNd - AYVWWNS SISATIVNV TVOILSILVIS INA d1aV 1 IWIL YOS “sdO 11V GNV1SI SNOaV SNOILVLS SMO 11V 768 al 4 NOILVLS SMO 20 ; p 70+ 4MN4 p : amsn sg0 TV B 0 4MNd aNy1sI , @msn SNOUV 0°6LF 4MNG 9 ALF 9 aMsn SdIHS T1V ¢"8F €°001F 4MNd LF 1°S6F SF ° aMsn y LAF 6°S9F €°GF 4MNd e'Z1F S*OLF Lilt aMsn c 0 LIF 0°99 0°OLF 4MN 8°91F 1°S9F LF aMsn Vv 930 ld NOIL534I0 LHO1GH GNIM JAVM NOIL53aIG GNIM LHO1ISH JAVM :[@) LAdNI NOILV1S SMO YOuds “S “W “8 Sv Id SGd4ddS GNIM - SNOILDIYIG GNIM - SLHDISH JAVM SNOSIYWdWOD SISATIVNV TVOILSILVLS XI aTaVvl 21 Ay hie * SESS aioe ssi] seen netav 1a ciilahaceaa ema 03 } 6M Wy = ScHDIse ave : Aoi ie a) Sve ty SuSINUS ¥ a ) “GNV1SI SNOYV GNV SNOILVLS SMO ’ SLNIOddIdS SILNVILV HLYON JO dVW e 0 ; 60 e : ; § “L ANSI 23 ) SIGNIFICANT WAVE HEIGHT, Hy,3 (FEET 40 35 30 N on ip’) (e) i) LEGEND @® AVERAGE A 36-HOURLY (1200Z) PREDICTION — OBSERVED WAVE HEIGHT ——- PREDICTED WAVE HEIGHT 5% AND 95% CONFIDENCE LEVELS l2 00 12 00 12 00 12 TIME 27 4 DATE FIGURE 2. COMPARISON OF OBSERVED (MEASURED) AND PREDICTED MARCH 1967 (ole) A ® 20: LOCATION OF A WITH RESPECT TO GRIDPOINT 20 WAVE HEIGHTS AT STATION A (USWB WIND INPUT) 24 SIGNIFICANT WAVE HEIGHT, H,,3 (FEET) 40 35 30 ins) oO iy) {o) a vac J ® LOCATION OF u WITH RESPECT TO GRIDPOINT 72 36 2 LOCATION OF I WITH RESPECT TO A GRIDPOINT 36 LEGEND @ AVERAGE A 36-HOURLY (I1200Z) PREDICTION — OBSERVED WAVE HEIGHT ——- PREDICTED WAVE HEIGHT K 8 5% AND 95 % CONFIDENCE 142 LEVELS TEE Coe EEES © 30-HOURLY (0600Z) PREDICTION GRIDPOINT i42 DECEMBER 1966 FEBRUARY 1967 MARCH I967 2 © 2 © BP G HM © PO P CO. BP oO iP TIME 4 5 6 27 28 4 5 6 DATE FIGURE 3. COMPARISON OF OBSERVED (MEASURED) AND PREDICTED ' WAVE HEIGHTS AT STATIONS I, J, K (USWB WIND INPUT). 25 40 SIGNIFICANT WAVE HEIGHT, H,,3 (FEET) . 35 30 i) o De) (eo) a LEGEND @® AVERAGE A 36-HOURLY (I200Z) PREDICTION — OBSERVED WAVE HEIGHT ——- PREDICTED WAVE HEIGHT 5% AND 95% CONFIDENCE LEVELS © 30-HOURLY(O600Z) PREDICTION FEBRUARY |I967 00 12 00 12 00 l2 (ojo) l2 TIME ail 4 DATE MARCH 1967 FIGURE 4. COMPARISON OF OBSERVED (MEASURED) AND PREDICTED WAVE HEIGHTS AT STATION K (USWB WIND INPUT). 26 SIGNIFICANT WAVE HEIGHT, Hy, (FEET) 40 Fi Simla el vir MEET @eal aT al 229: 0 ARGUS ®@ ISLAND 35 LEGEND LOCATION OF ARGUS ISLAND ® AVERAGE WITH RESPECT TO GRIDPOINT 229 A 36-HOURLY (1200Z) PREDICTION — OBSERVED WAVE HEIGHT -—- PREDICTED WAVE HEIGHT 30 5% AND 95% CONFIDENCE LEVELS © 30-HOURLY (O600Z) PREDICTION 25 20 15 (9) if A ~-—=-4-—— \ if 2) fi ~ X) 1 Y ra 10 + SS 2 I N DAN: Ne / by aN ® ) by a 2 s % DECEMBER I966 MARCH 1967 (e) 12 0O 12 00 12 00 0O 12 00 12 00 l2 0O {2 TIME is) 6 7 4 5 6 1 DATE FIGURE 5. COMPARISON OF OBSERVED (MEASURED) AND PREDICTED WAVE HEIGHTS AT ARGUS ISLAND (USWB WIND INPUT). 27 00Z SIGNIFICANT WAVE HEIGHT, Hi3 (FEET) 40 35 30 nN (3) ty (e) Go LEGEND ® AVERAGE — OBSERVED WAVE HEIGHT ——- PREDICTED WAVE HEIGHT 5% AND 95% CONFIDENCE LEVELS FEBRUARY 1967 MARCH 1967 TIME DATE FIGURE 6. COMPARISON OF OBSERVED (MEASURED) AND PREDICTED WAVE HEIGHTS AT STATION A (FNWF WIND INPUT). 28 SIGNIFICANT WAVE HEIGHT, Ay/3 (FEET) 40 35 30 25 N (o) Pie ae miele le healt ihr le leh teh otpt LEGEND ® AVERAGE © 30-HOURLY (O600Z) PREDICTION — OBSERVED WAVE HEIGHT ——- PREDICTED WAVE HEIGHT 5% AND 95% CONFIDENCE LEVELS FEBRUARY 1967 MARCH 1967 l2 00 l2 00 ‘(2 00 12 00 12 0Oo 12 00 12 TIME Bil 28 4 5 6 C4 DATE FIGURE 7. COMPARISON OF OBSERVED (MEASURED) AND PREDICTED WAVE HEIGHTS AT STATION J (FNWF WIND INPUT). 