Fall 1998

Special Seminar:

Testing GCM parameterization schemes: Sensitivity of the tropical water-vapour profile to GCM vertical resolution.

Adrian Tompkins

Max Planck Institute for Meteorology
Hamburg, Germany

4:30pm, Thursday, Aug. 6, 1998
Note: This seminar will be in LSC 4258, Psychology Wing

Special Seminar:

Representing Bottom Boundary Layers in Ocean General Circulation Models

Anand Gnanadesikan

Atmospheric and Oceanic Science Program
Princeton University

3:00pm, Thursday, Aug. 27, 1998
Please note special time.

Description: The bottom boundary layer is a hard problem for ocean general circulation models. Level coordinate models have trouble resolving the pressure gradient and in maintaining dense water properties. Isopycnal layer models have trouble getting the density contrast between the bottom layer and the interior right. Sigma-coordinate models have trouble with pressure gradient errors. This talk will explore a new approach in which a piecewise-sigma coordinate is turned on in appropriate regions, depending on the alongslope stratification of neutral density. It will demonstrate that this approach greatly improves the representation of bottom boundary layer physics. Impacts on a number of idealized problems and on a global general circulation model are presented.


Determining Shelf Circulation from Density Observations

Michael Dowd

Satlantic, Inc.

Thursday, Sept. 10, 1998
1:15 p.m.

LSC 4258 (Psychology Wing)
(NOTE Special time and place)

ABSTRACT: This talk will investigate the estimation of coastal and continental shelf circulation from scattered point observations of the density field. This problem has a long history in oceanography and is receiving increased attention due to advances in both numerical ocean modelling and data assimilation methods. The purpose of this talk is to motivate a dynamically consistent treatment of density and velocity. We first discuss (and illustrate) the commonly used 'diagnostic' method for determining circulation from density. Problem areas include the vorticity imbalance resulting from the interaction of the prescribed density field and the bathymetry, and the well-known barotropic indeterminacy (i.e. the 'level of no motion' problem). These shortcomings may be alleviated by allowing the density and velocity fields to co-evolve. Dynamically consistent density and velocity is then obtained, but at the expense of introducing significant complexity into both the dynamics and the underlying statistical estimation problem. We examine such a joint estimation procedure in the context of an idealized limited-area shelf circulation model, and propose an approach which estimates buoyancy fluxes across model open boundaries using density observations in the model interior.

Fully-nonlinear Solitary Internal Waves

Kevin Lamb

Department of Applied Mathematics
University of Waterloo

4:30pm, Thursday, Sept. 10, 1998

Abstract: Nonlinear solitary-like internal waves are common features in stratified coastal regions where they are generated by tidal currents over topographic features. Observed waves are often very large. Many features of these waves can be understood using weakly-nonlinear wave theories however these theories break down if the wave amplitude is too large. In this talk I will review current knowledge of fully-nonlinear solitary waves and make some comparisons between the predictions of weakly-nonlinear theory and fully nonlinear waves.


Seasonal Circulation on the Western and Central Scotian Shelf

Charles Hannah

Coastal Oceanography
Bedford Institute of Oceanography

TUESDAY, Sept. 15, 1998
1:15 p.m.

LSC 4258 (Psychology Wing)
(NOTE Special day, time and place)

ABSTRACT. Seasonal mean circulation fields for the western and central Scotian Shelf have been estimated through a synthesis of historical temperature and salinity observations, and a comprehensive shelf circulation model using realistic geometry and tidal forcing. The shelf-scale and bank-scale circulation will be discussed with an emphasis on the role of geography; in particular the proximity to major freshwater sources, the location of deep-water connections to the deep ocean, the distance from the tidally energetic Gulf of Maine, and the topography.


Predicting Subtidal Circulation over the Eastern Canadian Seaboard

Jinyu Sheng

Department of Physics and Physical Oceanography
Memorial University of Newfoundland

Thursday, Sept. 17, 1998
1:15 p.m.

