Alaska Coastal Ocean Forecast System - Regional (ALCOFS-R)
DISCLAIMER: This model is under experimental development. The results should not be used for navigational purposes or emergency planning under any circumstances. Please do not duplicate and use the result figures from this website without permission.
UPDATES
- Updated on 09-13-2020: A global forecast wave model using WW3 with IC4M2-IS1 is published on the section 8.
- Updated on 09-01-2020: Updated for using 1 hourly wind, pressure, ice forcing, previsouly using 3 hourly, and restarted SWAN+ADCIRC.
- Updated on 08-25-2020: Updated for adding the significant wave height forecast by using SWAN+ADCIRC with the IC4M2 method in SWAN.
- Updated on 08-07-2020: Updated to Version 2, daily forecast simulations just include 1 day hindcast and 5 day forecast.
- Updated on 06-13-2020: Updated the bathymetry by using GEBCO2020.
- Updated on 06-10-2020: ADCIRC Ver.55 is applied.
- Established on 04-28-2020: Testing Version 1, hindcast for 13 days and forecast for 5 days every day.
NOTES
The forecast system runs once a day, starts from 01:30, ends in around 04:00. The forecast animations are uploaded at 08:00, the water level figures are updated at 06:35, 12:35, 15:35, 18:30 Eastern (UTC−4) for comparing with the latest observational data from NOAA stations. The time showing in the figures are in UTC.
1. GFS-FV3 wind speed (5 day forecast)
2. GFS-FV3 atmospheric pressure (MSLET, Mean Sea Level Pressure) (5 day forecast)
3. GFS-FV3 ice concentration (5 days forecast)
4. ADCIRC surface water elevation (5 day forecast)
Animation, 2 hourly output. The simulation result in the animation apply ADCIRC with ice parameterization.
The following animation is an enlarged view for the surface elevation and wind speed vectors. The figures for this animation were generated by FigureGen v.49.
5. Maximum ADCIRC surface elevation for the 5 day forecast period
Zoom in by clicking the following locations:
・Whole Domain・Western Alaska・Cook Inlet・Northern Alaska・Inside Passage
6. SWAN+ADCIRC significant wave height (5 day forecast)
The IC4M2 method (Collins and Rogers, 2017) is utilized in SWAN for activating a source term Sice to represent the dissipation of wave energy by sea ice. Some parameters setting for SWAN: time interval=3600s, MDC=36(number of meshes in θ-space), MSC=40(one less than the number of frequencies). For the IC4M2 parameter, we use C2=1.06e-3, C4=2.30e-2 according to Meylan et al. (2014), which is for the case of ice floes, mostly 10 to 25 m in diameter, in the marginal ice zone near Antarctica. (Further consideration may need to be done in the future work since we are handling different ice zone)
7. SWAN+ADCIRC radiation stress gradient (5 day forecast)
8. Global GFS-FV3 wind speed, Global ADICIRC surface water elevation (5 day forecast) and maximum surface elevation for the 5 day forecast period
Global finite element mesh(GESTOFS v4.1.4 model, the same global model using at https://cera.coastalrisk.live): 8,452,486 nodes, 16,226,163 elements; compuational time step: 12s; computational wallclock time: about 2h30min by using 240 cores for 5 day forecast.
9. Comparison with stations and buoys
*blue: water level stations with observation; green: water level from model; yellow: wave buoys.Sitka Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9451600
The black line shows the observational data; the dashed blue line, the hindcast results; the solid blue line, the forecast results.
Skagway Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9452400
Yakutat Bay Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9453220
Anchorage Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9455920
Nikiski Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9455760
Adak Island Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9461380
Village Cove, St Paul Island Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9464212
Unalakleet Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9468333
Note that the Unalakleet station (ID 9468333) is located within an inlet near the mouth of Unalakleet River. The inlet itself is quite narrow (75m). We have noted during both winter and summer months significant unexplained variability in high tides and especially in low tides in the recorded signal, with low tides being limited. We suspect that shoal formation in the vicinity of the inlet and possibly ice formation at the inlet mouth significantly impact the measured tidal signal at the station.
Nome Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9468756.
Red Dog Dock Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9491094
Prudhoe Bay Station
https://tidesandcurrents.noaa.gov/stationhome.html?id=9497645
Kuskokwim Bay (5 day forecast)
Location:-162.0293, 59.3780.
10. Overview of the forecasting system
The western coast of Alaska has experienced a number of storm surge events ([1], [2]). In order to predict the water level around the western coast of Alaska, we developed this high resolution and continuously forecasting system. In this system, the ADCIRC [3] storm surge model acts as a main engine and incorporated air-sea-ice drag parameterization, which aim at higher accuracy for predicting water level during winter storm by considering sea ice coverage. For more detail about air-sea-ice drag parameterization, please refer to [5], [6], [7].