29 SIGNIFICANT WAVE HEIGHT, Hy; (FEET) 40 35 30 25 20 LEGEND ® AVERAGE © 30-HOURLY (O600Z) PREDICTION — OBSERVED WAVE HEIGHT ——~ PREDICTED WAVE HEIGHT 5% AND 95% CONFIDENCE LEVELS FEBRUARY 1967 MARCH 1967 12 00 12 oO l2 00 l2 oo 12 00 12 00 l2 TIME 27 28 4 3) 6 Te DATE FIGURE 8. COMPARISON OF OBSERVED (MEASURED) AND PREDICTED WAVE HEIGHTS AT STATION K (FNWF WIND INPUT). 30 00Z 40 35 30 25 20 SIGNIFICANT WAVE HEIGHT, H,,, (FEET) jefe) LEGEND ® AVERAGE ; © 30-HOURLY (O600Z) PREDICTION — OBSERVED WAVE HEIGHT --- PREDICTED WAVE HEIGHT 5% AND 95% CONFIDENCE LEVELS MARCH I967 l2 00 12 00 l2 0O 12 jefe) 12 (ofe) l2 00 12 TIME 4 5 6 1 DATE - FIGURE 9. COMPARISON OF OBSERVED (MEASURED) AND PREDICTED WAVE HEIGHTS AT ARGUS ISLAND (FNWF WIND INPUT). 3] os “(LAdNI 8MSN) SLHOISZH JAVM YOs WVYOVIG YALLVIS “Ol INNO! (1334) £"H ‘LHOSISH SAVM G3AYN3SEO Gb Ov ce o¢ G2 02 ans! Ol fe (0) ‘S| SNOYV 4 NOILVLS © f NOILVLS x T NOILVLS 9 V NOILVLS V QN3931 Ol S| fe) we H “LHSISH SAVM G3LdIGSud Ww (9) (ieee) ‘S80 vel 1429 + SWY 149 1—Ssvig oe Ge 32 OS “(LAdNI AMN&J) SLHOISH JAVM YOS WVYOVIG YILLVDS “LL ANOS Sv Si Snouv - 4 NOILVLS © f NOILVLS x V NOILVLS V QN3931 OV ce (1333) ®/'H ‘LHOI3H S3AVM G3AYaSEO O£ G2 02 S| Ol G ‘S80 99 4143 1°S + SWY 1470+ SVIG S wo (1334) *"H ‘LHOISH 3AVM G3L0IG aud O£ G¢ 33 PREDICTED WIND DIRECTION (DEG.) 200 BIAS + 46 DEG. RMS + 75.0 DEG. 160 105 OBS. 120 80 40 5 360 320 280 240 |-— 200 -— 160 120 LEGEND A STATION A © STATION I x STATION J © STATION K + ARGUS IS. 40 (0) 40 80 120 160 200 240 280 320 360 OBSERVED WIND DIRECTION ( DEG.) FIGURE 12. SCATTER DIAGRAM FOR WIND DIRECTIONS (USWB INPUT). 34 40 PREDICTED WIND DIRECTION ( DEG.) 200 160 120 80 40 360 320 280 240 200 160 120 80 5 40 BIAS + 52 DEG. RMS + 78.0 DEG. 62 OBS. i apl ie Bali ee ail ice T LEGEND A STATION A X STATION J © STATION K [A * ARGUS IS. (0) 40 80 120 160 200 ~ 240 280 320 360 40 OBSERVED WIND DIRECTION ( DEG.) FIGURE 13. SCATTER DIAGRAM FOR WIND DIRECTIONS (FNWF INPUT). 35 00! 06 ‘SI SNOYV 4 NOILVLS f NOILVLS I NOILVLS V NOILVLS Y QN3937 “(LAdUNI @MSN) SGJ3dS GNIM YOI WVYOVIG YALLVDS “vl ANNO! 08 OL (SLY) G33dS GNIM G3Au¥3S8O 09 oS Ov o¢ ‘Sd0 vil ‘SLY 1°01 + SWY ‘SL4 S —SVIS Ol (e) (e) rO nN (SLY) GA3dS GNIM G3L9IG3¥d fe) + fe) Te) 09 OL 00] *(LAdNI SMN4) SG94dS GNIM YOs WVEOVIG YALLVOS °SL FINO! (SLM) G33SdS GNIM G3AuSaSEO 06 08 OL 09 os Ov 0} 53 ‘S| SnouV > NOILVLS © f NOILVLS x V NOILVLS V GN3931 02 Ol ‘$d0 ¥9 ‘SLY S Ol + SNY ‘SLY v -— SVIE (S1H) GaadS GNIM G3191d3¥d 37 ENERGY DENSITY (ft2—sec.) 800 POWER SPECTRUM FORECAST: 12 HOUR DATE : 6 MARCH 1967 TIME : 1200Z LOCATION: STATION A 700 His OBS. 11.7" H13 PRED. 6.3’ 600 500 any fo) (o) 300 200 100 eS 4 8 2 6 20 - 24 28 32 LAG NUMBER FIGURE 16. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED— GREATEST CORRELATION (FNWF INPUT). 38 ENERGY DENSITY (ft.2 —sec ) 800 POWER SPECTRUM FORECAST: 18 HOUR DATE: 5 DEC. 1966 TIME : 1800Z LOCATION: STATION I 700 Hi,,3 OBS. 36.4 600 H 1,3 PRED. 33.6' 500 S fe) fo) W (e) fe) 200 100 4 8 l2 I6 20 24 28 32 LAG NUMBER FIGURE 17. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — GREATEST CORRELATION (USWB INPUT). 39 800 ) ~500 ENERGY DENSITY (ft*—sec 5 (e} Ol Oo (e) 200 100 POWER SPECTRUM 700 : ‘FORECAST: [2 HOUR DATE : 5 MARCH 1967 TIME : 1200Z LOCATION: STATION J 600 H1;3 OBS. 24.4' H\;3 PRED. 23.6' 4 8 l2 16 20 24 28 32 LAG NUMBER FIGURE 18. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — GREATEST CORRELATION (FNWF INPUT). 40 800 - POWER SPECTRUM FORECAST: 30 HOUR 700 DATE : 5 MARCH 1967 TIME : O600Z LOCATION: STATION K 600 Hj,3 OBS. 15.2 ~ 500 @ ” | N - (a & 400 4 LJ (2) > (&) a uJ ia 300 200 100 4 8 12 16 20 24 28 32 LAG NUMBER FIGURE 19. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — | GREATEST CORRELATION (USWB INPUT). 4] cS) ENERGY DENSITY (ft *—se 800 POWER SPECTRUM 700 FORECAST: 24 HOUR DATE: 5 DEC. 1966 TIME : OOOOZ LOCATION: ARGUS IS. H13 PRED. 12.4' rou (e) ©. 400 200 100 4 8 12 16 20 24 28 32 LAG NUMBER FIGURE 20. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — GREATEST CORRELATION (USWB INPUT). 42 ) ENERGY DENSITY (ft.2—sec. 800 700 600 ol fe) (e) S oO fo) (oN {e) (e) 200 100 l2 16 20 LAG NUMBER POWER SPECTRUM FORECAST: 6 HOUR DATE Seipe Ban S67 TIME : O600Z LOCATION: STATION J H,,3 OBS. 25.5' H,,3 PRED. 26.4 24 28 FIGURE 21. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — GREATEST CORRELATION - 6-HOUR FORECAST INTERVAL (FNWF INPUT). ie ENERGY DENSITY (ft.2—sec) Oo (e) (e) 800 POWER SPECTRUM 700 FORECAST: 12 HOUR DATE : 4 MARCH 1967 TIME = 1200Z LOCATION: STATION J H,,3 OBS. 22.9' Hyy3 PRED. 23.2’ (9) fe) oO 400 200 100 4 8 l2 16 20 24 28 32 LAG NUMBER FIGURE 22. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — GREATEST CORRELATION - 12-HOUR FORECAST INTERVAL (USWB INPUT). AA 500 ENERGY DENSITY (ft2—se 800 POWER SPECTRUM FORECAST: 18 HOUR DATE= 27 FEBS 1967, TIME : 1800Z LOCATION: STATION J 700 Hi,3 OBS. 20.8' H;3 PRED. 19.7° 600 bd (e) (e) 300 200 100 4 8 l2 16 20 24 28 SZ LAG NUMBER FIGURE 23. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — GREATEST CORRELATION - 18-HOUR FORECAST INTERVAL (FNWF INPUT). 45 ENERGY DENSITY ( ft.2—sec.) 800; POWER SPECTRUM 700 FORECAST: 24 HOUR DATE: 6 DEC. 1966 TIME =: OOOOZ LOCATION: STATION I H,,3 OBS. 35.7° His PRED. 34.8 600 500 S (e) (e) W fe) fe) 200 100 4 8 12 16 20 24 28 32 LAG NUMBER FIGURE 24. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED— GREATEST CORRELATION - 24-HOUR FORECAST INTERVAL (USWB INPUT). 46 c.) oO fe) (e) ENERGY DENSITY (ft.2—se 800 POWER SPECTRUM 700 FORECAST: 30 HOUR DATE: 6 DEC. 1966 TIME : O0600Z LOCATION: STATION LI - ry 1 H,;3 PRED. 32.1 400 200 100 4 8 12 16 20 24 28 32 LAG NUMBER FIGURE 25. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED— GREATEST CORRELATION - 30-HOUR FORECAST INTERVAL (USWB INPUT). 47 ENERGY DENSITY (ft.2— sec.) Oo (e) (e) 200 800 POWER SPECTRUM FORECAST: 36 HOUR DATE : 5 MARCH 1967 | TIME : 1200Z LOCATION: STATION K 700 H,,;3 OBS. 13.2 H,,;3 PRED. 15.8' 600 500 S fe) fo) 100 4 8 l2 I6 20 24 28 32 LAG NUMBER FIGURE 26. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED - GREATEST CORRELATION - 36-HOUR FORECAST INTERVAL (USWB INPUT). 48 ) ENERGY DENSITY (ft 2—sec Ww fo) (e) 800 POWER SPECTRUM FORECAST: 12 HOUR DATE : 5 MARCH 1967 TIME : OOOOZ LOCATION: STATION A 700 H,;3 OBS. 22.4' Hy3 PRED. 7.0° 600 500 BSS (e) (e) 200 100 4 8 l2 16 20 24 28 32 LAG NUMBER FIGURE 27. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — LEAST CORRELATION (USWB INPUT). 49 ) ENERGY DENSITY (ft*—sec 800 POWER SPECTRUM 700 FORECAST: 36 HOUR DATE: 5 DEC. 1966 TIME : 1200Z LOCATION: STATION I Ti ’ H,,3 OBS. 31.6 600 re Hj;3 PRED. 19.9' 500 S (e) (e) QA Oo (e) 200 100 4 8 12 16 20 24 28 32 LAG NUMBER FIGURE 28. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — LEAST CORRELATION (USWB INPUT). 50 c.) ENERGY DENSITY (ft.2—se oO fe) (e) 800 POWER SPECTRUM FORECAST: 24HOUR DATE : 6 MARCH 1967} TIME = OOO0Z LOCATION: STATION J 700 600 : : Hi3 OBS. 32.3' Hj,3 PRED. 13.6 oO fo) fe) S (e) (e) 200 100 4 8 12 I6 20 24 28 32 LAG NUMBER FIGURE 29. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED - LEAST CORRELATION (USWB INPUT). 