LSC 4258 (Psychology Wing)
(NOTE Special time and place)

ABSTRACT. It is a great challenge to make reliable predictions of subtidal circulation over the eastern Canadian seaboard. In examining barotropic shelf circulation, Sheng and Thompson developed a linear barotropic model and applied it to the Scotian Shelf. We found that our model has a significant predictive skill for wind-driven currents over the coastal region with timescales of a few hours to a few days. The model skill is further improved by assimilating oceanographic data into the model. The Sheng-Thompson model however is less successful for predicting shelf circulation with periods longer than a few weeks during which the ocean baroclinicity plays a more important role.

We recently developed a three-dimensional primitive equation numerical model known as CANDIE (CANadian version of the DIEcast model) and applied it to the eastern seaboard including the Gulf of St. Lawrence, Scotian Shelf and Gulf of Maine. Our immediate goal is to simulate the multiple-year circulation of the region, allowing the density field to evolve with the flow. We were encouraged to find the predicted circulation pattern reproduces many of the observed features including relatively strong coastal currents over the Gaspe the Scotian Shelf, strong along-isobath shelf break currents and eddies in deep water spun off the Gulf Stream.

Calibration and Validation of ERS-1 and ERS-2 scatterometer winds against marine weather buoys

Yuri S. Geshelin
Department of Oceanography
Dalhousie University


Fred W. Dobson
Department of Fisheries and Oceans
Bedford Institute of Oceanography

4:30pm, Thursday, Sept. 24, 1998

Abstract. Interpretation of the data delivered by the European Remote-Sensing Satellites (ERS-1 and ERS-2) constitutes a problem because of a lack of understanding of the backscattering process at high wind speeds. Whereas the relationship of radar normal cross-section to wind speed is well known at wind speeds less than 20 m/s, the ERS retrieval algorithms are not reliable at higher wind speeds. Another major problem in ERS calibration is the lack of accurate marine wind data at high wind speeds. We have some evidence that at winds greater than 20 m/s the calibration formulae may be improved. Some relationships are found between ERS values and concurrent buoy measurements. These relationships are demonstrated to be lag-dependent and differ for ascending and descending satellite tracks, which is accounted for by the specific spatial orientation of storm tracks in the study area. We present the method for assessing the noise level due to instrumentation error of the satellite scatterometer.

At high wind speeds, one major problem with the buoys is inadequate and unreliable measurements of wind speeds. Because the buoy is often sheltered by the waves, it fails to measure wind speeds correctly. Therefore, "sea truth" is no longer a valid calibration criterion. We discuss the behaviour of buoys of different shapes and sensor heights at rough sea states.

Forecast Error modelling in the 3D-Variational Data Assimilation System at AES

Mark Buehner

Department of Oceanography
Dalhousie University

4:30pm, Thursday, Oct. 1, 1998

Abstract: This talk will report on ongoing work to improve the 3-dimensional variational data assimilation (3D-Var) system at AES. This system is used in conjunction with atmospheric forecast models to produce routine weather forecasts for Canada. The analysis system is used to correct the state vector of the forecast model every 6 hours using all available observations. To make optimal use of the observed data we need a good estimate of the uncertainties (error statistics) associated with both the observations and forecast model. This talk will focus on recent work to improve the estimates of the error statistics associated with the forecast model.

On Using Coupled Numerical Models to Improve Weather Forecasts

Harold Ritchie

Atmospheric Environment Service

4:30pm, Thursday, Oct. 8, 1998

This presentation will start with a brief update on the Atlantic Environmental Prediction Research Initiative which has recently been established in Halifax. It will then continue with a detailed example based on a related study that is nearing completion, with the preparation of an article entitled "On the Use of Coupled Atmospheric and Hydrologic Models at Regional Scale" by Robert Benoit, Pierre Pellerin, Nick Kouwen, Harold Ritchie, Paul Joe and Rick Soulis.

The purpose of this study is to present the capacities offered by coupled atmospheric and hydrologic models as a new tool to validate and interpret results produced by atmospheric models. The advantages offered by the stream flow observations are different from those offered by conventional precipitation observations. It is demonstrated that the dependence effect between basins and sub-basins can be very useful, and that the precipitation integrator effect of the larger basins facilitates the evaluation of state-of-the-science atmospheric models by filtering out some of the spatial and temporal variability that complicate the point-by-point verifications that are more commonly used. The stream flow permits a better estimate of the water amount fallen over a region. Comparing the stream flow predicted by the coupled atmospheric-hydrologic model versus the measured stream flow proves to be sufficiently sensitive to clearly assess atmospheric model improvements resulting from increasing the horizontal resolution and altering the treatment of precipitation processes in the model.

