Where model and reality meet

Where model and reality meet 

February 26, Andvord Bay 64 50.58S, 62 32.76W

Written by: Dr. Lisa Hahn-Woernle

Hi everyone!  My name is Lisa Hahn-Woernle and I am working on the physical-ecological model of Andvord Bay as part of the FjordEco Project. It is my first time down here and everything on this trip has been super exciting for me so far. February 26 was exceptional though! I woke up at 6 am because I felt the boat speeding up. That meant that the mates decided to return to Howard Island. That ALSO meant that they decided to traverse the entry of the Gerlache Strait around 64 50.0 S, 64 W the whole night. Why would they do that? And why am I so excited about this? Well, I am trying to model the region around Andvord Bay in an ocean model that simulates the water movement as well as the biology in the ecosystem. The main advantage of modelling the fjord numerically is that we can study processes controlling the fjord that we otherwise cannot measure. For example, we can determine how important the winds are in driving water in and out of the fjord, and what effect meltwater from icebergs and glaciers could play on the physical and biological processes. Observations are essential to capture the actual state of the fjord, but they are mostly limited in space and/or time. Even though the model cannot resolve everything that happens in the fjord, it allows us to simulate the full basin over a continuous period of time when we are not present.  This will hopefully help us to understand the greater connections that lead to the snapshots we see in our observations.


Fig. 1: Model domain in the region of Andvord bay (red needle) and the Gerlache Strait. The coloured countours outline the topography. The model boundary sampled in Fig. 2 is along 64o W between 64.8 and 64.88 º S (northern and deep part of the western boundary).


Fig.2: Raw ADCP data passing over the northern part of the western model boundary for 6 times in the night of Feb 26/27 and twice until midnight of Feb 27 (red boxes). From this raw data it is hard to assign a specific location to a measurement. My intuition is that the inflow (positive/red features at the surface in the Ocean U plot) occurs over the deep part of the channel. The outflow would then occur in the shallower regions close to the coastline. The tides induce a strong outflowing current that leads to the reduction of the flow till the early hours of Feb 27. As the tides turn around, the flow picks up again.

We are aiming to model the ecosystem during only the summer months from November until the end of March. To start the model, we need to know what the typical state of the fjord would be in November. For example, what temperature and salinity profiles would be characteristic of the water in the fjord. If the region we modelis closedso there is no water flowing in and out of the fjord, temperature and salinity would be enough information needed to run the model (assuming we had the correct forcings like tides and meterology). As you can see in Figure 1, the region we model has 3 open boundaries: one in the west and two in the north-east. To run the model over several months, we need to have an idea about the water masses that flow in and out of these boundaries. What temperature and salinity do they have, what is their vertical structure, do they flow in or out? All of this will have an impact on the model results of the fjord itself. So, for the past months I worked my way through data of different ocean models and measurements to find information about the typical state of our study region. This includes seasonal signals of the water temperature and salinity, meteorological forcing such as winds and air temperature, tidal forcing, and also the strength and direction of the current that flows through the Gerlache Strait. Data down here are scarce, which makes it hard to define the initial model state and the forcing at its boundaries. The parts I have been struggling with the most are the currents that flow in and out of the region. And this brings us back to the morning of February 26. As the boat travels, an acoustic Doppler current profiler (ADCP) is measuring the currents below the hull. This instrument sends out an acoustic signal and estimates the velocity of the water below (up to a certain depth) based on the Doppler effect of the acoustic signal scattered back to the boat. Pretty neat! So we decided to take advantage of our nights and slowly steam back and forth along the model boundary close to Palmer Station. This way we could repetitively measure the cross-stream structure of the currents over a semi-diurnal (twice daily) tidal cycle. Since boats normally only pass along the currents into and out of the Gerlache Strait, this is a pretty unique data set. The cross-current structure will not only help for the boundary conditions of the model, but also allow us to study the importance of tidal currents in the region. By now, we collected data over 4 nights with 6 to 10 crossings per night, which makes a pretty neat dataset for everyone interested in the currents and tides of the region.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s