51 ) ENERGY DENSITY (ft.2 —sec 800 POWER SPECTRUM 700 . ; _| FORECAST: 36 HOUR | DATE: 6 DEC. 1966 TIME : 1200Z LOCATION: STATION K 600 H,,3 OBS.6.3' H,,3 PRED. 13.3° 500 400 200 100 4 8 2 G 20 24 28 32 LAG NUMBER FIGURE 30. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — LEAST CORRELATION (USWB INPUT). 52 c.) ENERGY DENSITY ( ft.2 —se 800 POWER SPECTRUM FORECAST: 30 HOUR DATE : 6 MARCH 1967 TIME : O600Z LOCATION: ARGUS IS. 700 H 3 OBS. 3.4 600 i H,,3 PRED. 6.3 o fe) (e) 400 200 100 4 8 l2 16 20 24 28 32 LAG NUMBER FIGURE 31. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — LEAST CORRELATION (FNWF INPUT). 53 ENERGY DENSITY (ft. 2—sec.) Oo fe) (e) 800 700 600 500 oS oe) (e) 200 100 FIGURE 32. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — I6 20 LAG NUMBER POWER SPECTRUM FORECAST: 6 HOUR DATE : 4 MARCH 1967 TIME : O600Z LOCATION: ARGUS IS. Hj,3 OBS. 4.0’ H1,3 PRED. 8.2 LEAST CORRELATION - 6-HOUR FORECAST INTERVAL (FNWE INPUT). 54 32 800 700 600 ) o {e) (@) BSS fo) (eo) ENERGY DENSITY (ft2 —sec. w (e) (e) 200 FIGURE 33. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — LAG NUMBER POWER SPECTRUM FORECAST: 12 HOUR DATE : 5 MARCH 1967 TIME : 1200Z LOCATION: STATION A H,,3 OBS. 19.6' Hi3 PRED. 6.2' LEAST CORRELATION - 12-HOUR FORECAST INTERVAL (USWB INPUT). 55 ) 800 POWER SPECTRUM 700 FORECAST: 18 HOUR DATE : 5 MARCH 1967 TIME : 1800Z © LOCATION: STATION J ae t 600 H,,3 OBS. 30.3 H1,3 PRED.15.0' ~500 ENERGY DENSITY (ft*—sec 8 5 (@) Ww Oo (e) 200 100 4 8 l2 I6 20 24 28 32 LAG NUMBER FIGURE 34. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — LEAST CORRELATION - 18-HOUR FORECAST INTERVAL (USWB INPUT). 56 800 POWER SPECTRUM FORECAST: 24 HOUR DATE : 5 MARCH 1967 TIME : OOOOZ LOCATION: STATION A 700 H 3 OBS. 22.4' H13 PRED. 8.5° 600 ) o fe) ENERGY DENSITY (ft. 2 —sec. 5 (e) WwW (e) (eo) 200 100 LAG NUMBER FIGURE 35. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — LEAST CORRELATION - 24-HOUR FORECAST INTERVAL (FNWF INPUT). 57 800 POWER SPECTRUM 700 FORECAST: 30 HOUR DATE : 4 MARCH 1967 TIME : 0600Z LOCATION: STATION J H,,3 OBS. 275° Hy,3 PRED. 13.4’ 600 —500 (3) ® wo | N . ze = = 400 r-4 Lu (ea) > [&) ag WW 4 lJ ow (e) Oo 200 100 4 8 l2 16 20 24 28 32 LAG NUMBER FIGURE 36. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — LEAST CORRELATION - 30-HOUR FORECAST INTERVAL (USWB INPUT). 58 c.) ENERGY DENSITY (ft.*—se 800 POWER SPECTRUM 700 FORECAST: 36 HOUR DATE : 5 MARCH I967 TIME : 1200Z LOCATION: STATION A Hiy3 OBS. 19.6 600 * . | H 3 PRED. 4.0 o (e) (e) f fe) fe) (oN ‘e) oO 200 100 4 8 l2 16 20 24 28 32 LAG NUMBER FIGURE 37. WAVE SPECTRA - OBSERVED (MEASURED) VERSUS PREDICTED — LEAST CORRELATION - 36-HOUR FORECAST INTERVAL (USWB INPUT). 59 * ' , “had yt 1 Ae ae 0a BIBLIOGRAPHY Baer, L., 1962: An Experiment in Numerical Forecasting of Deep Water Ocean Waves. LMSC-801296, Contract NOnr 285 (03), Office of Naval Research. Lockheed Missile and Space Compciny, California. , 1963: Ocean Wave Forecasting Study. Final Report. Lockheed Report No. 17501 to College of Engineering, Research Division, New York University as a Subcontract Under U. $. Naval Oceanographic Office Contract N62306-1042 with New York University. Bunting, D. C., 1966: Wave Hindcast Project North Atlantic Ocean. Tech- nical Report TR-183, U.S. Naval Oceanographic Office, Washington, D.C. Moskowitz, L. |., 1966: The Feasibility of the Argus Island Tower as a Deep Water Observation Site for the Automated Wave Prediction Program at the U. S. Naval Oceanographic Office. Paper