A look at the "solibore" field on the New England continental shelf

Dr. Jim Lynch

Department of Applied Ocean Physics and Engineering
Woods Hole Oceanographic Institution

4:30pm, Thursday, Oct. 15, 1998

Atmospheric Aerosols In The Northern Polar Region

Leonard A. Barrie

Atmospheric Environment Service

4:00pm, Thursday, Oct. 22, 1998
Room 5260, Psychology Wing
(Please note special time and place)

ABSTRACT. Atmospheric aerosols play a key role in several major environmental issues: climate change, , ozone depletion, pollution of ecosystems and human health. The northern polar atmosphere is underlain by a biologically-active ocean and surrounded by highly industrialized continents. In the cold half of the year, it is particularly susceptible to pollution that is transported on winds from Eurasia. This is superimposed on a natural background of sea salt from northern oceans, soil dust from Asian deserts and marine biogenic particles. Aerosol residence times are much longer in Arctic air mass(several weeks) than in air further south(several days). The concentration of aerosol mass is 10 to 20 times higher in winter than in summer. The north has av much different light regime than further south. It goes from 24-hours of sunlight in September to constant dark in December and then back to full light in April. At polar sunrise in March and April, dramatic changes in photochemically driven gas-to-particle conversion takes place. An ozone hole occurs at the ground through chemistry that resembles its stratospheric counterpart. However, instead of chlorine and bromine originating from perfluorocarbons, they come from sea salt halides that are chemically reduced by photochemical-heterogeneous reactions at the atmosphere/Arctic ocean surface. Volatile organic hydrocarbons, that have accumulated in the northen atmosphere during the dark, react in this environment to produce organic particulate matter. Atmospheric mercury is converted from a relatively inert gas to particulate mercury that is deposited to marine ecosystems much faster. The relative influence of sea salt, soil dust, marine biogenic and anthropogenic sources on northern aerosols is discussed. The role of aerosols in climate, ozone depletion and toxification of ecosystems is explored.

A framework for parameterizing transport by mesoscale eddies in ocean circulation models

Richard Greatbatch

Department of Oceanography
Dalhousie University

4:30pm, Thursday, Oct. 29, 1998

Some estimates of the subtidal circulation in the Western Bank region

Dr. Gleb Panteleev

Department of Oceanography
Dalhousie University

4:30pm, Thursday, Nov. 5, 1998

Abstract: Mean, density driven circulation in the Western Bank region were investigated by different approches: dynamic height, nonlinear diagnostic calculation and variational data assimilation method. Some results and comparison with current meter data will be presented.

Potential Vorticity Fluxes in a Shallow Water Model

Andrew Peterson

Department of Oceanography
Dalhousie University

4:30pm, Thursday, Nov. 12, 1998

Abstract: Recent work on new eddy parametrizations has focused upon the role of eddies in the transport of water mass and tracers either up the potential vorticity gradient or down the layer thickness gradient. We will use eddy resolving 1 and 2 layer models to test these hypothesis in a statistical sense.

3D numerical model of hydrodynamics and sediment transport using HOS method and GAS grid system

Qimiao Lu

Center for Marine Vessel Development and Research

4:30pm, Thursday, Nov. 19, 1998

Abstract: A 3D numerical modeling of hydrodynamics and sediment transport is developed by using a new numerical method, namely HOS method, and a new vertical grid system, namely GAS grid system, to achieve efficient and stable numerical environment. The HOS method is based on the operator splitting method and combined with Eulerian-Lagrangian method, FEM and FDM. The momentum equations and mass transport equation are partitioned into three subsystems by the operator splitting method according to physical and mathematical significance. Each subsystem is discretized using the different numerical method which can well represent the distinct physical subsystem. The special grid arrangement for the method can fully fit to complicated geometry to satisfy the boundary-fitted demands in coastal engineering applications. The method is proven to be unconditionally stable and efficient. To improve the numerical accuracy and capture sharp variations in sediment transport computations, a nonlinear vertical coordinate transformation, called "Gradient Adaptive Sigma" (GAS) grid system, has been developed in the model. GAS grid system transforms the varying water depths into unit of one and converges the vertical grids at the places where the vertical gradients of concentration are large. GAS grid system can increase the numerical resolutions in vertical and capture sharp variations of concentrations in transport modeling. Several numerical tests and applications are carried out by the developed model. The numerical results show that the developed model improve considerably the numerical accuracy, stability and efficiency.