11. Flowchart of the forecasting system
Fig.1 shows the flowchart for this forecasting system. In order to get a high quality initial condition to start the forecast, on the first day of the forecast simulation, we did a 12 days hindcast including 5 days spin up by using GFS reanalysis products (https://www.ncei.noaa.gov/data/global-forecast-system/access/grid-004-0.5-degree/analysis/) for 10 days and GFS forecast products (https://www.ftp.ncep.noaa.gov/data/nccf/com/gfs/prod/) for 2 days, after the 12 days hindcast, the forecast system goes into a daily loop simulation for 6 days including 1 day hindcast(green day in Fig.1) and 5 days forecast(blue days in Fig.1) using GFS forecast products. In Fig.1, the red 1d indicates the present day.
12. Mesh with floodplain (nodes: 1,648,088; elements: 3,211,583)
The ALCOFS mesh was generated by using OceanMesh2D.
Fig.2 shows the mesh resolution, for the parameters of ocean side mesh used in the OceanMesh2D:
min_el=1000, max_el=20000, wl=300, dt=2, grade=0.2, R=2, slp=20, dis=0.35, max_el_ns=1000,
and several specify bounding boxes for bays and inlets (Cook Inlet, Bristol Bay, Aleutian Islands, Kotzebue Sound, St. Lawrence Island) are set for min_el=200.
For the floodplain side, there are two floodplains in the mesh,
the first one is located at the north of Alaska, the second one is at Yukon Delta:
for the first one min_el=200, the second one min_el=100 and the other parameters are set the same as ocean side.
For the bathymetry data, we blended the GEBCO2020, the Smooth Sheet bathymetry of Aleutian Islands, Cook Inlet, Central Gulf of Alaska, Norton Sound, and the Southern Alaska Coastal Relief Model(CRM) bathymetry around Lynn Canal area. Fig.3 shows the mesh bathymetry; the dark green areas above 0 meter are floodplains.
13. Forcing boundary conditions and simulation conditions
8 tidal harmonic constituents are forcing on the open ocean boundaries,
4 semi-diurnal: M2, N2, S2, K2;
4 diurnal: K1, Q1, O1, P1.
The simulation is based on air-sea-ice drag parameterization incorporated ADCIRC ver. 55, the following are some settings for the parameters in ADCIRC.
ICS=-22, IM=511112, NTIP=2, TAU0=-3, DTDP=2, NTIF=8, NWS=14 (w/o ice), NWS=14014 (w/ ice), NWS=14314 (w/ ice&SWAN)...
14. References
- W. Blier, S. Keefe, W.A. Shaffer and S.C. Kim: Storm Surges in the Region of Western Alaska, Monthly Weather Review, Vol.125, 1997.
- R.S. Chapman, S.C. Kim and D.J. Mark: Storm-Induced Water Level Prediction Study for the Western Coast of Alaska, ERDC/CHL Letter Report, 2009.
- ADCIRC: https://adcirc.org/.
- SWAN+ADCIRC: https://ccht.ccee.ncsu.edu/swanadcirc/.
- Brian R. Joyce, William J. Pringle, Damrongsak Wirasaet, Joannes J. Westerink, Andre J. Van der Westhuysen, Robert Grumbine, Jesse Feyenc: High resolution modeling of western Alaskan tides and storm surge under varying sea ice conditions, Ocean Modelling, Vol.141, 101421, 2019.
- Christof Lüpkes, Vladimir M. Gryanik, Jörg Hartmann, and Edgar L Andreas: A parametrization, based on sea ice morphology, of the neutral atmospheric drag coefficients for weather prediction and climate models, Journal of Geophysical Research, Vol.117, D13112, 2012.
- G. Ling, D. Wirasaet, J. Westerink, D. H. Richter, B. Joyce, W. Pringle, M. T. Contreras Vargas, K. R. Steffen, C. N. Dawson, A. Fujisaki-Manome, E. Myers, S. Moghimi, S. V. Vinogradov, A. Van der Westhuysen, A. Abdolali, and R. Grumbine: Studies on Parameterizations of Sea Ice Effect in a Storm Surge Model for Western Alaska,
https://ams.confex.com/ams/2020Annual/webprogram/Paper370912.html, AMS, 2020.
- C.O. Collins and W.E. Rogers, A Source Term for Wave Attenuation by Sea ice in WAVEWATCH III ®: IC4, NRL Report NRL/MR/7320--17-9726, 2017.
- M.H. Meylan, L.G. Bennette and A.L. Kohout: In situ measurements and analysis of ocean waves in the Antarcticmarginal ice zone, Geophysical Research Letters, Vol.41(14), pp. 5046–5051, 2014.
- GLOCOFFS (GLObal Coastal Ocean Flood Forecasting System): https://wpringle.github.io/GLOCOFFS/.