Presented at the Conference on Marine (Oceanic) Meteorology, September 7-9, 1966, Virginia Beach, Va. , 1967: Evaluation of Spectral Wave Hindcasts Using the Automated Wave Prediction Program of the Naval Oceanographic Office. Naval Oceanographic Office, Washington, D. C. Informal Report No. 67-78. (Unpublished Manuscript). , W. J. Pierson, Jr., and E. Mehr, 1962: Wave Spectra Estimated from Wave Records Obtained by the OWS Weather Explorer and the OWS Weather Reporter, (I), (Ii), (III). New York University, Research Division, Department of Meteorology and Oceanography, Technical Report Prepared for U. S. Navy Oceanographic Office Under Contract N62306=1042. ji Pierson, W. J., Jr. and L. J. Tick, 1965: The Accuracy and Potential Uses of Computer Based Wave Forecasts and Hindcasts for the North Atlantic. Proceedings of Second U. S. Navy Symposium on Military Oceanography . , and L. Baer, 1966, Computer Based Procedures for Preparing Global Wave Forecasts and Wind Field Analysis Capable of Using Wave Data Obtained by Spacecraft; Sixth Simposium, Naval Hydrodynamics ACR-136, Office of Naval Research-Department of the Navy. 6] APPENDIX SYNOPTIC WEATHER CONDITIONS Period | - 4 December to 6 December 1966 Synoptic Weather Conditions On the first day there were two frontal systems in the North Atlantic Ocean, one in the western region, which was undergoing frontogenesis, and one in the eastern area, which was dissipating. A low pressure system was undergoing cyclogenesis just off the southeast coast of Greenland with a series of waves along its associated fronta! system extending southwestward across Florida. This cyclone developed very rapidly with its center moving northward between Greenland and Iceland to the north of Iceland by 5 December. Gale force winds were reported as far out as 500 miles south from the 956-millibar low pressure center. On the 5th the winds were increasing to whole gale force in the southwestem quadrant as the storm continued to move northward along the Greenland east coast. At this same time a new cyclone was developing on the frontal system about 400 miles east southeast of Cape Sable. This storm was blocked by an anticyclone to the east, however, and did not become of any great consequence. By 1200Z on the 6th, the Greenland storm was beginning to fill, as it reached almost total occlusion. At station | winds reached 46 knots for about 36 hours following the frontal passage and wave heights to 36.4 feet (significant height) were recorded by the wave meter. Very high waves were recorded from 1800Z on the 5th to 1200Z of the 6th. At station K the strongest winds were 18 knots with wave heights to 16.8 feet. Station K remained in the warm sector of the cyclone throughout the period because the cold front never extended this far south. Period II - 27 February to 28 February 1967 Synoptic Weather Conditions Early on the 27th a low pressure center, which was rapidly deepening, was moving east northeastward to about 150 miles south of Iceland. By 1200Z on the 27th, the low had deepened to 958 millibars and 60-knot winds were reported, A frontal system extended from the storm center to the south and southwest across the southwestern North Atlantic. As the cyclone center deepened further to 952 millibars and moved northeastward to the east of Iceland, the frontal system moved rapidly eastward to the east of the British Isles by 0000Z of the 28th. The cold front passed over station J between 0600Z and 1200Z on the 27th, but the strongest winds of 