Title: TBA
AGU Rehearsal

Carolyn Smyth, Steve Henderson, and Phil MacAulay

Department of Oceanography
Dalhousie University

4:30pm, Thursday, Nov. 26, 1998

Simulation of Internal Tides: Experiments with the Free-Surface CANDIE Ocean Model

Youyu Lu

Department of Oceanography
Dalhousie University

4:30pm, Thursday, Dec. 3, 1998

Abstract. The CANDIE (Canadian version of Diecast) ocean model has recently been modified to include a free surface. In this talk I will present results from two experiments using the CANDIE model. The first experiment deals with an idealized situation in which (i) the flow field varies only in the vertical and one horizontal direction (ii) water depth is constant in both the deep and shallow water and with a linearly sloping shelf break in between (iii) the density is initially uniform horizontally with constant vertical gradient (iv) perturbations are so small that most nonlinear terms can be dropped from the model. I will show that the results from CANDIE are in good agreement with analytic solutions obtained using the linear theory of Craig (1987). In the second experiment we apply the fully nonlinear, three-dimensional model to the Gulf of St. Lawrence using a summer density climatology to initialize the model's density field. The boundary conditions are obtained using an incremental data assimilation method in which a linear shallow water model is used to minimize the difference between observed and predicted sea surface elevations using the adjoint method. The model generated co-amplitude and co-tidal charts agree well with the assimilated tidal amplitudes and phases. More encouragingly the predicted tidal currents are in good agreement with an independent set of current measurements.

Climate of the Northwest Atlantic as seen from hydrographic data

Igor Yashayaev

Ocean Circulation Section
Ocean Sciences Division
Bedford Institute of Oceanography

4:30pm, Thursday, Dec. 10, 1998

Abstract. Not long ago we deeply believed that
1) there is a quasi-equilibrium state of the ocean or a norm, that can be used to reveal and trace any anomalous events occuring in the marine environment,
2) we are able to construct such state by more or less sophisticated averaging of existing observations. Analyses of the most recent data show that we have to review this concept and be more careful using the "climatology" in both data analysis and models.

The upper waters on the either of the Subpolar Front have different tendencies and dominant scales of the interannual-to-interdecadal variability.

The Labrador Sea Water (LSW) has greatly changed both its properties and distribution over time. It reached a maximum in temperature and salinity in the late 60s - early 70s, while during the 90s it has become colder, fresher, denser and deeper than at any time in this century. The tendency changed around 1994 and now the LSW is steadily getting warmer, saltier and lighter.

Between the early 70's and 90's North-East Atlantic Deep Water (NEADW) and Denmark Strait Overflow (DSOW) also became colder and fresher over a large portion of the North-West Atlantic.

The remnants of Antarctic Bottom Water (ABW) entering the subpolar gyre have been getting colder in the past two decades, but the rate of this change was less than that for DSOW, resulting in a change of spatial gradients in the bottom layer.

The change in the the properties of the upper, intermediate and deep waters throughout the North Atlantic resulted in changes of dynamic heights and potential vorticity, those of a great importance for understanding and modelling the ocean dynamics.

Capsize Analysis for Ships with Water Shipping on and off the Deck

C.C. Hsiung

Centre for Marine Vessel Development and Research

4:30pm, Thursday, Dec. 17, 1998

Abstract: A prediction method of water shipping on and off the deck was developed and the computed results were validated by model tests. The amount of water trapped on the deck varies with time. The shallow water wave equations are solved to obtain the force caused by water flow on deck. The nonlinear equations of ship motion were solved in the time domain to study the capsize of vessels with the varying mass of water flow on deck.