43 knots were reported at 1800Z of that day. Wave heights reached their maximum of 31 feet at 0000Z 28 February. Station A was in the northern and western sectors of the cyclone as the low center passed to the south of the station early on the 27th. Period Ill - 4 March to 6 March 1967 Synoptic Weather Conditions On the first day, a cyclone with a 956-millibar center, moving north- eastward, was located about 300 miles southwest of Iceland. Whole gale winds extended to 600 miles over most of the southern sector. At the same time a new cyclone was developing about 300 miles southeast of Cape Race, Newfoundland with whole gale winds to 300 miles from the center by early on the Sth. Frontal systems extended southward and southwestward from each of the two cyclones. By 0000Z on the 5th, the low to the southwest of Iceland had moved directly across Iceland in a northeastward direction and began to fill. The new low developing east of Cape Race moved rapidly eastnortheastward passing to the north of the British Isles and into the Norwegian Sea early on the 6th. It did not develop into as intense a cyclone as the more northerly one that crossed Iceland. From the beginning of the period, station J was in the cold air behind the frontal system associated with the deep low center which later crossed Iceland. At this time J was experiencing winds up to 48 knots, while wave heights reached a maximum about 18 hours later of 29.2 feet. Station A, at the beginning was also experiencing winds of about the same speed but the wave heights did not reach a maximum until 0000Z on the 5th when they were recorded at 22.4 feet. At station K, however, the winds were calm at this time. With the approach of the second cyclone on the 5th, the winds at station J reached a maximum of 55 knots just prior to the passage of the cold front associated with the system. The wave heights recorded at J were as high as 32.3 feet at 0000Z on the 6th. The highest wave heights were also recorded at station K at the same time but they were only 16.5 feet. The strongest winds at K were 27 knots at 1800Z 6 March. As the second storm moved rapidly eastward, the winds at station J died down quickly to 15 knots by 0600Z on the 6th. A-2 "Z0000 “9961 YdaWwIDIG 9 4 LYVHD YAHLVIM JDVAYNS DILAONAS GMSN *L-V UND A-3 02 * as i. ly ) CG "Z0000 “2961 AYVNYGIS 82 “LYWHD YIHLVIM JOVAYINS DILAONAS GMSN “Z-V JUNO! A-4 *Z0000 “2961 HOUVW 9 ‘LYVHD YIHLVAM JDVAYNS DILdONAS AMsSN “€-V FNS A-5 602-aL AMUN EH] PT] oI pup Buiyung “> “q :sioyynD UDBDO I!JUD]HW YOK] ay} 10} JaPOW UO!JOIPAIg SADA [DDWEWARY paziayndwos b jo uolyon|pAW uy punys, snBiy diys sayjoaM ud2a90O Jajaw aADM ausogdiys uo!jo1peid ADM |DOIeWIN| JyBiey eADM jupd1y1UuBIs winijyoads aADA, 602-UL Z4{IMO>SOW “| *7 pup Buiyung “> *q :sioujnD UDBIO BJU Hy YOK] eu} 10} japoyy UO!JOIPA1g SADA, JODWEWNHY peziiayndwos b jo uolyon|DAG uy pupjs] snBuy diys sayyoam ups90 Jajyaw aADM ausogdiys uoljoIpaid aADM |DOIJaWN| JyBiey eADM jund141UBIS winsyoads aADAA\ *pyop ynduy auiij-|pad jo syas Jua1a4J1pP OM} UBaMjag UMoYs aJD suos!DdWo> *JJOJs BADM D puD siajaw aADM ausoqdiys wos} spio9a1 8ADM Buisn apowi aiaM sasAjpuD }D914 =SI}D}g “paqiiosap s! wosBod uoljo1paid aADM |Po1ewinu pazisayndwoo D yo uolyDN|DAA *(60Z-YL) Seoueseje1 g ‘sajqo4 pup sain6iy Buipnjou! saBod 99 ‘Q761 Alor “Z1IMOAsOW “| J2UOIZ pud Buiyung “> pypuog Aq ‘NW4IDO DILNVILV HLYON JHL YO4 TAGOW NOILDIGIYd JAVM WWITAWNN GAZIILNdWOD V 410 NOILVNIVAA NV 2214JO D14dosBouDa0G |PADN “Ss “fp *p4yDp jndu} awij-jpa1 yo syas JUa19441P OMY UBaMjaq UMOYs 21D suos!uDdWo> *4JD4S 2ADM D pud siajaw aADM ausogdiys Woy Spiode1 ADM Buisn appw aiam saskjpuD {p91} =S1}DJS “paqiiosap s| wosBo1d uosjo!peid aADmM |Pol4eunu paziieyndwoo p jo uolyON|DAW "(60Z-YL) SPoueseya4 g ‘sa}qn4 pup sein61y Buipnyou seBpd 99 4Q76| Ajnr ‘Z\IMOASOW “| [@UOI] pu Buiyung “> pjpuog Aq “NWIDO SILNVILV HLYON JHL YO4 TAGOW NOILIIGIYd JAVM WWOLIWNN GAZIYLNdWOD V 40 NOILVNIVAT NV 291}40 914dpsBouns9Q |DADN| “Ss “A 602-aL ZUIMOSOW, “|| “1 pup Burjung “> °q ‘sioujnd UDBDO F!JUD|JY YNON) ayy 105 JaPO/ UOLJOIpasg SADAA JDOWEWNNY paziayndwo> b jo uoljpN|DAW uy punys] snBiy diys s94yDaM UDa90 Jaya BADM ausogdiys Uo!jo1paid SADM jDOLaWNY JyBiay aADM yun9141UBi¢ winsyoads SADA 602-UL ZYIMOSSOWN lll pup Buijung “> *q ‘s10yjnD UDBDO JUD] Hy YOK] ayy 10} ;apoyy UO! Ipalg PADMA |DDHEWAKY peziuayndwo> bp jo uolypN|DAW uy punys} snbiy diys sayjoam uD2a90 Jaya aADM auloqdiys uo!4j91paid aADM jDOLaWNN| JyBley aaom yunoiy1UuB Is winsyoads aADAA 9 S v € (4 l *pjbp ynduy awi4-jped jo syas jualajjIp OM} UBaMjag UMOYs 21D suos!IDdWo> *4JJD4JS @ADM D puD siajew aADM ausogdiys Woy $p10981 BADM Bulsn appwi aJaM sesAjoUD {91} =SI}DJg “paqiiosap s} wosBosd uoljoipaid aADM |Pol4euinu paziseyndwioo bp yo uolypnjDAR *(60Z-YL) Se2Uesajo1 g 4s9|qD4 pup sain614 Buipnyou! saBod 99 ‘761 Ajnr 4ZJIMOHSOW “| J2UO!T pud Buijyung “> pjouog Aq ‘NWIDO DILNVILV HLYON JHL YO4 TAGOW NOILDIGAYd JAVM TWOISWNN GAZIWALNdWOD V 410 NOILVNIVAA NV a214JO 214dosBoudaxGO |PADNY “Ss “ *pypp ynduy awij-|pa1 yo syas jualejsIp OM, UBaMjaq UMOYs 21D suOs!IDdWoO> *JJDJS 2ADM D pud siajaw @ADM ausoqdiys Woy $piode1 8ADM Bulsn appw alam sasAjpuD {091} =SIJDJg ‘“paqisosap s! wosBo1d uolyo1peid aADM |Pol4auinu pazidayndwios p jo uolyON|DAJ *(60Z-UL) Se2uesaje1 g 4s0|qD4 pup soin6iy Buipnyjour saBod 99 476) Ajnr 4ZJIMOHSOW “| [@UO!] puD Buljung *> pjDuog Aq “NWIDO DILNVILV HLYON JHL YO4 TAGOW NOILOIGAYd JAVM IWOLSWNN G4AZIYALNdWOD V 4O NOILVNIVAA NV 89141 14dbsBouvasGO |PADN| *S “/N 60Z-UL ZJIMOASOW “| * 7 pup Buijyung “> “q ‘sioujynD UDBIDO JUD} YOK] ayy 10} jaPOW UO!JOIPalg SADA, [DIEWINHY paziayndwos pd jo uolyon|DAq uy pupys] snbiy diys sayjDaM uDedO Jajaw @ADM ausogdius uo!jo1peid BADM |DOIAWN| 4yBieay adm juvs1y1uBi¢ wnijyoads SADA, 602-UL ZHMOXSOW) “1! 1 pud Buiyung “> *q :suoujnd UDBDO FI4JUD]4}W YON) ayy 104 Japo; UO!4{DIPAlg SADA |DOWEWAHY peziiayndwo> b yo uolyoN|DAZ uy pups} snBiy diys 1a4jD9M uD200 JajaW 2ADM ausogdiys uoljoIpaid ADM |DOIJaWN| 4yBiey aADM yund141UB IS winijoads SAD *pyop ynduy awi4-jpes jo syas JUa1a4,1pP OM} UBEeMjaq UMoYs 21D suos!IDdWo> *JJDJS BADM D pub siajaw 2ADM auogdiys Woy $pi09a1 @ADM Bulsn appwi asaM sasAjpuD D514 =SIJDJ§ “paqiuosap s! woiBoid uoljo1paid aAnm JDO 14auinu paziseyndwoo p yo uolyDN|DAA "(60Z-YL) Seoueiejo1 g ‘sajqQn4 pub sein6iy Buipnyjous seBod 99 40764 Ane 4ZJIMOASOW “| [@UO!I7T PUD BulyuNng “> pjDUCG Aq “NV4D0 DILNVILV HLYON JHL Os TAGOW NOILDIGIYd JAVM IWOINAWNN GAZIYILNdWOD V 4O NOILVNIVAT NV 891340 91ydosBounda9G9 |DADN “Ss “A "pyPp yndu! auwiiy-jpa1 yo syas JUaJa}JIp OMY UaMjaq UMOYs 21D suOs!1DdWO> *4JD4S BADM PD puD siajaw aADM ausogdiys Woy Spiode1 8ADM Bulsn appw asam sasAjpuD {D914 =S14DJ§ “peqlsosap si woi1Boid uoljo!paid aADm |PolMewinu paziieyndwos p jo uolyon|pAq *(60Z-YL) Se2ueseje1 g ‘sajqny pub seinBiy Buipnyou! saBod 99 ‘Q761 Ajnr *Z\IMOHSOW “| [2UO!7 pud Bulyung *> pypuog Aq ‘NV4D0 SILNVILV HLYON 3HL YO4 TAGOW NOILDIGIYd JAVM TVOINIWNN GAZIWILNdWOD V 4O NOILVNIVAI NV B21}JO 21ydosBouna0Q [DAD “Ss “fp 602-UL Z4IIMO>SOW “] “J pup Buijung “> “q :sdoujnD UDBDO BFUD]HY YON) ayy 40} jaPOW UO!JOIpalg SADA, |DIWEWNH| peziuajndwos bd jo uoljpn|pAq uy punys] snBiy diys ayyoem uDs90 Jaya aADM ausogdiys uoljoipaid aADM |DOLEWN| JyBiey aAnm yupo1y1UuBI¢ winsyoads aAD\A 602-UL ZHIMO>SOW “| ~ 7] pup Buiyung *> “q :suoujnD UDBDO FIJUD] JW YHOK) au4 105 jePO/W UO!JOIpadg SADAA JDOWEWNH| peziajndwos bd jo uoljpn|DAq uy punjs| snBiy diys seyyoam upesG Jaya aADM ausoqdiys uo!4j91pasd aADM |DOLeWN| JyBiey aAnm yund1y41UB Is winsjoads aADAA\ *pyDp yndul auiij-jpad Jo syas JUa1a4JIP OM UBaMjaqg UMOYs 21D suosIDdWO> *4JJDJS BADM D puD siajaw S9ADM ausogdiys wos} $pi09e1 aADM Buisn apow aiaM sasAjpuD {D914 =SI}DJS “paqiiosap si wosBoid uoljoipeid aAnm jPol4auinu paziieyndwos p yo uolyHN|DAA *(60Z-YL) Se2ueteya1 g “sejqn4y pup sein61y Buipnyouy saBod 99 ‘Q761 Ajnr 4ZJIMOASOW *| J2UO!] pur Buiyung “> pjpuog Aq “NW4I90 DILNVILV HLYON JHL YO4 TAGOW NOILDIGIYd JAVM TWOLSWNN G3ZIYLNdWOD V 40 NOILVNIVAI NV 891440 O14dpsBounda9GQ |PADNY “Ss “f] *D4JDp jndu! aui4-jDa1 yo syas JUa1a441P OM} UBaMjag UMOYs 21D suos!ADdWO> *J4JDJS BADM D pub siajaw ADM ausogdiys Woy $pi0981 BADM Bulsn appw asaM sasAjoUD |D91} =SIJDJS “paqiiosap s! woisBoid uoljo1peid aAnm |P2l4auinu pazitayndwos pb yo uolypN|DAZ *(60Z-YL) Se2ueseje1 g “sejqnD4 pup seinBiy Buipnyou! saBod 99 ‘Q761 Alar “ZJIMOMSOW “| [@UO!7 pub Buljung *> pyouog Aq ‘NW3D0 SILNVILV HLYON JHL YO4 TAGOW NOILDIGIYd JAVM WOTIWNN G4ZIYILNdWOD V 40 NOILVNIVAI NV 291430 14ddsBouDadGO |DADN| “Ss “ ayy’ UNCLASSIFIED Security Classification DOCUMENT CONTROL DATA-R&D (Security classification of title, body of abstract and indexing annotation must be entered when the overail report is classified) 1. ORIGINATING ACTIVITY (Corporate author) 2a. REPORT SECURITY CLASSIFICATION UNCLASSIFIED U. S. NAVAL OCEANOGRAPHIC OFFICE 3. REPORT TITLE AN EVALUATION OF A COMPUTERIZED NUMERICAL WAVE PREDICTION MODEL FOR THE NORTH ATLANTIC OCEAN 4. DESCRIPTIVE NOTES (Type of report and inclusive dates) Technical Report 5. AUTHOR(S) (First name, middle initial, last name) Donald C. Bunting Lionel 1. Moskowitz 6- REPORT DATE Ja. TOTAL NO. OF PAGES 7b. NO. OF REFS } July 1970 66 8 8a. CONTRACT OR GRANT NO. 98a. ORIGINATOR’S REPORT NUMBER(S) -709-ET-RLA . PROJECT NO. TR-209 HF 05-552-304 9b. OTHER REPORT NO(S) (Any other numbers that may be assigned this report) - DISTRIBUTION STATEMENT This document has been approved for public release and sale; its distribution is unlimited. - SUPPLEMENTARY NOTES 12. SPONSORING MILITARY ACTIVITY U. S. Naval Oceanographic Office Washington, D. C. 20390 - ABSTRACT Procedures used to evaluate a computerized numerical wave prediction program are described. Statistical analyses were made using records from shipborne wave meters or a wave staff at five different locations in the North Atlantic and machine-made predictions of wave spectra for forecast intervals up to 36 hours. Comparisons are shown between two different sets of input data. The results of the evaluation indicate that automated numerical wave spectral predictions are feasible and that the forecasts are within a reasonable degree of accuracy for forecast intervals up to 36 hours. DD 2"..1473 (Pace) UNCLASSIFIED 1 NOV 65 S/N 0101-807-6801 Security Classification UNCLASSIFIED Security Classification LINK A ? oy Gone Lae ee Baie Wave spectrum Significant wave height Numerical wave prediction Shipborne wave meter Ocean weather ship Argus Island DD J2r",.1473 (Back) UNCLASSIFIED (PAGE 2) : Security Classification